Citation
Physicians' opinions about patient involvement in health and medical care decisions and telephone-based decision support

Material Information

Title:
Physicians' opinions about patient involvement in health and medical care decisions and telephone-based decision support
Creator:
Howell, Jerry Wayne
Place of Publication:
Denver, CO
Publisher:
University of Colorado Denver
Publication Date:
Language:
English
Physical Description:
xiii, 352 leaves : illustrations ; 28 cm

Subjects

Subjects / Keywords:
Physician and patient -- Decision making ( lcsh )
Medical care -- Decision making ( lcsh )
Telephone in medicine ( lcsh )
Physicians -- Attitudes -- United States ( lcsh )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Bibliography:
Includes bibliographical references (leaves 343-352).
Thesis:
Health and behavioral sciences
General Note:
Department of Health and Behavioral Sciences
Statement of Responsibility:
by Jerry Wayne Howell.

Record Information

Source Institution:
|University of Colorado Denver
Holding Location:
|Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
44075519 ( OCLC )
ocm44075519
Classification:
LD1190.L566 1999d .H69 ( lcc )

Downloads

This item has the following downloads:


Full Text
PHYSICIANS OPINIONS ABOUT PATIENT INVOLVEMENT IN
HEALTH AND MEDICAL CARE DECISIONS AND
TELEPHONE-BASED DECISION SUPPORT
by
Jerry Wsyne Howell
B.A., University of Northern Colorado, 1977
M.S., University of Colorado at Boulder, 1986
A thesis submitted to the
University of Colorado at Denver
in partial
fulfillment
of the requirements for the degree of
Doctor of Philosophy
Health and Behavioral Sciences
1999


1999 by Jerry Wayne Howell
All rights reserved.


This thesis for the Doctor of Philosophy
degree by
Jerry Wayne Howell
has been approved
by
Donald C. Iverson
flirt
Date


Howell, Jerry Wayne (Ph.D., Health and Behavioral Sciences)
Physicians Opinions About Patient Involvement in Health and Medical Care
Decisions and Telephone-Based Decision Support
Thesis directed by Professor Deborah S. Main
ABSTRACT
Survey research (n = 914) and key informant interviews (n = 21) were used to
assess physicians opinions about: 1) colnsumers/patients being more informed and
taking a more active partnership role in iliedical care decisions, 2) the incorporation
of telephone-based decision support (TB:DS) into care management, and 3) the effect
a more active consumer/patient partnership and TBDS will have on health outcomes,
patient satisfaction, doctor-patient relationships clinical efficiencies, utilization, costs,
and malpractice liability. Die percentage of physicians agreeing that patients want
more information about health and medical issues and more active involvement in
medical decision making was 90% and 88%, respectively. The percentage of
physicians agreeing that physician and patient together should make decisions about
minor and major medical conditions wasj 56% and 68%, respectively, with 66%
agreeing that they would like their patients to take a greater role in self-managing
chronic conditions. Physicians believed jthat a more informed patient would result in
better medical decisions leading to improved health outcomes, patient satisfaction,
doctor-patient relationships, utilization efficiencies, and a potential reduction in
medical liability. However, physicians varied in their preference for involving
patients in decision making and acknowledged that patients have varying preferences
for involvement. They believed that the most prudent clinical approach is for the
physician to be sensitive to each individual patients preference for information and
involvement. Opinions varied regarding (the purposes, potential impacts, and reasons
for why health plans had implemented TBDS services. Logistical regression analyses
showed that primary care physicians werie more likely than referral specialists to:
agree that their patients should take a greater role in treating minor conditions; agree
that their patients should take a greater role in self-managing chronic conditions; be
knowledgeable about TBDS; have a more positive opinion about TBDS; believe that
IV


the health plans implemented TBDS to respond to interests of the business
community, to more efficiently care for capitated patients, improve quality of care,
and member satisfaction; believe that TjBDS would decrease the number of
medically inappropriate visits and unnecessary office calls. Explanations for these
findings, implications for working with physicians, and recommendations for future
research are discussed.
This abstract accurately represents the content of the candidates thesis. I
recommend its publication.
Signed
Deborah S. Main


DEDICATION
To my God, who provided endurance an d a quiet place for the soul by reminding me
that, For the moment all discipline seems painful rather than pleasant; but later it
yields the peaceful fruit of righteousness to those who have been trained by it
(Hebrews 12:11).
To my mother, Alice; my father Wayne; my sister, Cindy; and my brother, Byron; for
inspiring me with living faith, courage, iind patient determination when faced with the
sometimes overwhelming challenges of life. Thank you for your unconditional love.
To the very special group of outstanding educators, coaches, and mentors who have
encouraged me to complete an advanced degree. I am grateful to all of you for your
mentoring and your inspiration. Youve
the better.
made a difference in shaping my life for
To my wife, Cynthia. Despite the many conflicts I faced on the outside and the many
fears, anxieties, and uncertainties within, her enduring love and vibrant devotion were
always there stirring me to remain strong and of good courage, to be neither
frightened, nor dismayed (Joshua 1:9); but to finish the race that had been set before
me. Thank you for the reassurance and strength I found in the gentleness of your love
and for all the sacrifices you had to make while I completed this work.


ACKNOWLEDGEMENTS
The writer wishes to extend his sincere thanks to:
Dr. Main for her unwavering confidence in me, her boundless enthusiasm, her
sage leadership, mentoring and coaching in directing this thesis.
Dr. Main, Dr. Iverson, Dr. Green Dr. Lynch, and Dr. Stulp for working
together as a multidisciplinary team in sharing their wealth of knowledge and
experience to create and nurture a learning environment that should
characterize all dissertation experiences.
technical assistance in data management and
Joe McGloin for his patience and
statistical analysis.
Lu Sandoval for her tenacity and
informant interviews.
Sherry Holcomb for her expertise
administration and data collection.
David Nash, M.D., Zach Gerbarg
responsiveness in transcribing the key
and attention to detail in survey
M.D. and Mark Earnest, M.D., for their
critical review of the qualitative analyses.
Barb and Don Day for providing a home away from home.
Pepper Hatch for his leadership in
flying high cover and providing
consistent support and encouragement to complete this work
All the physician subjects who graciously committed time and energy to this
study.


CONTENTS
CHAPTER
1. INTRODUCTION.
Specific Aims.......
Background and Significance.
Overview of Research Design and Setting.
Study Design....
Study Site.....
Telephone-Based Decision Support Services.
Key Questions...
Reconciling Quantitative and Qualitative
Analysis.........
Organization of the Dissertation
2. REVIEW OF LITERATURE
Geographical Variatiohs and Appropriate
Use of Health Care.
Demand-Side Interven
Perceived Need anc
ions.
Self-Care
Chronic Disease Management.........
Patient Preference and Shared Decisions.
. 1
..2
.3
..8
..9
11
12
16
18
18
20
20
24
24
27
29
vm


Patient Preference for Information and Involvement
in Decision Making.............................
32
Physician Perception about Patient Desire for
Information and Invo vement............................52
Doctor-Patient Communication and Relationships:
Challenges/Barriers ahd Effects on Outcomes...
Doctor-Patient Communications: Challenges
and Barriers...............................
54
55
Doctor-Patient Communications: Effects on
Outcomes........................................65
Models for Physician-Patient Communication
and Relationships. J...............................71
Diffusion of InnovatiJns: Constructs, Components,
and Determinants...j..................................79
Components of Diffusion.............................79
Adoption Rate Determinants..........................82
Diffusion of Innovation in Health Care
85
Summary and Implications for Research..................117
3. QUANTITATIVE METHODS AND RESULTS
119
Introduction
119
Methods
120
Subjects.......................
Physician Opinion Survey (POS)
120
120
IX


POS Administration and Data Collection
Protocol..............................
122
4.
Analysis and Statistical Methods
Sample Description...................
Description of Variables....
Selection of Statistical Tests
Results.........................
Sample Description..........
Description of Variables
Logistic Regressions...
QUALITATIVE METHODS AND RESULTS
Introduction...............................
Methods....................................
Sample Selection.........................
Physician Opinion Interview (POI) Guide
Interview Structure and Administration...
Analyses.............!.....................
Transcription, Coding, Segmentation.....
Data Analysis and Codebook..............
Interpretation..........................
123
123
123
126
129
130
132
133
153
153
154
154
156
157
158
159
160
161
x


Results
163
Subjects............................
Opinions About Patient Involvement in
Medical Decisions...................
163
165
Opinions About Telephone-Based
Decision Support...............................214
Closing Questions
242
Synopsis: Overarching Themes.......................246
Theme 1: Physicians Support Idea of
More Informed Active Patients....
Theme 2: Improved Health Outcomes: Chronic
and Major Illness ......................
Theme 3: Improved Patient Satisfaction
..........................
Theme 4: Improved Doctor-Patient Relationship
(Patient Preference).............................
246
247
249
250
Theme 5: Improvedj Clinical Efficiencies for Minor
and Chronic Illnesses................................. 252
Theme 6: Reduced Health Care Costs for Minor
and Chronic Conditions......................
255
Theme 7: Reduced Potential for Medical
Liability.........
Theme 8: Self-Efficacy Improved with More
Informed Active Patient.......................
256
257
Summary: Themes
1-8
257
xi


Theme 9: Variance in
Preferences.........
Physician and Patient
Theme 10: TBDS Uncertain or No Impact on
Physicians Practices and Patients............
Theme 11: TBDS Lack of Support by
Practicing Physicians Will Diminish Its Value
258
.261
262
5. DISCUSSION.....................................
Key Quantitative and Qualitative Results..
Physicians Opinions about Shared Decision
Making..........I......................
Physicians OpiniJns about Telephone-
Based Decision Support.................
Commentary on Implications for Working
with Physician Organizations...........
Limitations
Recommendations for Future Research
Conclusions
APPENDIX
A. PHYSICIAN OPINION SURVEY (POS)...
B. PHYSICIAN OPINION INTEVEW (POI).
266
266
267
277
291
304
314
319
321
328
xii


C. QUALITATIVE ANALYSIS CODEBOOK............334
D. VARIABLES WITH MISSING VALUES............340
REFERENCES
343


CHAPTER 1
INTRODUCTION
During the last decade of U.S. socio-cultural history, the consumer has played
an increasingly greater role in shaping the marketplace. The baby-boomer generation
that has shaped virtually every other sector of the U.S. economy is emerging as the
new consumer for the healthcare marketplace as well. This new consumer is
generally more informed; more demanding about access, convenience, and service
quality; generally more assertive in the doctor-patient relationship. In essence, they
want to be more self-sufficient and independent. For this new consumer to be
effective in shaping the healthcare marketplace, they will need access to technological
tools and processes that can provide information and that can improve their
effectiveness in the doctor-patient relationship.
Consequently, this new era of consumerism in healthcare has stimulated an
interest in the concept of demand management and has spawned a rapidly growing
increase in consumer health management and decision support services. Demand
management has been defined as the use of decision and self-management support
systems to enable and encourage consumers to make appropriate use of medical care
(Vickery & Lynch, 1995). One of the consumer health management services that has
emerged has been telephone-based decision support (TBDS). Through a 1-800
1


number, such services provide 24-hour per day access to specially trained nurse
counselors. Using on-line decision suppjort protocols or algorithms, the nurse
counselors provide decision support, self-jnanagement support, psychosocial support,
education and information to consumers/patients on virtually any health care topic.
The market potential for these types of services has been estimated to be somewhere
between 100-200 million lives (Equitable Securities Research, 1996).
Even though the opinion of the physician may stands as a critical variable in
determining the ultimate impact of more ii formed, actively involved patients and
TBDS, reviews of published literature revealed no systematic assessment of
physicians viewpoints about these issues.
This research will advance our
understanding about how physicians view enhanced participation of patients in medical
decision-making and the integration of TBDS services into care management. This
information will provide key insights on the implications for working with physicians
in optimizing the integration of these components into the delivery of health care.
Specific Aims
A multi-method approach was used to address four central issues: The specific aims
of this research were:
2


Aim #1: To ascertain primary care and specialty physicians opinions about
more informed, actively involved patients as partners in health and medical care
decisions.
Aim #2: To assess primary care and specialty physicians opinions about the
impact they believe consumers/patients bring more informed and taking a more active
partnership role in health and medical care decisions will have on: the quality of the
doctor-patient relationship, clinical efficiencies, health outcomes, patient satisfaction,
health care utilization, health care costs, and malpractice liability.
Aim # 3: To assess primary care and specialty physicians opinions about the
incorporation of TBDS services into care
management.
Aim # 4: To assess primary care and specialty physicians opinions about the
impact they believe a TBDS service will have on: the quality of the doctor-patient
relationship, clinical efficiencies, health outcomes, patient satisfaction, health care
utilization, health care costs, and malpractice liability.
Background and Significance
Proposals and strategies for health care reform (generally cost containment)
have been derived chiefly from a supply-side economic paradigm. Supply-side
strategies have historically emphasized global budgets, consumer tax incentives, cost-
shifting to consumers, capitated payments for physicians, technology management,
3


hospital bed capacity control, use of gatekeepers, utilization review, and other forms
of "managed care." In contrast, little emphasis has been given to integrating the
biological sciences with demand-side social, cultural, psychological, and health risk
factors which influence health care decision making, health care behaviors, and health
care utilization that ultimately impact health outcomes, patient satisfaction and health
care costs.
Recently, however, the concept of demand management has received
significant national interest with the incorporation of demand-side strategies into
managed care delivery systems. Demand management has been defined as the use of
decision and self-management support systems to enable and encourage consumers to
make appropriate use of medical care (Vickery & Lynch, 1995).
The supposition behind demand-side strategies is that health care decisions and
health care seeking behaviors are influenced by a complex set of psychological, social,
and cultural factors which are more related to perceived need and patient preference
rather than to morbidity (Vickery & Lynch, 1995). It appears that the access an
individual has to information and resources influences perceived need and patient
preference for services which in turn influences behaviors associated with health and
medical decisions (Berkanovic Telesky, & Reeder, 1981; Bradley, Zia, & Hamilton,
1996; Dentzer, 1991; Greenfield, Kaplan, Ware, Yano & Frank, 1988; Levine &
Britten, 1973; Lorig, Mazonson & Holman
1993; Mayo, Richman & Harris, 1990;
4


Vickery, Kalmer, Lowry, Constantine, Wright & Loren, 1983). That is,
a well-informed patient will make different decisions than a less-informed patient and
will subsequently utilize health care resources differently.
This recent interest in demand-side strategies has subsequently spawned a
rapidly growing increase in consumer health management and decision support
services. One such service has been TBDS. Through a 1-800 number, such services
provide 24-hour per day access to specially trained nurse counselors. With the aid of
sophisticated on-line decision support protocols or algorithms, the nurse counselors
provide decision support, self management support, information, and psychosocial
support to consumers/patients for: (a) triage or pre-visit decisions regarding acute
minor illnesses and injuries (e.g., infections, headache, back pain, etc.); (b) shared
decisions with health care providers concerning major therapeutic and diagnostic
procedures (e.g., surgery, hospitalization,
chemotherapy); c) chronic disease
management (e.g., diabetes, asthma, hypertension); and d) lifestyle management (e.g.
exercise, stress, weight management, clinical preventive services).
There are over 60 million Americans enrolled in some form of managed care.
Capitation has increased in prevalence as mechanism for reimbursement. For managed
care to be ultimately successful, however,
it must mature beyond strictly a mechanism
for cost containment to one focused on measuring and managing value. Market
dynamics suggest that employers and government as the ultimate purchasers of care
5


and patients as the ultimate consumers of care are looking beyond just cost as a basis
by which they will select health care services. Even though cost will remain a primary
consideration, they are looking to other indicators of value and quality such as patient
satisfaction, access, convenience, and health outcomes.
With the advent of a more informed and assertive consumer, the health care
industry is ripe for consumer health and decision support services. The market
potential for TBDS has been estimated to be at least 100 million lives and as many as
200 million (Equitable Securities Research, 1996). However, despite the recent
growth of TBDS services, to date we know very little about their impact on the
quality of the doctor-patient relationship, clinical efficiencies, health outcomes, patient
satisfaction, health care utilization, health care costs, and malpractice liability. As
these types of services gain visibility and become more widely incorporated into health
care delivery systems as a "demand-side" strategy, an increasingly sophisticated
American population of consumers, patients, physicians, purchasers of medical care
(government and business), and insurers will pose questions about the medical efficacy
and efficiency of such services.
Physicians, in particular, are interested in clinically appropriate questions. Do
consumers/patients want more information about health and medical care decisions?
Do consumers/patients want increased invo
decisions? Will TBDS services improve the
vement in health and medical care
quantity and quality of information
6


desired by consumers/patients? If so, what impact will such services have on: the
quality of the doctor-patient relationship, clinical efficiencies, health outcomes,
patient satisfaction, health care utilizatioijL, health care costs, and malpractice liability?
Consumers and patients are desiring more information related to health and
medical care decisions. In 1988, Greenfie
prophetically stated, "We are entering an
knowledge among American patients.. .
d, Kaplan, Ware, Yano, and Frank
era of unprecedented sophistication and
In the everyday practice of medicine, will
involvement of the patient represent a critical disruption or will it promote a new kind
of relationship that enhances the roles of physician and patient in generating improved
outcomes (p. 457)? Anecdotal reports about TBDS decision support services have
indicated that they support the doctor-patient relationship, enhance patient satisfaction
and save money without sacrificing quality of care (Barnett, 1995).
Even though the opinion of the physician stands as a critical variable in
determining the ultimate impact such decilion support interventions will have on
physician-patient relationships, clinical efficiencies, patient outcomes, medical care
utilization, and costs, reviews of published literature revealed no systematic assessment
of physicians viewpoints about these types of services. Nor has anyone tried to
ascertain the implications of their opinions
on the implementation, utilization, and
ultimate impact of such services on health care quality, cost, and access factors.
7


Consumer health management support services, including TBDS, have
experienced rapid growth (Barnett, 1995; Equitable Securities Research, 1996).
However, there is a dearth of information regarding the opinion of physicians about
such services and the effect physicians o pinions might have on the ultimate
effectiveness of these services. This research is one of the first to: (a) Advance our
understanding about how physicians view enhanced participation of patients in medical
decision-making and the integration of TBDS services into care management and (b)
provide key insights on the implications for working with physicians in optimizing the
integration of such services in the delivery of health care.
Overview of Research Design and Setting
This research has been designed to be primarily descriptive and hypothesis-
generating. A multi-method approach using both quantitative (survey research) and
qualitative (key informant discussions) methods has been utilized to address the
specific aims of this study. This section provides an overview of the study design and
study site as well as a description of the te
ephone-based decision support services that
were implemented within the community where the physician subjects for this study
were practicing. Key questions explored in this study are presented. The
reconciliation of quantitative and qualitative data are discussed.
8


Study Design
A multi-method approach was chosen to enhance methodological rigor.
Patton (1990) has emphasized the value of triangulation or the use of multiple
methods in the research and evaluation of a given phenomena. Triangulation or
multiple methods can involve the utilization of different kinds of quantitative and
qualitative data and methods. Therefore, both quantitative and qualitative approaches
were employed concurrently to assess provider perceptions and opinions.
Survey research and key informant interviews were used to ascertain
physicians opinions about:
1. The increased involvement of patients as partners in health and medical
care decisions.
2. The impact they believe consumers/patients being more informed and
taking a more active partnership role in health and medical care decisions
will have on: the quality of the doctor-patient relationship, clinical
efficiencies, health outcomes, patient satisfaction, health care utilization,
health care costs, and malpractice liability
3. The integration of TBDS services into care management and the impact
they believe a TBDS service will have on: the quality of the doctor-patient
9


relationship, clinical efficiencies, health outcomes, patient satisfaction, health care
utilization, health care costs, and malpra ctice liability. Table 1.1 denotes the specific
questions from the Physician Opinion Survey (Appendix A) and the Physician Opinion
Interview (Appendix B) used to address each of these issues.
Table 1.1. Physician Opinion Survey and key informam interview questions.
Issue 1 Physician Opinion Survey Key Informant Interview
Patient involvement Bl, B2, B3, (ci, C2 1, 2, 4, 5, 6, 13
Doctor-patient relationship D4b 4, 5, 6, 9
Clinical efficiencies D4a 4, 5, 6, 9
Health outcomes D3, D4b 3, 4, 5, 6, 9
Patient satisfaction D3, D4b 4, 5, 6, 9
Patient retention D4b
Continuity of care D4b
Health care costs B2,D3 4, 5, 6, 9
Malpractice liability D3, D4a 4, 5, 6, 9
Physician reimbursement D4a
Utilization and appropriateness B2, D4a 4, 5, 6, 9
TBDS Service Dl, D2, D3, El [" 7,8, 9, 10,11, 12, 14
1 0


Study Site
The study site was the community of Rochester, New York (NY). Located in
upstate NY on Lake Ontario, Rochester has a largely stable population base (+3%
from 1980-1996) of just over one million with a demographic composition (16%
minority) comparable to similar sized urban areas in the Northeast United States. It has
an export-oriented, high technology manufacturing-based economy. Unemployment
rate is usually a percentage point or more lower than the U.S. average with the cost of
living slightly above the national average.
Prior to this study, the Rochester system received recognition as a model for
health care delivery. Rochester's success \yas attributed primarily to strong regional
planning and cooperation that focused primarily on supply side management initiatives
including: community rating and continuous open enrollment, near universal coverage,
technology assessment and control, limited hospital bed capacity, inter-hospital
cooperation, and a moderate managed care presence. Health insurance and medical
care were provided primarily by two carriers and their respective independent practice
associations (IPAs): (a) Blue Cross and Blue Shield of the Rochester Area
(BCBSRA) and the Rochester Community
IPA (RCIPA) which owned 73% of market
share, and (b) Preferred Care, an independent HMO, and the Rochester IPA (RIPA)
which owned 13% of market share at the time of this study. There were
1


approximately 2100 total physicians comprising both RCIPA and RIP A with 95% of
the total physicians providing services for both BCBSRA and Preferred Care.
Telephone-Based Decision Support (TBPS) Service
Approximately 80,000 insured families from seven major employer groups and
two major health plans in Rochester were
offered a TBDS service. This service had
just been implemented at the time of this study. Available 24 hours/day and toll-free,
the telephone-based nurse counseling service was staffed by nurse counselors trained
in assisting, educating, and supporting people faced with choices about health and
medical care. The counselors were required to be registered nurses with a minimum of
five years of current clinical experience and a minimum of a bachelor of science degree
in nursing.
Nurses who staffed the TBDS service received 240 hours of structured
education and training. The training focused on interactive learning, designed to
familiarize them with concepts of communication, communication skills, and
applications of these skills in interactions vrith callers. Training emphasized non-
directive, non-prescriptive techniques of communication as well as patient-centered
rather than provider-centered communications. Second, the nurse counselors were
trained in psychosocial assessment and self-care alternatives, and how to apply them to
12


the phone counseling process. Third, they received training on how to use computer
hardware and software which comprised the services information system.
The training curriculum for the nurse counselors was comprised of three
phases. Phase I consisted of 160 hours of initial training in communications,
information systems, and protocols for researching medical information. Phase II
embodied an intensive 80-hour mentorship with a senior nurse counselor which
included performance audits and feedback. Phase IQ incorporated another 8-10 weeks
of modified mentorship with a senior nurse counselor, also with performance audits
and feedback. Following this formal, initial training, nurse counselors received 60
hours of continuing education per year on
information systems, medical and health
topic reviews, and research protocols; par;idpated in weekly case review meetings;
and took part in quarterly in-service training sessions focused on customer service.
The primary objectives of the nurse counselors were to provide non-directive,
non-prescriptive decision support, self-management support, social support, simple
information requests, and referrals to local
agencies over the phone on virtually any
health and medical issue: (a) triage or pre-provider decisions concerning acute minor
illness or injuries (e.g., headache, back pain, sprained ankle); (b) major medical
decisions shared with health care providers (e.g. surgery, chemotherapy, advanced
imaging, hospitalization); c) chronic disease management (e.g. diabetes, asthma,


hypertension); (d) lifestyle changes to reduce risk of disease and injury (e.g. body fat
loss, exercise, stress management).
In addition to the TBDS service, all recipients of the service received the
following member communication materials: (a) an introductory packet (letter of
introduction, brochure, wallet cards, refrigerator magnet, initial quarterly newsletter)
regarding the telephone counseling service; (b) a copy of the medical self-care
reference guide, Take Care of Yourself; c) eight postcard reminders to use the service;
and (d) a quarterly newsletter. In February, May, August, and November, subjects
received the quarterly newsletter. The quarterly newsletters emphasized specific,
usable information for personal health and medical care decisions regarding pre-
provider triage, major medical, chronic disease management, and healthy lifestyles.
Each month in between newsletter issues,
subjects received one of the reminder
postcards promoting the benefits of calling the service for a wide range of health and
medical issues. The self-care reference guide featured clinical algorithms for over 100
common medical problems for adults and children. Promotional and educational
materials were supplemented with comprehensive and coordinated employer and
insurer communications through e-mail, voice-mail, employee letters, and newsletters
explaining and promoting the service as pint of the benefit plan being offered to them.
14


Physicians in the community were provided information about the TBDS
service as one component of a comprehensive and coordinated community-wide
communications plan developed and implemented by executive leaders from the
participating provider, employer, and insurer organizations. Five to six months prior
to the intervention, the governing boards as well as the various standing committees of
the two participating IP As were provided
with information about the community-wide
demand management initiative and the TBDS service. Three months prior to the
intervention, every physician in each of the IP As (RCIPA and RIP A) received a letter
from their respective elected physician executives explaining the service, asking for
their support and inviting them to participate in a community physician town
meeting planned for three weeks prior to the implementation ofthe TBDS. This
letter was accompanied by a roster of the physicians participating patients. Two
months prior to the intervention, each of the IP As as well as the county medical
society included an article about the commbnity-wide demand management initiative
and the TBDS service in their respective physician newsletters. During this same time
frame, a press conference describing the objectives of the initiative, the community
aspect of the project, and the TBDS service was held. A subsequent press release was
distributed to local and national media.
Three weeks prior to the start date
provided the opportunity to attend one of t
or the TBDS service, all physicians were
Wo community physicians town
15


meetings jointly sponsored by the county medical society and the two IP As. During
these town meetings, information regarding demand management, implementation of
the TBDS service, evaluation of the service, and employer perspectives were shared by
a panel of physician, employer, insurer, and vendor executive leaders. Physicians
opportunity to express their views and to
vo weeks prior to the intervention, a
attending the meetings were afforded the
have questions or concerns addressed. T
physician advisory panel comprised of physician executive and clinical leaders
representing the IP As and the county medical society were provided with the
opportunity to test the TBDS service.
Kev Questions
Important questions explored in this study included:
1. What are physicians opinions about patients wanting more information about
health and medical issues?
2. What are physicians opinions about patients wanting more involvement and taking
a more active role in the health and medical care decision-making process?
2a. Who should primarily make decisions about treatment for minor and major
health issues?
2b. What are physicians opinions about patients taking a greater role in the self-
management of chronic disease?
16


3. What are the opinions of physicians about the impact they believe a more
informed, active consumer/patient will have on: the quality of the doctor-patient
relationship, clinical efficiencies, heal
:h outcomes, patient satisfaction, health care
utilization, health care costs, and malpractice liability?
4. What are the opinions of physicians about the incorporation of telephone based
decision support (TBDS) for patients/consumers into the delivery of health care
services?
5. What are the opinions of physicians auout the impact they believe the
incorporation of TBDS into the health care delivery process will have on: the
quality of the doctor-patient relationship, clinical efficiencies, health outcomes,
patient satisfaction, health care utilization, health care costs, and malpractice
liability?
6. Do physicians opinions differ by age,
volume, and reimbursement structure?
7. What are the primary reasons physicians believed that the health plans implemented
TBDS services?
8. Which independent variables (age, gender, specialty, practice structure, visit
volume, and % of patients capitated) are the most significant predictors of
physicians opinions?
gender, specialty, practice structure, visit
1(7


Reconciling Quantitative and Qualitative lAnalvsis
The final step integrated the resul
s of both quantitative and qualitative
analyses to summarize physicians opinions about these key issues. Methods
triangulation is often accompanied by the
risk of divergent quantitative and qualitative
data. The subsequent challenge is developing a single, well-integrated analysis and
interpretation. For this study, steps were! taken to minimize this risk. The questions
for the Physician Opinion Interviews (qualitative data) were derived directly from the
Physician Opinion Survey (quantitative data). That is, the interview questions
obtained physicians opinions on the same issues as the survey, but were formatted to
derive more in-depth, information-rich understanding of that issue through the
interview process. Nonetheless, where discordance between quantitative and
qualitative data occurred, the quantitative data were used as foundation for
interpretation while the qualitative data provided additional explanatory power.
Organization of the Dissertation
This chapter introduces the signific;
ance of the study, states the specific aims,
provides an overview of the research design, and outlines the organization of what is
to follow. Chapter 2 presents a review of literature pertinent to health care
utilization, demand-side strategies for improving health care, physician and patient
18


perspectives about more informed and actively involved patients as partners in health
and medical issues, and a review of factors that influence the diffusion of new
innovations in health care. Chapters 3 anjd 4 are dedicated, respectively, to
quantitative and qualitative methods, analyses, and results. Chapter 5 presents a
discussion of the integrated quantitative and qualitative results including limitations,
implications, recommendations for future research, and conclusions.
19


CHAPTER 2
REVIEW OF LITERATURE
A review of literature pertinent to this study is presented in this chapter. The
following topics are discussed: (a) health care utilization, (b) demand-side strategies
for improving health care, (c) physician and patient perspectives about more informed
and actively involved patients as partners in health and medical issues, and (d) a review
of factors that influence the diffusion of new innovations.
Geographical Variations and Appropriate Use of Health Care
In the current debate on health care reform, three very salient and inextricably
linked questions central to health policy and strategy have been posited: (a) "Will we
need to ration effective care?" (Brook &
,ohr, 1986); (b)H Do the geographical
variations in the use of services have any clinical significance?" (Brook, Lohr,
Chassin, Kosecoff, Fink & Solomon, 1984); and c) "Is it prudent to focus exclusively
on how to manage the supply of medical services, or does a new concept demand
management deserve serious consideratic
Defining, measuring and classifyinj
delineated by Brook, Kamberg, Mayer-Oakes, Beers, et al. (1990):
n as well?" (Vickery & Lynch, 1995).
appropriateness has been recently
20


A service is provided app ropriately when its benefits
(expected positive consequences) exceed its costs
(expected negative consequences).... Based on
literature review, we have delineated three approaches
by which to judge whether a service is appropriate. ..
Benefit-risk approach... a service is judged to be
appropriate only if the hejilth benefits to the patient exceed
the negative health consequences to the patient by a wide
enough margin so that the procedure is worth doing. ...
Benefit-cost approach.. 1 extends the benefit-risk approach
by considering the resource cost of providing the service....
Implicit approach.. .standards forjudging appropriateness
may have been used, but no clear, explicit definitions are
specified.... Each of these approaches... can be used to
classify a service as being {underused, misused, or overused.
The underuse of a procedure, office visit, or hospital stay
is the failure to use these services when they are indicated....
Misuse implies that services are incorrectly used ....
Overuse is the use of a procedure, office visit, or hospital
Admission that is not indicated (p 3).
What have we learned from health services research about the appropriateness
of medical care utilization? Vickery and
ries (1994) have pointed out that up to 70%
of all physician visits for new problems may be medically unnecessary. Referencing a
significant and growing body of evidence from medical care utilization research,
nationally prominent health services research scientists, Brook and Lohr, estimated
that, "... perhaps one-third of the resources devoted to health care today are being
spent on ineffective or unproductive care"
(1990) stated that, "Virtually every study
double-digit levels of inappropriate care.
(1986, p. 6). After conducting a literature
review regarding the appropriateness of acute medical care for the elderly, Brook et al
included in this review found at least
'eriiaps as much as one-fifth to one-quarter
21


of acute hospital services or procedures were felt to be used for equivocal or
inappropriate reasons and two-fifths to one-half of the medications studied were
overused in outpatients" (p. 225). In addition to these studies of overuse, they noted
that the few studies that examined underuse or misuse of services also documented the
existence of these phenomena.
Furthermore, they noted that for t
le services examined in these
appropriateness studies, very little is known about the effects of such utilization on
health outcomes.
This can be viewed pessimistically as an indictment of
the current U.S. health care system or optimistically as
a window of opportunity. Even while awaiting more
data on appropriateness, m ich work needs to be done to
decrease the extent of inappropriateness in the U.S. health
care system. Depending on one's beliefs, that work could
take the form ... at the clinical level of better education or
dissemination of informatio
u to patients.. (1990, p. 240).
Eddy (1990, 1991, 1993) has tended to support this perspective by pointing out that
resolving our societal conflict with health care costs will require connecting value to
cost and addressing unchecked demand for volume and intensity of services by
involving patients as partners with physicians in clinical decision making.
Vickery and Lynch (1995) have prc
volume and intensity of service utilization t
posed one such opportunity for addressing
irough the concept of demand
management. They have defined demand management as, "The use of decision and
self-management support systems to enable and encourage consumers to make
22


appropriate use of medical care" (p. 4-5). Vickery and Lynch (1995) pointed out that
the supposition behind demand management is that health care decisions are influenced
by a complex set of psychological, social, and cultural variables, most of which are
unrelated to morbidity (nature and severity of the illness).
Similar to Brook & et al, (1990), they have defined appropriate decision
making as that which optimizes benefits while minimizing risk. However, Vickery and
Lynch (1995) have provided an intriguing, expanded, and intuitively sound
psychosocial-cultural dimension to appropriateness. They have indicated that two
components of demand, perceived need and patient preference, are central to demand
management and hold considerable potential for enhancing more appropriate
utilization of health and medical care services. For example, in regards to perceived
need, they have pointed to research by social scientists such as Berkanovic (1981) who
has estimated that 42% of the variance in decisions to seek health care can be
accounted for by psychosocial variables (personal beliefs, social networks etc.) with
morbidity explaining only 12%.
Recent evidence has indicated that
an individual's access to information and
resources is an important variable in health care decisions and subsequent health care
utilization. That is, a well-informed patient will make different decisions than a less-
informed patient. Since several of the mor; popular contemporary telephone-based
decision support services generally include pre-provider triage (self-care and perceived
23


need), shared decisions for major medical issues (patient preference), and chronic
disease management; the empirical basis for each of these decision-support
components will be reviewed.
Demand-side1
Interventions
This section will review the literature pertinent to various types of demand-
side interventions. Particular highlight will be given to the impact of these
interventions on health outcomes, patient satisfaction, health care utilization, health
care costs, cost-benefit, and clinical efficiencies
Perceived Need and Self-care
Vickery and Lynch (1995) have hypothesized that self-care interventions
directly influence an individual's perception of the need for health care by addressing
the individual's capabilities to assess and manage medical problems outside of the
formal medical care system. As noted by Vickery and Lynch (1995), there is a
considerable amount of evidence supporting the effectiveness of self-care interventions
in decreasing demand for medical care.
Vickery, Kalmer, Lowry et al. (1983) conducted a prospective, randomized,
controlled trial of self-care educational inteiventions within a health maintenance
organization to determine the effect of self-care written communication materials on
24


ambulatory health care utilization. Statistically significant decreases of 17% and 35%
were found for total medical visits and minor illness visits, respectively. They also
estimated that these decreases in health care utilization resulted in a savings of $2.50
to $3.50 for each dollar invested in the educational interventions.
Long, Kraines, Brown and Richardson (1985) conducted a study on the impact
of a workplace health education program on reducing medically unnecessary
outpatient visits. The purpose of the study was to evaluate a health education
program's ability to help employees maki more informed decisions about when to seek
professional medical care at the physicians office or the emergency room. The main
finding of this study was that the self-help intervention decreased self-reported
physician visits (17% or 2.0 visits/househjold/year) in first-dollar-fee-for-service benefit
plans. However, in contrast to the Vickery et al. (1983) study, no reduction was
found among HMO members. The researchers also examined whether or not these
decreases in visits resulted in poor medical outcomes. They tested this indirectly by
examining increased hospitalization and outpatient visit rates during treatment periods
of the study. They reported that, "Neither of these results was observed although the
power for detecting such changes in this sudy is not great... In future investigations
of self-care educational programs, more attention must be given to the long-term
health benefits .. ." (Long et al. 1985, p. 1053).
25


studied the impact of self-administering b
Utilizing a case-control pre-post study design, Levine and Britten (1973)
ood products in 45 hemophiliac patients
selected from a cohort of patients with congenital coagulation disorders routinely
followed through the New England Medical Center in Boston. The intervention
consisted of theoretical instruction for patients and family members on hemophilia
treatment and complications, instruction o
as suggested dosages for minor and major
n sterile technique and venipuncture as well
hemorrhage, and two to four supervised
infusions. The self-therapy intervention resulted in a reduction of 74%, 89%, 76%,
and 45% for absenteeism, hospital days, outpatient visits and health care costs,
respectively, for the 45 patients.
Most intriguing were the subjective evaluations completed by study
participants. Levine and Britten (1973) reported that, "All 45 patients indicated their
complete satisfaction and felt that home therapy had greatly improved their lives.
None wanted to change the program .... Numerous patients volunteered the
information that they were enjoying their relationships with the physician and hospital
staff for the first time. Several attributed this to their own participation in health care
and their decreased dependency needs" (pp
198-199). This patient feedback tended
to suggest that overall health status and quality of life improved significantly as a result
of patient participation in self care. However, except for the limited morbidity,
26


mortality, and this informal patient satisfaction survey information, no formal
evaluations of patient-assessed functional status or health outcomes were conducted.
Chronic Disease Management
Greenfield, Kaplan, Ware, Yano, and Frank (1988) studied the effects on blood
sugar control and quality of life in diabetics who participated in medical decision
making. The intervention consisted of a 20-minute coaching session while the patients
were waiting to see their physicians. The components of this session included: a)
following an algorithm to review with the patient the last recorded medical visit, b)
assisting the patient in identifying medical decisions most likely to arise during the
visit, c) encouraging the patient to focus on treatment decisions that could be affected
by lifestyle and/or preferences, d) rehearsing negotiation skills, e) encouraging patients
to ask questions of their physician, f) reviewing obstacles that can impede negotiations
with their physician, and g) rehearsing how to overcome these obstacles.
Glycosylated hemoglobin (HbAi c) values for the experimental group decreased
from 10.6 +/- 2.1% at pre-intervention to 9.1 +/- 1.9% (p<0.01) after the intervention.
There was no change in (HbAic) values for the control group from pre-intervention
(10.3 +/- 2.0%) to post-intervention (10.6
+/- 2.22%). These researchers reported
that the experimental patients were approximately twice as effective as controls during
post-intervention in the elicitation of information from physicians, reported
27


significantly fewer functional limitations (mobility, role, physical, lost work days, self
care), and assessed their health (overall perception of health, health concern, number
of current health problems) more favorab
Mayo, Richman, and Harris (1990;
y-
determined the effect of an outpatient
program designed to reduce re-admissions and hospital days for asthmatic
exacerbations among adults with asthma. The intervention was classified as an
intensive outpatient treatment program which consisted of a vigorous medical regimen
combined with an educational program which emphasized teaching patients aggressive
self-management strategies in case of marked asthma exacerbation. The patient
education program did not involve any written, lecture, or audiovisual materials.
Patient education was completed during repetitive contacts between the patient and
physician. Hospital readmission rates and
lospital day rates per patient per month
were reduced by 70% and 64%, respectively.
In commenting on the fact that written materials, lectures, and support groups
were not used in this study, Mayo et al. (1990) reported that,". . it rapidly became
evident that the face-to-face interaction of :he medical care provider and patient
resulted in the most meaningful transfer of information.... Preliminary results
suggest that a nurse practitioner is as effective as a physician in reducing readmission
rates among difficult adult asthmatics" (pp.
870-871).
28


Long, Mazonson, and Holman (1993) studied the effects of a four-year
Arthritis Self-Management Program (ASMP) on pain, physical disability, physician
visits, and cost-benefit in osteoarthritis (OA) and rheumatoid arthritis (RA) patients.
Studies were also conducted to ascertain
effects.
how the ASMP achieved the measured
Results indicated that change scores from baseline through four years for the
two ASMP groups were persistent and statistically significant (p<0.01). Despite
deterioration in physical abilities (mean increase in disability by 9%), the mean
decrease in pain and visits to physicians was 20% and 40%, respectively, over the four
perceived self-efficacy to cope with the
years. Concurrently, the mean increase in
consequences of arthritis was 25% above
the four-year baseline level. Financial
analysis indicated a four-year savings of $648 per RA patient and $189 per OA patient
resulting in benefit-cost ratios of 12:1 anjd 3.5:1 for RA and OA patients,
respectively.
Patient Preference and Shared Decisions
In what has become one of the most well-known and highly publicized patient
preference and shared-decision making interventions to date (Dentzer, 1991;
Winslow, 1992), patients with benign prostatic hypertrophy (BPH) who had been
provided with information regarding the benefits and risks of "watchful waiting"
29


versus elective surgery were more risk-aversive than physicians. When BPH patients
(n=284) at Kaiser Penmanente HMO in Denver viewed an interactive video describing
the benefits and risks of watchful waiting versus elective surgery for BPH, all but 30
chose watchful waiting over surgery. This patient preference for watchful waiting
over elective surgery approximated a 44% reduction in the rate of prostate surgery
among HMO members in the first year. Similarly, surgical rates initially plummeted
60% at Group Health of Puget Sound HMO in Seattle.
Lerman, Brody, Caputo, Smith, Lazaro, and Wolfson (1990) constructed a
Perceptions of Involvement in Care Scale
(PICS), a self-report questionnaire that
assessed patients perceptions of doctor and patient behaviors that occur during a
routine medical visit. Specifically, this instrument was designed to ascertain patient
perceptions about doctor facilitation of patient involvement, level of information
exchange, and patient participation in decision making. Subjects used in the
development of this instrument were primarily HMO patients attending a primary care
office at an academic medical center. The
and black (80%) with a mean age of 39 ye
A major finding of this study was i
subjects were predominantly female (62%)
ars.
lat physician facilitation of patient
involvement and actual patient involvement in decision making were significantly
related to patients satisfaction with care. Additionally, physician facilitation of patient
involvement and information exchange were consistently and significantly associated
3
0


with patients perceptions of post-visit cl ranges in their understanding, reassurance,
perceived control over their illness, and expectations for functional improvement.
Also, patients who reported that they were more involved in decision making with the
doctor were more satisfied with their physicians technical competence. These
findings reinforced previous research (Wasserman, Inui, Barriatua, Carter &
Lippincott, 1984) which demonstrated the important role physician facilitative
behavior can and should play in producing improved therapeutic benefits for patients.
An intriguing study by Brody, Miller, Lerman, Smith, and Caputo (1989)
examined the relationships between patients perceptions about the roles they played
during medical visits and their subsequent attitudes about their illnesses, treatments,
and self-rated improvements. This study involved 117 adult patients from a general
internal medicine faculty practice which provided adult primary care to a
predominantly HMO population. The mean age of the patients was 32 years; 62%
were female, 85% were minorities, 92% had HMO insurance, and a majority
presented with symptoms suggestive of re atively minor illnesses (sore throat, vaginal
discharge, musculoskeletal pain) at the time of the study.
Questionnaires assessing the roles
patients wanted to play during the medical
visit, the role they later perceived they played, illness outcomes, patient attitudes about
their illnesses, and patient satisfaction with their physician were administered before
the medical visit and then again one day and one week after the visit. Their physicians
31


also completed a questionnaire which provided information about the seriousness of
the problem and expected alleviation of patient discomfort and dysfunction one week
later.
Forty-seven percent of the patients reported playing an active role while 53%
of the patients reported playing a passive role. One week following the medical visit,
active patients reported statistically significant (p = 0.04) less discomfort, greater
alleviation of symptoms, and more improvement in their general medical condition
than did passive patients. One day post-visit, active patients reported less concern with
their illnesses (p = 0.04), a greater sense of control of the illness (p = 0.04), and more
satisfaction (p = 0.02) with their physicians than did passive patients. The researchers
concluded that their findings were consistent with others who have demonstrated the
impact that increasing a patients participation in care can have on improved disease
control, higher quality of life, and a more
satisfactory doctor-patient relationship.
They speculated that patients increased sense of control and self-efficacy may be
possible explanations for their findings.
Patient Preference for Information and
Involvement in Decision Making
The traditional doctor-patient relat
Parsonian, paternalistic approach whereby
ionship has been characterized by a
the patient has played a predominantly
passive, dependent role and the physician has assumed an autonomous, professionally
32


dominant role in the health and medical care decision-making process. However, a
variety of social, political, and cultural forces in combination with technological
advances in communication and information management point toward a changing
doctor-patient relationship characterized
t>y increasing consumer/patient participation
in decisions affecting both health and medical care services.
Although the previously referenced studies have provided evidence that the
inclusion of patients in the health and medical care decision-making process can result
in more appropriate use of health care resources, decreased health care costs,
increased patient satisfaction, improved control of disease, and better clinical and
functional outcomes; it is also important to address the fundamental questions related
to the consumer/patients role in clinical decision-making. That is, to what extent do
consumers/patients prefer to be more informed and what type of information is desired
about health and medical care decisions? Io what extent do consumers and patients
want to be more actively involved in what type of health and medical care decisions
once they have been afforded specific information? And to what extent should
physicians encourage active and collaborative patient participation in health and
medical care decisions?
Cassileth, Zupkis, Sutton-Smith, and March (1980) examined the degree to
which 256 cancer patients desired to be more informed and to participate in decisions
regarding their care. Using an Information Styles Questionnaire, patient preferences
3
3


regarding involvement in medical decisions were measured using two alternatives
one indicating a preference for their doctor to make decisions about their medical care,
the other indicating a preference to participate in decisions about their medical care.
Similarly, preferences for quantity and quality of information regarding diagnosis and
treatment were ascertained.
The most notable findings from tliii
1. Patients who desired detailed i
is study were:
brmation about their treatment in
infi
comparison to those who didnt seek such information were younger, white, better
educated, and more recently diagnosed.
o participate in treatment decisions were
those who preferred to leave treatment
2. Those patients who preferred
younger and better educated compared to
decisions up to their doctor.
3. Even though age was the only factor that distinguished those patients who
wanted more information and active involvement in their own care from those who
preferred a minimum of information and active involvement, a majority of patients in
each age group preferred active participation and detailed information.
4. There was a strong correlation between preferences for maximum
information and the desire to participate in decisions.
3
4


5. Because not all patients desired an active involvement in the medical
decision-making process, physicians should be attentive to patient preferences for such
roles.
Strull, Lo, and Charles (1984) examined to what extent 210 hypertensive
patients from a community hospital clinic an HMO, and a Veterans Administration
outpatient clinic preferred more information about hypertension and preferred to
participate in medical decisions. Concurrently, the patients physicians were asked to
estimate their patients preferences for more information about hypertension and
preference to participate in medical decisions. The researchers also studied whether or
not patient demographic and other health care related factors affected patient
preferences. Similar to other studies, Strull et al. utilized five different preference
alternatives encompassing active, collaborative, and passive decision making
styles to ascertain how decisions are usually made and patient preference for how
decisions should be made.
The major finding of this study was that clinicians underestimate patients
desire for more information and discussions about therapy but overestimate patients
desires to participate in decisions. That is
41% of the patients preferred more
information about their illness, but clinicians underestimated patient desires for
discussion about therapy in 29% of cases;
and while 53% of patients preferred to
35


participate in treatment decisions, clinicians estimated that their patients desired to
participate in decision making in 78% of
the cases.
In terms of how decisions are usually made, 63% of patients reported that the
clinician usually makes the decision while clinicians reported such decision making in
only 20% of the cases. Additionally, 80% of clinicians reported that the patient
participates to any extent in decisions while only 37% of patients reported that they
participate in decision making to any extent.
In terms of how decisions should be made, 47% of the patients preferred that
the clinician make the therapeutic decisions without patient participation.
Approximately one-third of the patients indicated that the physician should make the
decision while strongly considering the pa
ients opinion. Less than one quarter (19%)
of the patients indicated a preference to share equally with the physician in making
decisions and 3% indicated a preference for making decisions themselves. In contrast,
physicians believed that patients wanted to participate in decision making in 78% of
the cases. In only 22% of the cases did th
physician believe the patient wanted the
clinician to make the decision without the patient. Because of possible physician-
patient discordance about desires for more information and preferences for medical
decision-making, and, given the fact that enhanced patient participation may result in
improved outcomes, this study acknowledged that physicians may want to find explicit
36


ways to discuss these issues with patients in order to facilitate communication and
decision making.
Using a study population of 312 patients randomly selected from a hospital-
based primary care clinic, Ende, Kazis, Ash, and Moskowitz (1989) developed the
Autonomy Preference Index (API). The API measures patients preferences to be
informed and their desire to make medical decisions. The API consisted of two scales
one to assess information seeking and one to assess decision making. For each of
the two scales, totals were linearly adjusted to range from 0 to 100, where 0
corresponded to complete lack of desire for information seeking or for decision
making and 100 corresponded to the strongest possible desire for each of these
dimensions.
On the 0 to 100 scales for information seeking and preference for decision
making, the mean scores were 79.5 and 33
2, respectively. There were no significant
correlations between patients desires for information and their desires for decision
making or with patient satisfaction. Consistent with other studies, they found stronger
preferences for active involvement in decision making in younger, more educated,
higher income individuals, and less preference for involvement in decision making
among the more severely ill. Interestingly,
after adjusting for sociodemographic
factors and severity of illness in a regression model, only 19% of the variance in
decision making preferences could be explained. The authors concluded that patients
37


had a strong desire to be well informed, but, in support of other studies, relatively
weak preferences for participating in medical decisions.
The findings of this study tended
o support the view that patients prefer a
doctor-patient relationship whereby the doctor keeps the patient informed and
engaged in decision-making processes, but does not require the patient to function as
the principal decision maker. The researchers pointed out that a great deal of the
individual preference among patients for information and decision-making roles may
be more related to ingrained specific cultiral and ethical variables rather than the
sociodemographic factors measured in this study. To this point,
they, like Strull et al. (1984), acknowledged that physicians should have explicit
discussions with individual patients regarding information and decision-making
preferences.
Sutherland, Llewellyn-Thomas, Lockwood, Tritchler, and Till (1989) used a
convenience sample of 52 patients (35 women and 17 men) with a mean age of 48.5
years who were undergoing treatment for cancer (radiation therapy or chemotherapy)
to study their desire for information and preferences for participation in treatment
decisions. Preferences for information were assessed with two instruments the
Health Opinion Survey (HOS) and the Information Seeking Questionnaire (ISQ).
Preference for participation in treatment decisions was assessed using a questionnaire
adapted from Strull et al. (1984).
38


The results showed that although
sought information, 63% of the them felt
most patients indicated that they actively
the physician should take the primary
responsibility in decision making, 27% fe t decision making should be an equally
shared process, and 10% felt they should take a major role. The authors pointed out
that this discrepancy between patient preferences for information compared to their
preference for a limited role in decision making may be best explained by the patients
desire to satisfy an aspect of psychological autonomy that does not necessarily
include participation in decision making clinicians should be sensitive to situation-dependent preferences for information and
decision-making on the part of individual patients.
Another study assessed patient desire for information and involvement in
decision making using the Autonomy Preference Index (API) and the Health Opinion
Survey (HOS). The subjects were 167 university-based primary care outpatient
patients with benign prostatic hyperplasia,
back pain, or mild hypertension (Nease and
Brooks, 1995). Assessments from both instruments indicated that patients desire for
information exceeded that for involvement
in decision making. Consistent with the
findings by Ende et al. (1989), the API decision-making score in this study decreased
as the seriousness of the clinical condition increased. That is, as the seriousness of the
hypothetical clinical condition increased, patients desire for involvement in decision
making decreased.
39


Consistent with previous studies,
substantial variation among patients in their desire for information and for their
this research pointed out that there is
involvement in medical decisions. That is,
although sociodemographic factors
(younger age, female gender, higher education) may be associated with greater desires
for information and decision making, a large part of the variance for patient responses
to these two aspects of medical care remains unexplained. Similar to others (Ende et
al. 1989; Sutherland et al. 1989; Strull et ll. 1984; Cassileth et al. 1980), these
researchers pointed out that even though patient desire for information is high, the
substantial variation among individual patients for involvement in decision making
should serve as a caution against rigid policies and recommendations about the degree
to which patients should be involved in medical decisions.
Beisecker and Beisecker (1990) studied 106 rehabilitation medicine patients
(42 men and 64 women ranging in age from 17-85 years) to determine information
seeking behaviors and decision-making preferences using the Desire for Medical
Information and the Locus of Authority in Medical Decision-making Scales. The
results indicated that patients desired information about a wide range of medical
topics, but when communicating with their physician they did not engage in many
information-seeking behaviors. Furthermore, patients believed that medical decision
making authority should rest primarily with
the physician rather than with the patient.
Finally, it was noted that patient information-seeking behaviors were more closely
40


associated with situational variables such as length of interaction, diagnosis, and
reason for the visit than with patient attitudes or sociodemographic characteristics.
The researchers surmised that information-seeking behavior was attenuated in
this study because the subjects were primarily patients with chronic diseases and
disabilities. Like others, they also emphasized that situational factors (diagnosis,
presence of a companion, reason for visit etc.) surrounding the doctor-patient
relationship may better explain patient information seeking behaviors and desires to be
involved in medical decision-making than patient sociodemographic variables. They
suggested that future research examine the impact of situational variables on
information seeking behaviors and preference for involvement in medical decision-
evaluated patient preferences about
making.
Bradley, Zia, and Hamilton (1996)
involvement in medical decision making and factors that might influence these
preferences in a predominantly white female population. They used a survey
instrument that included 10 specific case scenarios depicting different acute and
chronic medical situations that varied in severity. For each scenario, patients were
asked to select one of seven responses suggesting varied levels of physician/patient
decision making. These responses ranged rom very passive (I prefer that my doctor
tell me what to do) to collaborative (I prefer that my doctor and I make the decision
together) to very active (I prefer that I make the decision without any information or
4


recommendation from the doctor) preferences for involvement in decision making by
the patient.
The survey responses in this study1 indicated that patients favored the physician
as decision maker. However, the preferences for involvement in decision making
varied by scenario. The authors acknowledged that patient decision making
preferences were scenario specific and probably illness specific. Patients wanted less
control in the hypertension scenario and more control in the cancer scenario.
Consistent with other studies (Cassileth et
al. 1980; Strull et al. 1984; Ende et al.,
1989), increased age and lower educational levels were related to increased emphasis
on the physician as primary decision maker whereas lower age and increased
educational levels were related to increase d patient control. Similar to others (Nease
and Brooks, 1995; Ende et al. 1989; Sutherland et al. 1989; Strull, 1984; Cassileth et
al. 1980), this study suggested that: a) pre:
making depends upon the situation and b)
erence for patient involvement in decision-
physicians must learn to be both facilitators
of decisions as well as decision makers depending upon the patient and the situation.
Hack, Degner, and Dyck (1994) examined the quantitative and qualitative
relationships between preferences for specific medical information and preferences for
involvement in making treatment decisions
in 35 women (aged 32-83 years) with stage
I and II breast cancer who were 2-6 months post-diagnosis. Using a card sort
technique developed by Degner and Sloan (1992), they explored patient preferences
42


for specific medical information and preferences for involvement in making treatment
decisions. This card sort technique provided patients with five different degrees of
preferred involvement in medical decisions ranging from active (I prefer to make the
final selection about which treatment I receive) to collaborative (I prefer that my
doctor and I share responsibility for deciding which treatment is best for me) to
passive (I prefer to leave all decisions re
gar ding my treatment to my doctor). They
also developed a card sort measure to determine patient preferences for diagnostic,
prognostic, and treatment-related information desired by the patient. The information
preference card sort procedure addressed the specific manner by which patients want
physicians to provide them with information (quantity, detail, media, type of
descriptive terminology to be used).
Overall, the findings suggested that patients who desire an active role in
choosing their treatment also desired more detailed information about their medical
care. That is, those patients who preferred a more active preference for involvement
in decision making also preferred to receiv e more detailed information regarding their
care while those patients who preferred a more passive style preferred either a brief
explanation or did not want any information at all. However, consistent with other
studies regarding patient variability (Ende et al. 1989; Sutherland et al. 1989; Strull et
al. 1984; Cassileth et al. 1980), Hack et al.
pointed out that there were nine passive
patients who wanted detailed information regarding their treatment.


Second, all of the patients who pr
ferred active involvement in decision making
and most of the passive patients wanted their physicians to provide them with all
treatment alternatives and the degree of risk associated with each alternative. Active
and collaborative style patients wanted complete information about all major
alternatives while the passive patients preferred to hear about the best single treatment
or hear nothing at all.
Third, education was the only sociodemographic variable significantly related
to decision-making role preference. Those patients who preferred an active decision-
making role were more likely to have graduated from high school whereas the non-
high school graduates preferred a more passive role. In contrast to other studies, age
(Cassileth et al. 1980; Strull et al. 1984; Ende et al., 1989), and illness severity
(Cassileth et al. 1980; Ende et al. 1989; Bradley et al. 1989), were not significantly
related to decision-making role preferences.
Fourth, the qualitative analyses disclosed that: a) patients actively engaged in
decision making wished to play such a role because it affords them a greater degree of
control over body, health, and life; b) the more passive patients wanted to be kept
informed, but at the same time it was important to place faith and trust in then-
physicians because only physicians possess
necessary to formulate treatment decisions
the medical knowledge and expertise
(p. 286); c) active patients were more like
to believe their physicians wanted them to be more active in medical decisions; d)
44


being pressured to play a more active role in decision-making processes was more
anxiety-producing for the passive patien:s; and e) both active and passive patients
confirmed that neither role is superior the best role is specific to the individuals
style and circumstances.
To this point, Hack et al. noted that several patients emphasized that physicians
needed to be sensitive to and respect the information and decision-making preferences
of each individual patient. Preferences for information and decision-making may differ
depending on milestones related to seventy of illness, elapsed time since diagnosis, and
treatment progression. Similar to other studies, the younger and better educated
patients expressed a greater desire for involvement.
Using the same card sort procedures, Davison, Degner, and Morgan (1995)
assessed information and decision making preferences of 57 men (mean age 71 years)
with prostate cancer who were 0-6 months post-diagnosis. The majority of the men in
this study were married (74%) and had less than a high school education (54%).
Twenty-nine (51%) had received no treatment.
Consistent with another recent study by Degner and Sloan (1992), a majority
of men (58%) in this study preferred a passive treatment decision-making role, while
23% and 19% preferred a collaborative and an active role, respectively. However a
high proportion of men wanted more information regarding likelihood of cure, types of
treatment available, and disease progression.
45


Speculations as to why the majority of men preferred to delegate treatment
decisions to their doctor included: 1) olde
r married male patients tend to involve their
wives in medical care decisions and subsequently tend to relegate to them a decision-
making role; 2) older patients may be less
authoritarian role of the physician; and 3)
likely to believe they should challenge the
older patients may actually prefer a more
authoritarian model in the doctor-patient relationship because they have a high level of
trust in their personal physician; 4) there may be a fear of assuming responsibility for
decisions that may lead to poor outcomes;
and 5) they may believe they lack the
knowledge and information necessary to make a good decision.
As has been true of most of the studies reviewed in this section, Davison and
colleagues (1995) point out that information seeking and decision-making preferences
are determined by a variety of sociodemogjaphic, intra-personal, and situational
factors. Thus, clinicians should be explicitly sensitive to individual differences in
regard to needs for information and involvement in medical decisions. Davison and
associates (1995) also sounded an appropriate call for the critical and urgent
need for research in determining the role that wives and the family unit play in health
and medical care decisions.
Degner.
Beaver, Luker, Owens, Leinster,
Degner and Sloan (1992) card sort procedup
in treatment decision making in 150 women
, and Sloan (1996) also used the
e to examine the degree of control desired
(mean age 54.8 years) newly diagnosed
46


(2.5 weeks) with breast cancer. Their primary aim was to ascertain whether these
women preferred to play an active (patie
nt-controlled), collaborative (jointly-
controlled between patient and physician^, or passive (provider-controlled) role in
decisions regarding breast cancer treatment. For a majority (66%) of the newly
diagnosed women, lumpectomy was the planned treatment with mastectomy the
planned treatment for the remaining 34%. Two hundred women with benign breast
disease served as a comparison group.
The most significant finding of this study was that a majority of the newly
diagnosed women preferred a passive role, leaving treatment decisions to their doctor;
whereas the women with benign breast disease (in the comparison group) preferred a
more collaborative role. For the newly diagnosed group, the variables found to be
weakly correlated with decision-making preference were age, with older women
preferring a more passive role; education, with lower levels of education associated
with preference for a more passive role; ar d social class, with lower social classes
preferring a more passive role. However, in regression analysis, age and level of
education were not predictive of decision making preferences. Even though social
class was predictive of decision making preferences (p = 0.04), its influence was
relatively weak accounting for only 2% of the variance. In the benign group, age and
social class were weakly correlated with preferences for decision making. A family
47


history of breast cancer and marital status;
preferences in either group.
Beaver and colleagues speculated
Using supportive information from other
were not associated with decision making
about the explanations for these findings,
cancer studies and preferences for decision
making, it was argued that potentially life threatening illnesses may make patients
more passive. They also noted that even
hough a majority of the women with newly
diagnosed breast cancer preferred a more passive role in decision making, there were
still many who wanted some form of involvement. They cite that a lack of appropriate
information could also be one of the reasons that women felt uncomfortable in playing
an active role in decision making.
Similar to previous research cited in this review, they emphasized once again
that the most prudent clinical approach is for the clinician to be sensitive to and take
into account each individual patients preferences for decision making involvement.
They reiterated that there seems to be no simple predictive formula for determining
patients preferences for decision-making. To assist clinicians with such assessments,
they suggested that the card sort procedure
tool into the everyday practice of medicine.
could easily be integrated as a screening
The authors indicated that future studies
were underway to determine the effect of time since diagnosis on decision-making
preferences.
48


A particularly interesting and noteworthy study was conducted by Thompson,
Pitts, and Schwankovsky (1993) regarding the situational and demographic influences
on the preference for involvement in medical decision-making. Participants for the
study were 459 randomly selected members of an HMO who were predominantly
white (87%), female (68%), low to middl e income (49%), married (65%), had some
college (63%), with a mean age of 54 years. To assess patient preference for decision
making they used the Ende et al.(1989) Autonomy Preference Index (API) and the
Desire for Involvement Questionnaire (DIQ) they developed specifically for this study.
Thompson et al. (1993) underscored the significant difference between the two
instruments. The API is constructed to ask about patient preference in regard to
medical situations that generally fall within the province of a medical professional.
The DIQ is designed to ascertain desire fol input into situations where medical
expertise is not required. Four vignettes of varying severity (knee injury, high blood
pressure, cancerous growth, and stress hea daches) each requiring two treatment
choices with varying side effects, trade-off between quality and length of life, and
effects on patients lifestyle were used to assess preferences.
As hypothesized, one situational factor was found to be strongly related to
patient desire for involvement. Respondents expressed a greater desire for involvement
in decisions that did not require medical expertise (DIQ) and significantly lower
preferences for involvement in decisions that required medical expertise (API). With
49


this finding, Thompson and colleagues su ggested that the conclusions reached in other
studies (Strull et al. 1984; Ende et al., 1989; Sutherland et al. 1989; Beisecker and
Beisecker, 1990) about patients not wanting to be involved in medical decisions are
misleading.
Additionally, in an analysis to see if experience with a condition impacted the
desire for involvement, they found that those who had hypertension were less likely to
want decisional involvement than those who did not have hypertension. They
speculated that experience with a chronic condition and all of its associated difficulties
may lessen the desire for decisional responsibility.
From a clinical practice viewpoint,
out the documented advantages of having
Thompson and associates (1993) pointed
patients participate in medical decisions
while at the same time conceding that clinicians need to recognize that there are
individual patient differences in decision-making preferences. They urge clinicians to
be sensitive to these differences and subsec
patient involvement is most likely to be pre
In summary, it is clear that patients
uently to the types of decisions for which
'erred and beneficial.
prefer and desire information about the
diagnostic, prognostic, and therapeutic aspects of medical decisions. What isnt so
clear is to what extent patients want to be actively involved in medical decisions. The
literature reviewed here has presented conflicting findings. The evidence has
suggested that there are certain sociodemographic and situational factors which have
50


been consistently associated with decisio n-making preferences: 1) age, with younger
individuals preferring a more active role; 2) education, with more highly educated
individuals preferring a more active role; 3) gender, with men preferring a more
passive role than women; 4) marital status, with married men preferring a passive role;
5) prognosis, with more seriously ill patients preferring a more passive role. Two
critical variables that need further study include time since diagnosis and the role that
wives and the family unit play in health and medical care decisions.
It is clear from this literature review that information seeking and decision-
making preferences are determined by a variety of sociodemographic, intrapersonal,
and situational factors. The most prudent clinical approach is for the physician to be
sensitive to and take into account each individual patients preferences for information
and for involvement in decision-making.
Finally, the intriguing work by Thompson et al. (1993) indicates that there may
be some significant methodological issues associated with how we are assessing
patient preference for involvement in medical decision-making. That is, respondents in
their study expressed a greater desire for involvement in decisions that did not require
medical expertise as measured by the Desire for Information Questionnaire (DIQ) and
significantly lower preferences for involvement in decisions that required medical
expertise (API). Like the API, the Locus of Authority Scale utilized in the Beisecker
and Beisecker (1990) study also required medical expertise to answer many of the
5


questions. What information should be covered in the patients history? Whether there
is a need for medication? Which drug to use? What dosage to use? Whether to change
medications? Whether to have the patient undergo diagnostic tests? Whether to treat
the patient in the hospital or on an outpatient basis? Thompson et al.s (1993) opinion
that misleading conclusions have been rea ched in other studies (Strull et al. 1984;
Ende et al., 1989; Sutherland et al. 1989; Beisecker and Beisecker, 1990) about
patients not wanting to be involved in medical decisions warrants important
consideration.
Physician Perception about Patient Desire for
Information and Involvement
There is a substantial body of evidence regarding patient preference for
However, there seems to be a coinciding
perception about how active patients
information and medical decision making,
scarcity of information regarding physician
should be in seeking information or in participating in health and medical care
review some of the more pertinent
decisions. The purpose of this section is to
physician information that may provide further insight on the matter.
As noted previously, Strull et al. (1984) examined to what extent 210
hypertensive patients from a community hospital clinic, a free-standing, HMO and a
Veterans Administration outpatient clinic preferred more information about
hypertension and preferred to participate in
medical decisions. Concurrently, they
52


determined how well clinicians estimated their patients preferences about these
aspects of medical care and whether or not patient demographic and other health care
related factors affected patient preferences. In terms of how decisions are usually
made, 63% of patients reported that the clinician usually makes the decision while
clinicians reported such decision making in only 20% of the cases. Additionally, 80%
of clinicians reported that the patient participates to any extent in decisions while only
37% of patients reported that the patient participates in decision making to any extent.
In terms of how decisions should be made 47% of the patients preferred that
the clinician make the therapeutic decisioJs without patient participation. Approx-
imately one-third of the patients indicated that the physician should make the decision
while strongly considering the patients opinion. Less than one quarter (19%) of the
patients indicated a preference to share equally with the physician in making de-cisions
and 3% indicated a preference for making decisions themselves. In contrast,
physicians believed that patients wanted to
the cases. In only 22% of the cases did the
participate in decision making in 78% of
physician believe the patient wanted the
clinician to make the decision without the patient. Because there was relatively little
demographic information characterizing the physicians in this study, very little else
about physician perceptions regarding patient information seeking and involvement in
decision-making can be extrapolated from this information.
53


Doctor-Patient Communication & Relationships:
Challenges/Bamers and Effect on Outcomes
One of the very appropriate and valid concerns that physicians have when
health plans begin to consider the implementation of 24-hour telephone-based nurse
counseling (TBDS) services is, What impact will this service have on the doctor-
patient relationship? Preserving the physician-patient relationship in an era of
managed care has become a central concern for most physicians (Emanuel and Dubler,
1995). In studying the perceptions and opinions that physicians have about
contemporary TBDS services, its imperative that we have a fundamental
understanding of the communication dynamics, challenges, and success factors which
comprise the doctor-patient relationship.
Traditionally, the doctor-patient relationship has been predominantly a
paternalistic one in which the patient has assumed a very passive, dependent, and
cooperative/acquiescent role while the physician has assumed a very active/directive,
autonomous, and professionally dominant/controlling role. Recently, however,
ethical, social, economic, and legal concerns in combination with cultural shifts and
advancements in communications and information management technology have begun
yet another era in the social transformation of American medicine (Starr, 1982).
This era is advocating a more informed consumer of health and medical care; more
patient autonomy and self-determination in medical decisions; and a mutual,
54


Collaborative partnership between physicians and patients in communicating with one
another and in subsequently making health and medical care decisions.
In their review of the literature on doctor-patient communication, Ong,
DeHaes, Hoos, and Lammes (1995;p. 903) stated that, Communication can be seen
as the main ingredient in medical care. In their treatise, they examined this issue in
three domains: purposes of medical communication, communicative behaviors, and
the influence of communicative behavior^ on patient outcomes. This will serve as a
useful framework for our review.
Doctor-Patient Communications: Challen
zes/Bamers
Ong et al. (1995) identified three
between doctor and patient: 1) to develop
different purposes of communication
an inter-personal relationship, 2) to
exchange information, and 3) to make decisions. Since perspectives on decision
making have been previously reviewed in ihis report, attention will be focused
primarily on physician-patient inter-persoqal relationships and the associated
information exchange processes.
The development of interpersonal relations and the information exchange
process has included the actions of gather! ig, giving, receiving, interpreting, and
evaluating verbal and non-verbal information by both patients and doctors. Physicians
and patients seek and need information for
obvious reasons. Physicians need different
55


types of information for diagnostic, prognostic, and treatment planning purposes. As
reviewed previously, patients have a strong desire and preference to know as much
about their medical situation as possible
patients need for information by stating,
have to be met when visiting the doctor:
what is the matter, where the pain comes
understood (to know the doctor accepts
Ong et al. (1995; p. 904) have captured the
From the patients point of view, two needs
the need to know and understand (to know
from) and the need to feel known and
him and takes him seriously). Speedling
and Rose (1985; p. 116) have supported this perspective by stating that, The
physicians expertise in curing the illness is rarely seen to count more than his or her
facility to care for the patients psychosocial needs.
Several factors have contributed to documented discordance in information
exchange between patients and physicians
Observational research (Beckman and
Frankel, 1984) has indicated that the physician interrupts the patients beginning
narrative after only 18 seconds; and the first complaint presented by the patient may
not be the most critical one! It has been noted that 60-80% of medical diagnoses and
treatment decisions can be made without the need for expensive diagnostics just on the
basis of a complete history and listening to
Harrison, Mitchell, Prichard, and Seymour
what the patient has to say (Hampton,
1975; Sandler, 1980). Sir William Osier,
the father of modem internal medicine, astutely implored physicians to listen for the
patient will reveal the diagnosis (Osier, 1899).
56


In support of the previous observations, DiMatteo (1994) outlined four
barriers or limitations to physician-patienjt communications: 1) physician use of
medical jargon resulting in patient lack of understanding and misunderstanding; 2) lack
of active listening to patients by physicians; 3) the desire the patient has for
information is usually not fulfilled by the provider; and 4) societal expectations and the
influence of medical tradition make it difficult for patients to effectively assert their
needs for information and participation in
decision-making. Facilitating patients in
telling their story about their illness experience and then actively listening to what they
have to say are critical success factors for
optimizing the doctor-patient relationship.
As noted previously, patients almost always want as much information as
possible about their situation. However, because physicians have tended to
underestimate this desire for information (Strull et al. 1984), they may have
subsequently provided insufficient informa tion to the patient to fully satisfy his/her
needs. Ong et al. (1995), like DiMatteo (1994), have pointed out that another
contributor to patient dissatisfaction with the exchange of information is rooted in a
lack of concordance between patient perception and physician perception about the
type of information exchanged. Physicians often view medical information as that
which is objective (diagnostic test results, prognosis, type of disease, treatment
procedures) whereas patients define medical information in terms of its personal
relevance to their illness experience (what is the likelihood of cure? what is the effect
57


of treatment on my job? social life? family? sexual activity? risk of disease to my
family?).
To this point, Frederikson (1995) has emphasized that a more effective
strategy for physician consultation with patients is one which attends to the concepts,
perceptions, and views of the patient rather than providing vast amounts of
standardized information. Utilizing her model of information exchange for medical
consultations, she studied information-exchange tasks in a self-selected sample of 35
general practice doctor-patient dyads.
The consultation evaluation instrument was a questionnaire which asked the
patient to rate the level of importance (essential, desirable, unimportant) for the
following information exchange tasks: exploring patient ideas, dealing with patient
concerns, exploring patient expectations r
the effects of problems, explaining the etiology, explaining treatment, discussing side
effects, and outlining a prognosis. The fo
sgarding diagnosis and treatment, discussing
patients: satisfaction, patient understanding of health problems, doctors response to
the patient, perceptions of appropriateness
owing outcomes were also rated by the
of treatment offered, treatment intentions,
development of the relationship between doctor and patient, and change in concern
and information provision.
The information exchange tasks most strongly correlated with patient
satisfaction were: dealing with patient concerns, exploring patient expectations,
58


exploring patient ideas, and discussing effects of problems. Fredrikson (1995) noted
that the quality of information provided by physicians to patients is more important
than quantity and that physicians should tjarget information toward the specific
individual needs of the patient.
von Friederichs-Fitzwater, Callahan, Flynn, and Williams (1991) analyzed 30
transcripts of conversations between physicians and patients in four different settings
(AIDs clinic, hospice program, family practice clinic, hospital room/doctors office).
The purpose of this study was to identify relational control patterns in the
communications between physicians and patients. The primary emphasis was on how
(relational dimension) physicians communicate with patients rather than on what (the
content dimension) physicians communica
A total of 5,968 utterances were o
ed to patients.
ded for format (assertion, question, talk-
over, noncomplete and other) across nine relational categories (support, nonsupport,
extension, answer, instruction, order, disconfirmation, topic change, and initiation-
termination). Communications between physicians and patients were concentrated
into three of the nine possible relational categories: extension, support, and answer
with predominantly symmetrical exchanges
The researchers concluded that the
interactions were primarily doctor-centerec. or disease-centered as opposed to patient-
centered as the physicians frequently changed the topic, asked more questions and
talked more than the patient.
59


Further analysis also revealed that there was a tendency on the part of patients
toward a more domineering behavior. However, further analysis of the conversations
reflected a lack of the dialectic labor of understanding, of listening and
interpreting, of clarifying and acknowledging what has been said, and responding (p.
30) by the physicians. Additionally, when the patients in this study raised an emotional
issue, physicians tended to respond with a question about physical symptoms or
change the subject, therefore, controlling Ihe meaning of the illness.
The findings of this study supportid previous relational research and the
frequent complaint by patients that they are not permitted to tell their stories. The
authors suggested that physicians could develop a more patient-centered partnership
model of care by allowing patients to talk about their illness experience; providing
more empathy and direct support; and facilitating patient communication around topics
which are particularly meaningful to the patient such as feelings and expectations.
Suggestions for further research in relational communications included: the
significance of gender on relational communications and the influence of
physician gender on patient credibility; the influence of physician attitudes about
certain topics; and how the physician-patient relationship develops over time.
Lerman, Daly, Walsh, et al. (1993) studied 97 patients (mean age of 55 years)
with Stage I or Stage II breast cancer who had all undergone primary surgical
treatment and who were scheduled to receive postoperative radiation or
60


chemotherapy. This study evaluated patient-provider communications in relation to
patient coping styles and psychological distress.
Patient-provider communications were evaluated using the three subscales
from the Medical Interaction factor of the Cancer Rehabilitation Evaluation System
(CARES): 1) Problems Obtaining Information from Medical Team; 2) Control of
Medical Team; and 3) Difficulty Communicating with Medical Team. The percentage
of patients reporting one, two or three, four to six, and seven to ten communication
problems were 84%, 39%, 35%, and 10%, respectively. The most commonly reported
problems were difficulty understanding plysicians (49.5%), difficulty expressing
feelings (46.3%), desire for more control over the medical team (45.3%), and difficulty
asking physicians questions (42.6%). The se patients who were more distressed, who
were less optimistic and who felt more helpless about their disease were most likely to
report communications problems.
Even though it was not possible to
relationships, Lerman et al. (1993) specula:
determine the direction of causal
ed that one mechanism that might be
responsible for the relationship between communications problems and psychological
distress was that patients who were able to
concerns were more likely to elicit informa
ask questions and voice their needs and
ion and receive personalized advice, which
reduced uncertainty and provided reassurance. They suggested that both patient-
oriented interventions and physician-oriented interventions which enhance physician-
6


patient communications may be of value
treatment. Patient-oriented interventions
m improving patients adjustments to
would include those interventions that
enhance patient skills in asking questions and expressing concerns. Finally, they
recommended that future research includ^ prospective randomized trials which
evaluate the impact of interventions to increase patient and/or physician
communication skills.
Not only are patients frustrated with physician-patient communications, it also
appears that physicians may be just as frusjtrated. Levinson, Stiles, Inui, and Engle
(1993) undertook a study of the nature of frustrating visits from the practicing
physicians perspective. This study examined feelings of frustration with patient visits
in 1,076 practicing physicians who were attending a voluntary workshop on physician
patient communications.
Analysis from the 25-item questionnaire developed specifically for this study
revealed several key findings:
1. Thirty six percent reported frustration in up to one quarter of patient visits.
2. The perceived cause of the frustrating visits was most frequently attributed
to the patient in more than 50% of the frustrating visits.
3. The most significant patient-attributed communication problems were: lack
of trust or agreement by the patient, patient presenting with too many
problems, lack of patient adherence, lack of understanding on the part of
62


both patient and physician, di
manding or controlling patient, patient had
special problems (alcohol, drugs, chronic pain).
Another communication problem identif|ed by physicians in frustrating visits was a
feeling of distress on their part during or ifter seeing the patient.
For each of the seven communicai ion problems identified by the physicians,
Levinson et al. (1993) discussed how each could be addressed by developing
communication skills that are oriented toward a more patient-centered, collaborative,
mutual participation model of clinical care. Characteristics of this model included: use
of open-ended questions, allowing the patient to talk uninterrupted at the outset,
active listening, repeating and summarizing the patients problems, avoiding the use of
medical jargon, and involving the patient in decision-making about treatment options.
They particularly noted the danger of the tendency for physicians to blame the patient;
and, concurrently emphasized the importance of discussion and education for
physicians in enhancing physician-patient clmmunications.
White, Levinson and Roter (1994) provided insight into one of the most
common challenges encountered by practicing physicians and one of the frustrations
identified by physicians in the Levinson et al (1993) study: the oh by the way...
phenomenon which occurs at the end of the medical visit. The objectives of the White
et al. (1994) study were to define and describe the physician-patient communications
during the closing phase of the medical visit
and to identify the types of
63


communication during the visit that might have contributed to the introduction of a
new problem at the close of the visit.
Audiotapes of 88 outpatient visits
with 20 primary care physicians were
analyzed for communication patterns using the Roter Interactional Analysis System.
The patients were predominantly white, highly educated (80% had some college
education), females (61%).
Closure was initiated by physicians
whether or not there were any more questions 25% of the time. In 21% of the
86% of the time and physicians asked
closures, patients presented new problems
not previously discussed. Both physician
and patient behaviors throughout the visit were associated with the presenting of new
problems during closure. The specific phj
sician behaviors associated with the
reduction in the occurrence of new problems at closure were: orienting the patient to
the flow of the visit, giving patients more information about the treatment plan, and
asking about patient beliefs. The specific patient behaviors associated with reduced
occurrence of new problems at closure wers: talking more about their therapy and
sharing emotions and feelings. Interestingly, the findings of this study are closely
related to the observational studies by Beckman and Frankel (1984) demonstrating
that patients are interrupted on average 18 seconds into their opening narrative; and,
the first problem presented may not be the primary problem.


Similar to other suggestions and
models for a more participatory, collaborative,
patient-centered approach to medical care, White et al. (1994) proposed a set of
communication skills to use before and during closure which may be helpful to
physicians in decreasing the frequency of new problems presented at closure. Before
closure: orient patient to the flow of the visit, address emotions and psychological
issues, explore patient beliefs, check for understanding, ask if the patient has additional
questions. During closure: orient the patient to the closure, provide a summary of the
visit, complete a final check for understanding, clarify the plan, provide encouragement
and reassurance, and consolidate rapport/caring.
Finally, a noteworthy practical observation was made regarding the concern
with time pressures in the everyday practice of medicine. These researchers pointed
out that instead of asking open-ended questions and whether or not the patient had
additional questions during closure, it may be more efficient to ask these questions
during the visit and before closure begins. Implicitly embedded in this observation is
the possibility that by using the aforementioned communication skills, overall clinical
efficiencies in terms of total encounter time may be achieved as well.
Doctor-Patient Communications: Effects on Outcomes
We have entered into an era of accountability in health care where the ultimate
purchasers of care (government and employers) are placing emphasis on value
65


enhancement. The notion of value enhancement recognizes that the cost of delivering
medical care and the associated resource
jtilization can be reduced even as quality is
improved. Because quality and value enhancement may become central to provider
differentiation and competitive advantage
in the market place, the importance of
providers understanding the impact of physician-patient communications on quality,
value enhancement, and outcomes is paramount.
Different dimensions have been used to define patient outcomes: satisfaction,
clinical indicators, compliance/adherence, quality of life, health status, patient-derived
measures of functional status, etc. The purpose of this section is to review some of
the impacts of doctor-patient communications on various outcome dimensions.
In their review of the literature, Onj
salient findings regarding the relationship o
et al. (1995) noted some of the more
f patient satisfaction, compliance, and
health outcomes to physician-patient communications:
Patients are frequently dissatisfied vrith the information they receive and
physicians tend to underestimate paiient desire for information.
Information-giving by providers to patients was significantly related to patient
satisfaction.
Time spent in discussing preventive care was positively associated with patient
satisfaction.
More time spent in the medical interview was positively associated with patient
satisfaction.
A more dominant and controlling physician communication style was
associated with less patient satisfaction.
66


Non-verbal behaviors such as frequent eye contact, appearing interested etc.
were positively associated with patient satisfaction.
Positive associations have been noted between patient-centered care behaviors
and patient satisfaction, compliance, resolution of patient concerns.
Compliance has been associated with more information-giving and positive
talk.
Patient-centered behaviors were related to higher self-reported compliance.
More patient controlling behaviors, less physician controlling behaviors and
more information-giving by physi patients were related to improved
and subjectively.
patient health status clinically, functionally,
In cancer patients, lack of information seems to be related to increased coping
problems, depression, anxiety, and uncertainty.
In her commentary on the effects of the physician-patient relationship on the quality of
health care, DiMatteo (1994) denoted similar findings:
Patient satisfaction with their physicians behavior has been positively
associated with the continuity of tile relationship.
More adequate histories have been| associated with higher patient satisfaction
and less delay in reporting symptoms.
Enhanced perceived personal control over health with collaboration and
TBDS.
Increased patient responsibility for jmedical decisions have been related to
lower likelihood of blaming the physician if things go wrong.
Increased adherence by patients when physicians provide more information.
Physician interpersonal characteristics and non-verbal communications have
been associated with patient health behaviors.
67


Patient outcomes can be effected by positive expectations about efficacy by
both the physician and the patient.
Improved recovery from surgery, decreased use of pain medications, shortened
hospital stays, improved clinical si:atus, improved symptoms, and enhanced
management of chronic conditions have all been associated with effective
physician-patient communication.
Roberts, Cox, Reintgen, Baile and Gibertini (1994) studied the influence of
physician communication on the psychological adjustment and decision-making of 100
post-surgical breast cancer patients. The Cancer Diagnostic Interview Scale (CDIS)
was used to assess the patients perceptions of their surgeons behavior during
the discussion in which they were informed of their cancer diagnosis. The SCL-90-R
Scale was used to assess psychological status.
Results from the CDIS assessment
revealed that the physicians caring attitude
was perceived as the most important factor related to psychological adjustment while
information giving was a much weaker factor. Those physician behaviors most closely
associated with a caring attitude were: I felt my doctor cared about me as a person,
my doctor understood my fears and concerns and my doctor is a warm and caring
person. The authors concluded that the use of psychotherapeutic techniques by
physicians can have a significant positive effect on a patients psychological well-
being. Based on this research and that from the social sciences literature, they
provided guidelines for what the surgeon should do in conducting the cancer
diagnostic interview: select an appropriate tune and place with the ideal being a
68


prearranged return office visit, verbalize empathy with direct eye contact, allow the
patient sufficient time to absorb the news and react emotionally, provide information
about the cancer and the treatment options, engage the patient in decision-making
about treatment.
In a very comprehensive study, Fox and Stein (1991) utilized random digit
dialed telephone interviews to investigate the effect of physician-patient
communication on mammography utilization in a random sample of 1,057 black,
Hispanic, and white women (mean age o:
53 years). Seven questions from the
interview were used to secure information about the quality of physician-patient
communication regarding mammography
Results revealed that 71% ofHisp anic women had never had a mammogram
and that more than half (55.3%) of the Hispanic women over age 50 years had never
had a mammogram. The most important predictor of whether women of all racial
groups had a mammogram, at any time or within the last year, was whether their
doctors had discussed mammography with them.
The data also suggested that within the physician-patient communication
process, physicians were not effectively communicating information about screening
and were discussing screening less frequently with Hispanic women than with white or
black women. The authors offered several possible explanations for this. First, fewer
than half of the Hispanic women had annu
d clinical breast exams that reduced the
69


opportunity for mammography referrals. Second, physicians were not using pediatric
visits for their children to offer family care. Third, Hispanic women tended not to
view the doctor as the most believable source of health information. Fourth, Hispanic
women cited more barriers to and anxieties about the use of mammography than did
black or white women. Finally, language differences may have created communication
difficulties. With the rapid population growth and purchasing power of the Hispanic
population in many concentrated areas ofjthe U.S. (Los Angeles, Houston, Phoenix,
Chicago, New York and others), the implications for provider organizations are
apparent.
In summary, the research reviewed here has provided evidence that patients,
and to some extent physicians, have been
patient communications and the physician
dissatisfactions have been borne primarily
frustrated with several aspects of doctor-
patient relationship. These frustrations and
out of traditional doctor-patient
relationships characterized by a Parsonian, physician-centered, paternalistic model
whereby the patient has played a predomii
antly passive, dependent role and the
physician has assumed an autonomous, professionally dominant and directive role in
the health and medical care decision-making process.
It is clear from the evidence presen
and non-physician researchers continue to
ted here that patients as well as physician
advocate a physician-patient relationship
and communication style characterized by informed collaboration, mutual
70


participation, and patient-centeredness. Most importantly, it is also clear from the
evidence presented here that such approaches can have profound effects on societal
outcomes of interest: enhanced patient satisfaction, improved clinical and functional
health status for patients, cost efficiencies, and possibly, enhanced job satisfaction for
physicians.
Given these implications, Ong et
al. (1995) noted that an important goal would
a theoretical framework would be useful
approach for improving physician-patien
be to establish a systematic theory and model of doctor-patient communication. Such
for: 1) providing a systematic, organized
communications and relationships; and 2)
generating hypotheses regarding such relationships. Thus, the purpose of the
following section is to review some models of doctor-patient communications and
relationships which have been championed by researchers and practitioners.
Collectively, these models may point us toward the theoretical framework and
practical models needed for advancing knowledge in this area; and subsequently, for
the successful reengineering of doctor-patient communications and relationships.
Models for Physician-Patient Communication and
Relationships |
In creating interpersonal relationsnips, there have been different opinions on
how to define and develop an optimal interpersonal relationship in the physician-
patient dyad. As early as 1956, Szasz and
Hollender described three types of
71


physician-patient relationships based largely on the degree of control required by each
participant for certain medical circumstances: activity-passivity, guidance-cooperation,
and mutual participation. According to Ong et al. (1995), some believe it is primarily
a social relationship; some claim that the relationship is determined by its therapeutic
qualities or client-centeredness; some describe the ideal as one which integrates
both patient-centered and physician-centered approaches; while others advocate a
eful models have been advocated for
itionship.
991) noted that the term patient-centered
Hunt, Joyce, Marinker, and Woodcock in
mutuality in the relationship. Several us
structuring the optimal doctor-patient rek
von Friederichs-Fitzwater et al. (1
medicine was first introduced by Balint,
1970. This was a contrasting model to the traditional illness-centered/disease-
centered/doctor-centered model. Patient-
follows:
centered medicine has been characterized as
In the patient-centered method, the physician attempts to
Ascertain the patients agenda... the essence of this
method is that the physiciaji tries to enter the patients
world, to see the illness through the patients eyes. This
is done by behavior that in\|ites and facilitates openness
by the patient. The doctorjs goal is to understand the
patients expectations, feelings, and fears.... bringing the
patients problems into the diagnosis, (von Friederichs-
Fitzwater et al.,1991;p,18)
von Friederichs-Fitzwater et al. (1991) pointed out that there has been no operational
definition of the model.
72


In 1977, Engel advocated a biopsychosocial model of disease and healing
which viewed disease as an open system process that incorporated mind, body, and
environment. This model was also in contrast to the traditional mind-body dualism
prevalent in the traditional paternalistic niomedical model. Keller and Carroll
(1994;p.l31) asserted that, While Engils biopsychosocial model of disease and
healing has received significant attention in primary care as an idea, it has received less
attention as a guide for physician behavior.
Emanuel and Dubler (1995) have characterized the ideal physician-patient
relationship as one that encompasses: choice, competence, communication,
compassion, continuity, and no conflict of interest. Their ideal tilted in favor of a
more patient-centered, collaborative, mutual participation approach: in terms of
choice, they have advocated for an informed patient who is provided choice among
treatment alternatives; in terms of communication, they have advocated for listening,
understanding preferences for information and involvement in decision-making,
assisting patients in dealing with the illness experience as well as the disease, and a
facilitative role for physicians in guiding sind directing patients through the illness
experience by recognizing values and feelings as well as life, family, and job
circumstances. From a legal perspective, :hey also pointed out that this type of
communication style can allay adverse events which may eventually lead to malpractice
litigation. In terms of compassion, they emphasized the importance of empathy; in
73


terms of continuity, they acknowledged the investment of time necessary for
prevention and chronic disease management; in terms of no conflicts of interest, it was
expected that the patients well-being take precedence over the physicians own
personal interests.
Keller and Carroll (1992) have id
influence the physician-patient relationshi
lentified three perspectives which they believe
iip: the physicians philosophy of disease and
health, the role which the physician assumes in the relationship, and the
communication strategies employed by the physician. They have advocated the E4"
model for physician-patient communications. This model goes beyond the traditional
biomedical tasks of find it and fix it to one which unites the communication tasks of
engage, empathize, educate and enlist wit
the model of Emanuel and Dubler (1995),
r the biomedical tasks of find and fix. Like
this model has favored a non-patemalistic
approach to physician-patient communications that emphasizes physician sensitivity to
patients preferences for information and involvement in decision-making.
Similar to Emanuel and Dubler (1995), Delbanco (1992) has advocated for a
more patient-centered view of doctor-patient interpersonal relationships. His
approach includes seven dimensions of care which should be considered when dealing
with each patient as a unique individual: respect for patients values, preferences and
expressed needs; communication and education; coordination and integration of care;
physical discomfort; emotional support ana alleviation of fears and anxieties;
74


involvement of family and friends; continuity and transition. Delbanco has asserted
that this approach should be: woven into the clinical encounter in bits and pieces until
practitioners become more comfortable with it; incorporated into the early years of
training in medical school; adopted by patients to guide encounters with physicians;
used to guide quality of care feedback from patients.
DiMatteo (1994) proposed a model based on collaborative TBDS. This model
was characterized as one which includes: mutual participation; explicit sharing of goals
and expected outcomes by both patient and physician; information exchange about
various treatment alternatives and their p otential effects; and negotiated treatment
decisions. To effect this model, she proceeded to identify the required physician and
patient behaviors: establishment of a con ducive atmosphere; allow patients to tell their
stories; physicians respond to patients illness experience with concern, empathy,
immediacy, and interest; physician encouragemen t of patients to ask questions;
physician exploration of patient goals, outcome expectations, values and quality of life
components; patient and doctor flexibility in considering various health management
strategies; physicians should assist patterns in organizing
information given to them; and patients should be allowed sufficient time to make
decisions.
Ballard-Reisch (1990) advanced a
making for physician-patient interactions.
theoretical model of participative decision
This model has three phases that guide the
75


doctor-patient interaction: the diagnostic phase; the exploration of treatment
alternatives phase; and the treatment decision, implementation, and evaluation phase.
In turn, each of these phases includes several stages.
The diagnostic phase includes information gathering and information
interpretation by both patient and physician. During this phase, the nature of the
relationship between the doctor and patient is negotiated. Within this negotiation,
there are four relational options: patient abdication, patient autonomy, collaboration,
or termination. Ballard-Reisch emphasized that collaboration is the ideal model for
participative decision making. The exploration of treatment alternatives phase
included the following stages: exploration of alternatives and criteria, establishment of
alternatives, and weighing of alternatives against criteria. The treatment decision,
implementation, and evaluation of implemented treatment (phase III) included the
stages of: alternative selection, decision implementation, and evaluation of
implemented treatment. Each of the three phases incorporates outside input from
other professionals, family members, and support groups.
Ballard and Reisch (1990) emphasized that the physician behaviors most
crucial to this model include: information sharing; gathering information about the
patients goals and expectations, values, beliefs, lifestyles, and sociocultural
relationships; active listening and providing feedback about understandings and
interpretations; involving family members and friends where appropriate; and
76


collaborative decision-making. The authors contended that viewing the doctor-patient
relationship as a three-phase process wnich progresses through eight distinct stages of
decision-making: 1) provides a structured framework upon which physician-patient
communications can be improved; and 2) tenders decision analysis researchers with a
framework for analyzing the quality of medical decisions.
Fredrikson (1995) has proposed an input-process-outcomes information-
exchange model of medical consultation. Central to the model and to the core of the
process component is information exchange which, like Ballard Reischs model
(1990), emphasizes a very flexible, parti Jipative flow and exchange of information
between doctor and patient in regards to
diagnostic and treatment information and
involvement in decision-making. Both mpdels have embodied input, process, and
outcome components.
The input component incorporates both patient and physician input. From the
patients perspective, this includes: frame of reference; motivations, goals, needs,
expectations, and personal information. Ijrom the physicians perspective, this
includes: motivations, goals, needs, expectations, and medical information. The
outcomes component is linked to the process component and encompasses specific
patient satisfaction outcomes related to factors of: patient understanding,
appropriateness of treatment, and physician relationship.
77


Kreps, OHair, and Clowers (1994) provided a conceptual model of
communications and how communications influences outcomes. They have referred to
this model as the Transformation Model
of Communications and Health Outcomes.
Theyve pointed out that this model is based on the systems theory transformation
model of input-process-output. In systems theory, inputs are the antecedents which
provide the raw materials for the processes necessary to accomplish the outputs. In
their model, antecedent conditions included health problem or risk and consumer
attitudes, beliefs, and expectations. The process component encompassed the
communication activities engaged in by tie patient and the physician:
provider/consumer message strategies, language used, non-verbal cues, channels, and
media. The outcomes of the communication and medical care activities comprised the
output component of the model: cognitive, behavioral, and physiological outcomes.
Finally, Ong et al. (1995) provided a theoretical framework which is very
analogous to the inputs-processes-outputs models described previously. Their model
includes three components: a) background variables, b) actual content of
Included in the background variables
ationship, types of patients and doctors,
dual content of communications component
communication, and c) patient outcomes,
component are: culture, doctor-patient rel
disease characteristics. Contained in the
were the actual communicative behaviors including both instrumental and affective
behaviors. Entailed in the patient outcomes component were short-term,
78


intermediate, and long-term outcomes in cluding: satisfaction, compliance, recall and
understanding of information, health status, and psychiatric morbidity.
Diffusion of Innovations: Constructs. Components.
and Determinants
Two of the primary aims of this study were to: 1) advance our understanding
about how physicians view enhanced participation of patients in medical decision
making through the integration of TBDS services, and 2) glean key insights on the
implications for working with physician organizations in optimizing the integration of
such services into medical care delivery systems. Since TBDS and the involvement of
patients in playing a more informed and active role in the medical decision-making
process can be considered a new innovation in the delivery of health care services,
the framework for studying the diffusion of innovations may assist in the fulfillment of
these aims. Therefore, the purposes of this section are to: review the basic com-
ponents and premises underlying the difiJsion of innovations, identify benchmark rates
for the adoption of new innovations and, where possible, delineate
factors/determinants influencing these rates.
Components of Diffusion
What is the diffusion of innovation and what are the major premises
underlying this concept? Diffusion has been defined as:
79


The process by which an innovation is communicated
through certain channels over time among the members
of a social system. The four main elements are the
innovation, communication channels, time, and the social
system. ... An innovation is an idea, practice, or object
that is perceived as new by an individual or an organization.
(Rogers, 1995, p. 10-11).
Innovations in health care include such t lings such as new drugs, treatment guidelines,
preventive services, procedures, diagnostics, therapeutics, and technologies. The rate
of adoption of an innovation over time hps been consistently demonstrated to follow
an S-shaped curve among various audiences including physicians, public health
administrators, teachers, and farmers (Rogers, 1995a).
This epidemiological S-shaped adoption curve characteristically has an early
slow phase involving a few innovators and early adopters. This is followed by an
accelerated middle phase whereby the rate of adoption usually takes off following
adoption of the idea by the early adopters. The rate of adoption in the final and third
phase is slow ending in incomplete penetration. In association with this adoption
curve, Ryan and Gross (1943) have described the different populations of adopters
within a given social system. They are characterized in five different categories.
The innovators generally comprise approximately 2-3% of the group. They
have been characterized as: well endowed with resources, venturesome, risk taking,
fascinated with novelty, willing to experiment and transcend geographical and social
boundaries to learn. Rogers (1995a) refers to them as cosmopolite. Locally, they


are generally somewhat disconnected socially, and are generally not the opinion
leaders. They play an important role by serving as the gatekeeper for the flow of new
ideas into the system.
The early adopters generally comprise around 12-15% of a group. They do
talk with innovators and selectively identify new ideas they would like to try out.
Locally, they are highly regarded for making prudent decisions in the adoption of new
ideas and they are socially well connected. They are regarded as significantly
influential opinion leaders with the early majority. For these reasons, they are often
utilized by change agents as the local evangelist for accelerating change.
The early majority comprise approximately one third of the group and are
characterized as very local in their perspectives, more focused on current needs and
local practical problems, and do not hold in high esteem lofty conceptual ideas,
science, or theory. Their innovation-decision making process has been characterized
as very deliberate and therefore slower than that of the innovators and early adopters.
They generally have a high state of readiness to hear about new innovations that are
relevant and practical for solving local problems from peers they know well and trust.
They look to the early adopters for indications of what is prudent to try in terms of
new innovations and serve as an important interpersonal network link between the
early adopters and late majority.
81


The next group, referred to as tne late majority, comprise approximately one
third of the group and have been described as skeptical about new innovations. They
distrust cosmopolite sources of knowledge,
for their learning channels, local proof, and
are more conservative in their views and
They look primarily to the early majority
cues in adopting new innovations. The late majority do not adopt new innovations
until they become the status quo and peep pressure eventually forces their adoption.
Finally, Ryan and Gross (1943) define the last 16% as the laggards. They
are the ultra-conservative traditionalists who are the last to change and to adopt new
innovations. They are isolated in their so
cial networks and are generally suspicious of
change, change agents, and innovations. The label given to this group may be an
injustice. Even though slow to adopt new innovations, this group is often useful and
valuable in making wise, systematic, calculated choices based on a rational view of
their own limited economic resources.
Adoption Rate Determinants
Rate of adoption has been defined as, The relative speed with which an
innovation is adopted by members of a social system and is measured by the number of
individuals who adopt a new idea in a specified period Rogers (1995a, p. 250). He
has identified at least five variables effecting the adoption rate of innovations.
82


First, the rate is influenced by how the potential adopter groups perceive the
attributes of the new idea. Rogers (1995a, pp. 250-251) emphasized that those
innovations that have the following attributes are more likely to see rapid adoption
when compared to those innovations that do not have these attributes.
1) Relative advantage the degree to which a new idea is perceived as better than
the idea that it supersedes.
2) Compatibility the degree to which an innovation is perceived a consistent with
the existing values, past experiences, and needs of potential adopters.
3) Complexity the degree to which a pew idea is perceived as relatively difficult to
understand and to use.
4) Trialability the degree to which an innovation can be experimented with on a
limited basis.
5) Observability the degree to which the results of an innovation are visible to
others.
Rogers (1995a) pointed out that 50-90% of the variance in adoption rates can be
explained by these five attributes. As we will see later, these attributes have indeed
been demonstrated to be influencing factors on the diffusion of innovations in health
and medical care.
83


A second variable influencing the rate of adoption is the decision-making process
itself. In the process of adopting a new innovation, individuals go through a decision-
making process that includes the following stages (Rogers, 1995a): 1) knowledge -
becoming aware that a new idea exists; 2) persuasion developing a favorable attitude
about the new idea; 3) decision detemiining whether to adopt the innovation; 4)
implementation putting the new idea into use; and 5) confirmation evaluating the
innovation for the purpose of continued
adoption, modification, or discontinuance.
Interestingly, this typology is conceptually congruent with the transtheoretical model
of behavior change that includes pre-con
emplation, contemplation, preparation,
action, and maintenance stages (Prochaska and DiClemente, 1983).
According to Rogers (1995a), three additional variables influencing adoption
rates include: 1) the nature of the communication channels within a social network,
2) the nature of the social system through which the innovation is diffusing, and 3) the
degree to which change agents are invol jed in promotional activities. He speculates
that when considering all five of these variables, the diffusion process is driven
predominantly by how the attributes are perceived and how this perception is
conveyed through the adopter groups interpersonal networks.
In addition to Rogers (1995a) classic social system theory of diffusion, Green
(1980) has provided another typology that holds significant potential for assessing
factors which influence the diffusion of innovations. It also provides a framework for
84


targeting interventions for enhancing the diffusion of a given innovation. In his
PRECEDE model, Green has classified these factors into three categories:
predisposing factors (physician knowleJge, attitudes, beliefs,), enabling factors
(skills/competence, resources, practice characteristics), and reinforcing factors
(reimbursement, peer support, positive feedback, enhanced self-esteem). The
application of this typology will be explored in a later section examining the diffusion
of preventive services.
Diffusion of Innovation in Health Care
The gap between medical science and medical practice (Greer, 1994) as well as
the gap between best achievable practices and actual practice (Berwick, 1998) is well-
established. Many medical innovations (new drugs, treatment guidelines, preventive
services, procedures, diagnostics, therapeutics, and technologies) may take years or
even decades to reach widespread adoption. Ford, Kaluzny, and Sondik (1990) have
poignantly noted that the time involved to achieve the full transfer of new treatments
continues to be problematic in medicine.
Even though an enormous body of scholarly research exists on the diffusion of
innovations, it is still not a well-understooL process. A practical standardized process
for success has not yet been codified. Therefore, it is not surprising to find wide
variation in the rates of adoption for new ideas in health and medical care services. The
85


purpose of this section is twofold. The arst is to review, where available, rates of
adoption related to new drugs, practice guidelines, and preventive services. The
second is to distill those determinants/factors that have been demonstrated to
predominantly influence diffusion and rates of adoption. Well begin by first
examining one of the earliest and most c
medical innovations.
assic sociometric studies on the diffusion of
Using pharmacy prescription records with information about the month in
which each doctor adopted the new drug. Coleman, Menzel, and Katz, (1966) studied
the spread of an antibiotic (tetracycline) among physicians in four Illinois communities.
They also collected data from physician respondents regarding other physicians in their
community with whom they had discussed the new drug. This information allowed the
researchers to trace the social networks through which the innovation was diffused.
The commercial sponsor of the study, Pfizer, provided academic detailers to
extensively promote the new drug.
This innovation spread rapidly reaching almost complete adoption in 17
months. Physicians reported that the academic detailers had little direct influence on
their decision to adopt. Instead, the researchers discovered that information about the
new drug spread primarily by word of mouth from physician to physician and,
generally, a physician did not adopt tetracycline until a respected peer had shared
his/her experiences with the new innovation. Greer (1994, p. 434) insightfully noted
86


that what is usually forgotten in this study is that the best predictor of general adoption
was not adoption by the physicians who were recognized as opinion leaders, but
adoption by those physicians who communicated with the most other physicians the
most often. She emphasized that, the more people talk and stimulate further talk, the
more there is opportunity for consensus, practical problem solving, and social pressure
to change.
In another prescription drug study, Peay and Peay (1988) examined the
adoption process for temazepam through interviews with 124 specialists and general
practitioners. They found that within 13 months after its release, 71% of the
physicians were familiar with the drug (average time between release and awareness
was about 17 weeks), 48% had prescribed it (average time between its introduction
and first prescription was 5.5 months), ana 27% preferred it to the alternatives.
In contrast to the Coleman, Menzel, and Katz (1966) study, they reported that
contact with academic detailers was the nlost consistent predictor of favorable
reception by physicians. Sixty-eight percent reported that their first news came from
commercial sources, 40% citing the detailers in particular. Only 28% reported
discussing this new drug with other doctors before deciding to prescribe. Eighteen
percent reported colleagues as the most important information sources in their first
decision to prescribe. Doctors who had more frequent contact with detailers (at least
once per week) were aware of the drug earlier, prescribed it earlier, and were more
87


Full Text

PAGE 1

ADAPTIVE EQUALIZATION byNamHo B.S., University of Colorado, 1986 A thesis submitted to the Faculty of the Graduate School of the University of Colorado in partial fulfillment of the requirements for the degree of the Master of Science Department of Electrical Engineering 1989

PAGE 2

To the_ GRADUATE SCHOOL OF THE UNIVERSITY OF COLORADO: Student's name: ______ Degree : ..... M==s ....... Discipline : Electrical Engineering Thesis Title : _____ ___ The final copy of this thesis has been examined by the undersigned, and .we fmd that both the content and the form meet acceptable presentation standards of scholarly work in the above discipline. Chairperson dol/ I ftfZ Date

PAGE 3

This thesis for the Master of Science degree by NamDaiHo has been approved for the Department of Electrical Engineering by Douglas A. Ross '1 i'lliam J. Wolfe

PAGE 4

Ho, Nam D. (M.S., Electrical Engineering). Adaptive Equalization. Thesis directe4 by Associate Professor Douglas A. Ross. In adaptive signal processing, one usually encounters identification or filtering problems. Consider the performance function J: J = 1: [e(i)]2 = 1: [d(i)-y(i)]2 where e(i) is the difference between the output d(i) of the system (or the plant) and the output y(i) of the model. If the model is a MA model then J has one and only one minimum point and its shape looks like a parabola in one dimension and a bowl in N dimensions (N >2). The objective of an identification problem (to find the parameter values of the model so that they are as close to those of the system as possible) will be achieved simultaneously with the objective of a filtering problem (to minimize J). If the shape of J is flat at the mtmmum point then Exhaustive Search technique (ES) is probably the best way to cope with this situation. ES is easy to understand, it offers very fast

PAGE 5

v convergence and gives very accurate estimates. This thesis will discuss several techniques to solve identification and filtering problems and ES will be investigated in detail. The form and content of this abstract are approved. I recommend its publication. Signed Douglas A. Ross

PAGE 6

ACKNOWLEDGEMENTS P would like to express my smcere appreciation to Dr. Douglas A. Ross for his support and encouragement during the preparation of this thesis. Special thanks are gtven to the student's parents, brothers, sisters and friends for their support.

PAGE 7

TABLE OF CONTENTS CHAPTER I. IN'TRODUCTION ................................................... 1 II. GENERALIZED INVERSE ...................................... 4 2-1 AR, MA and ARMA models . . . . . . . . . . 4 2-2 Formulation of Normal equation ................. 7 2-3 Solving Normal equation ........................... 10 2-3.1 Direct Inversion ...................... 10 2-3.2 Gauss elimination ...................... 11 2-3.3 Iterative technique .................... 11 2-3.4 Generalized Inverse .................. 12 2-4 Computer simulation ................... ............ 15 2-5 Conclusion .................................. .... ... .... 17 ill. RECURSIVE LEAST SQUARES ALGORITHM.... 18 3-1 Introduction ........................................... 18 3-2 Formulation of the problem ...................... 18 3-3 Recursive Least Squares algorithm ............ 20 3-4 Some properties of RLS ........................... 21 3-5 Computer simulation .. .... ... .... .... .... ..... ... 23 3-6 Conclusion ........................... .................... 31 IV. EXHAUSTIVE SEARCH TECHNIQUE (ES) .......... 32 4-1 Introduction .... .................... .... ...... .... ..... 32 4-2 Exhaustive Search technique ................... 34 4-3 Computation ............................................ 44

PAGE 8

Vll a) Exhau_stive Search ...... ..... ... ..... ... ........ .. 44 b) Recursive Least Squares ... .. . .. .. . .. . . 45 4-4 Sample size .... ... .. ..... .. .. . .. . .. . .. ... .. ... . 48 4-5 Step size .. ...................................... .......... ... 49 4-6 Input and output ... ...... ... .... ... ... ... .... ... ... . . 50 4-7 Noisy MA process........................................ 51 4-8 Over/Underestimate the order of the system.... 58 a) Overestimate the order of the system ...... 62 b) Underestimate the order of the system .... 62 4-9 Exhaustive Search for infinite impulse response (IIR) filter ................................... 64 4-10 Conclusion . .. .. . .. . . . . .. . . . . . . . . . .. . .. . 66 V. DISCUSSION AND CONCLUSION....................... 68 BffiLIOGRAPHY ....................................................... 70 APPENDIX A. THE PARAMETER VALUES OF THE WEIGHT VECTOR OF THE SYSTEM, THE ESTIMATES OF THE WEIGHT VECTOR OF THE MODEL, SUM OF RESIDUALS R, R.M.S OF R ... OF THE GRAPHS IN CHAPTERS II, III, IV ... .. .. . .. .. . .. .. . .. ... .. .. 72 B. PROGRAM LISTINGS ....................................... 85

PAGE 9

v 111 TABLE OF SYl\ffiOLS RLS = recursive least squares ES -exhaustive search D -number of data points I -number of iterations L = number of parameters SIN = signal to noise ratio = step size R -sum of residuals

PAGE 10

TABLE OF FIGURES FIGURE 2-1.1 2-2.1 3-5.1 Stochasic model .. .. .. .. .. .. .. .. .... .. .. .. .. .. .. .. ... .. .. .. .. . .. .. 5 Schematic diagram for identification or filtering problems 7 RLS 0=150 1=150 L=6 S/N=infinite parameter values vs number of iterations .. ............ 25 3-5.2 RLS 0=1000 1=1000 L=6 S/N=infinite R & log10R vs number of iterations ......... ..... ..... .. 26 3-5.3 RLS 0=1000 1=1000 L=6 S/N=10 parameter values vs number of iterations ..... ... .. .. .. 27 3-5.4 RLS 0=1000 1=1000 L=6 parameter values vs number of iterations ... .. .. .. .. .. 28 3-5.5 RLS 0=1000 1=1000 L=6 S/N=10S parameter values vs number of iterations .......... .. .. 29 3-5.6 RLS 0=1000 1=1000 L=6 S/N=IO, IQ3, 105 R & log10R vs number of iterations ..................... 4-1.1 The "parabola" shape of the performance function J in one d . un.enston .................................................................... 4-2 Schematic diagram of exhaustive search ......................... 4-2.1 ES 0 =50 1=10 L=6 S/N=infinite 30 33 35 dS = .1 parameter values vs number of iterations 40 4-2.2 ES 0 =50 1=10 L=6 S/N=infinite dS = .I R & log10R vs number of iterations....... 41

PAGE 11

X 4-2.3 ES D =50 1=10 L=11 S/N=infinite = .1 parameter values vs number of iterations 42 4-2.4 ES D =50 1=10 L=11 S/N=infinite = .1 R & log10R vs number of iterations ........ 43 4-7.1 ES D =50 1=10 L=6 S/N=10 = .1 parameter values vs number of iterations 54 4-7.2 ES D =50 1=10 L=6 S/N=103 = .1 parameter values vs number of iterations 55 4-7.3 ES D =50 1=10 L=6 SIN=10S = .1 parameter values vs number of iterations 56 4-7.4 ES D=50 1=10 L=6 S/N=10, 103, 105 R & log10R vs number of iterations ...................... 57

PAGE 12

CHAPTER I INTRODUCTION The field of recursive identification is often viewed as a long and confusing list of methods and tricks. Methods and algorithms have been developed in different areas with different applications. The term "recursive identification" is taken from the control literature. In statistical literature the field is usually called "sequential parameter estimation," and in signal processing the methods are known as "adaptive algorithms." Systems consist of a wide range of objects whose behavior we are interested in studying, affecting or controlling. The typical tasks are related to the study and use of the systems: control, signal processing (filter designing) and prediction. Many techniques have been developed in control theory, communications, signal processing and statistics for solving the problems involving these tasks. In order to use these techniques, some knowledge of the properties of a system i.e., model, must be known. The model of a system can be represented in one of several different forms. For examples, graphics models ( properties of the systems summarized in graphics or in tables), mathematical models ( mathematical relationships between certain variables of a system) are very important and necessary when complex design problems are involved.

PAGE 13

2 In this thesis, mathematical models: moving average (MA) and autoregressive moving average ( ARMA ) will be used. Basically, there are two approaches to building a mathematical model of a given system: the first way is to look at physical laws and relationships between variables inside the system; from there the model is constructed. Unfortunately, this is not always possible: it is time-consuming and may lead to an unnecessarily complex model. In addition, the system's properties may change in an unpredictable manner. The second way is to use the signal produced by the system; it consists of two general steps: choose a particular member of a family of candidate models, then use an algorithm to estimate the parameters of this candidate. The fundamental theorem in the decomposition of stationary time series proposed by Wold [1] says: The most general decomposition of a stochastic process is a moving average one (MA). An MA process can also be modeled by AR or ARMA; a finite order MA process requires an infinite order AR representation. AR and MA are special cases of ARMA. The correct choice of AR, MA, ARMA and their orders is very important. Mismodeling may lead to serious errors. Akaike information criterion (AIC) and Final prediction error (FPE) are among the choices for estimating the orders. Estimating (identifying) the parameters is usually done in two ways: a batch of data is collected first and used to identify the parameters; this is called off-line or batch processing identification. On-line or real time identification means updating the estimates of parameters as new data are available.

PAGE 14

3 The following 1s the synopsis of the thesis's content and order: chapter II Generalized Inverse This is an off-line technique. The identification problem which leads to normal equation is formulated. Several methods such as direct inversion, Gauss-elimination, and iterative technique are briefly discussed and finally generalized inverse is investigated. chapter ill Recursive Least Squares (RLS) Here is an on-line algorithm used for MA orily. RLS algorithm has been investigated extensively in technical literature and is mentioned here for the purpose of pointing out its pros and cons and showing, later on, how exhaustive search solves some of RLS's problems. chapter IV Exhaustive Search (ES) One of the problems associated with RLS occurs when the autocorrelation matrix R in normal equation RW=P is ill-conditioned. This is why ES came into being. ES works quite well for finite impulse response (FIR) filters but not so for infinite impulse response (ITR) filters. Number of computations, rate of convergence, etc., of ES are investigated in detail in this chapter. chapter V Discussion and Conclusion

PAGE 15

CHAPTER IT GENERALIZED INVERSE 2-1 AR. MA and ARMA models Many discrete-time random processes encountered in practice are well approximated by a rational transfer function model. The idea of representation of a discrete-time random process by a model is that a random process consisting of highly correlated observations may be generated by applying a series of statistically independent shocks," white noise, for instance, to a model (discrete-time linear filter. figure 2-1.1 on next page). H an input driving sequence x(i) and. the output sequence y(i) that is to rriodel the discrete-time random process are related by the linear difference equation: y(i) Where q L b(k)x(i-k) k=O p L a(k)y(i-k) k=l a(O), a(l), a(2), ... a(p ), b(O), b(l), ... b(q) are constants, this general model is termed an autoregressive moving average (ARMA) model. The transfer function H(z) between input x(i) and output y(i) is a rational function: H(z) B(z)/A(z)

PAGE 16

White noise Input sequence x(i) ... MODEL Discrete-time linear filter with transfer function H(z) figure 2-1.1 : Stochastic model ... Discrete-time random process Output sequence y(i) 5

PAGE 17

6 p q A(z) = L a(k)z-k B(z) = L b(k)z-k k=O k=O It is assumed that A(z) has all its zeros within the unit circle of the z plane so that the filter is stable. If all the a(k) coefficients except a(O) = 1 vanish for ARMA parameters then: y(i) = q L b(k)x(i-k) k=() the discrete-time random process y(i) is a moving average of .order q; it is denoted as MA(q) process. The model is called MA or all-zero model. Likewise, if all the b(k) coefficients except b(O) are zero in ARMA model then: p y(i) L a(k)y(i-k) + b(O)x(i) k=l The output sequence y(i) is now an autoregressive process of order p; it is denoted as AR(p) process and the model is an AR or all -pole model. Note that MA model is often known as finite impulse response (FIR) filter while AR, ARMA models are usually known as infinite impulse response (llR) filters [2], [3].

PAGE 18

2-2 of Normal eguation Consider the following figure: x(i) SYSTEM d(i) e(i) = d(i)-y(i) MODEL y(i) figure 2-2.1 : Schematic diagram for identification or filtering problems. Model : one can use AR, MA, ARMA, etc. In this case MA model (or FIR filter) is used. x(i) input d(i) : desired output or output of the system. y(i) : filtered output or output of the model. Assume the system itself is a FIR filter. The impulse response vectors of the system and the model are: 7

PAGE 19

W.* L respectively. 8 In identification problems, the objective is to obtain W L so that .W L is as close to W L as possible. In filtering problems, minimizing the sum of squared errors is the goal. The objective now is to find the "best" W L such that the sum of squared difference between d(i) and y(i): k k J L [e(i)]2 1: [d(i)-y(i)l2 i=L-1 i=L-1 is as small as possible. J is called the performance function (criterion function). For a given sequence of input vectors { X(i)} and scalars { d(i)}, J is a function of W L only and therefore is a measure of how well W L performs to produce y(i) which matches d(i). The choice of W L minimizing J is the value that has the best performance. Let the input vector be: XL(i) -[ x(i) x(i-1) ... x(i-L+1) ]T and the given data be: x(O), x(1), ... x(k), d(L-1), d(L), ... d(k), then

PAGE 20

J = J note: k k L [e(i)]2 = l: [d(i)-y(i)] 2 i=L-1 i=L-1 k k L [d(i)]2 -WLT L XL(i)d(i) i=L-1 i=L-1 k + WLT[l:XL(i) XLT(i) ]WL i=L-1 y(i) = 9 k [L d(i)XL T(i)]W L i=L-1 Define the autocorrelation matrix R and the correlation matrix P as following: k R = p = L d(i)XL(i) i=L-1 i=L-1 expand R and P out and write in more explicit form: k k :E x(i)x(i) :E x(i)x(i-1) i=L-1 i=L-1 k-1 R :E x(i)x(i) i=L-2 k :E x(i)x(i-L+1) i=L-1 k-1 :E x(i)x(i-L+2) i=L-2 k-L+l :E x(i)x(i) i=O

PAGE 21

10 k :E d(i)x(i) i=Ll k p = :E d(i)x(i-1) i=Ll k J :I: [d(i)J2 -wTp -pTw + wTRw i=Ll k J L [d(i)]2 2WTP + wTRW Where w = WL i=Ll To minimize J, take the derivative of J with respect toW and set it equal to zero: dJ/dW = -2P + 2RW = 0 RW = P (22.1) is called the normal equation. 2-3 Solving the normal equation 2-3.1 Direct inversion (2-2.1) The most straightforward approach to solving 2-2.1 IS the.

PAGE 22

11 direct inversion of R followed by its multiplication with P. The inverse of R is defmed as: R-1 -adjoint R/det R While this formulation is fine for theoretical work, it is hardly adequate for numerical work. As the order of R increases, the number of computations of R-1 gets out of hand; the limitations on word length of the machine used and the numerical accuracy of computational methods also introduce round-off errors which become significant if the number of computations is large. 2-3.2 Gauss-elimination Gauss-elimination (or sucessive elimination) consists of reducing the system of L equations in L unknowns (for example ) to a system of (L -1) equations in (L-1) unknowns by using one of the equations to eliminate one of the unknowns. This process repeats until only one equation and one unknown are obtained. Once this last unknown is found, the remaining (L-1) unknowns are calculated by back substitution. If matrix R does not have full rank, R -1 does not exist, direct inversion fails and so does Gauss-elimination. 2-3.3 Iterative approximation Given autocorrelation matrix R and correlation matrix P, the estimate of WL at (k+1) th iteration may be computed according to this scheme: WL(k+1) WL(k+1) = [ w 0(k+l) W1 (k+l) ... wL_1 (k+l) JT = [ 1-J.!R]. w L(k) + JlP w L(O) = 0 Jl : small positive number

PAGE 23

I: identity matrix It has been shown in [ 4] that if J.l is chosen so that: lim [ l-J.LR]k-----? 0 k----> infinity then lim WL(k) ->WL* k----> infinity 12 This scheme is not the only way. Others are steepest-descent, Newton methods and their modifications. 2-3.4 Generalized Inverse Consider again equation (2-2.1): RW = P R is a linear transformation with domain and range spaces in matrix theory. By introducing arbitrary orthornormal bases for both domain and range, a simple representation (i.e., diagonal matrix) can be found for R. Let: v u = = ( Yt V2 ... VL] ( llt U2 ... UL] be new orthornormal sets of vectors used as bases for both domain and range respectively. The new V: W' . w In range space the new P' of P : coordinates W' of W with respect to VW' P' = UHP :. P = UP' P'

PAGE 24

In the new system, R becomes R' with: R' UHRV :o R UR'VH where R' = :] diag[cs1 CJ2 . cs") k by k matrix -I yHy -I = V(R'HR')VH 0 RHRv. = 0 0 1 -U(R'R'H) UH 0 RRHu. 0 0 I 13 cr.2v. 1 1 cri2 : eigenvalues of RHR and RRHo vi and ui are eigenvectors of RHR andRRH The positive square root of cr? i = 1, 2, 3 ... k, are singular values ofR. U and V are not unique because they consist of ui vi not uniquely determined. To solve RW =Pin the least squares sense means to find W so that IIRW-PII2 is minimum. II RW-PII 2 = II UR'VHW-P 112 = II UR'VHW-UUHP 112 Let Y = yHw and P' = uHp The problem now is to minimize IIR'Y P'll -2

PAGE 25

P' = [pl' P2' ... PL']T II R'Y-P'II2 = [( crlyl-pl') 2+ ( cr2y2-p2')2 + ( O'iYCPi')2 ]112 i = 1, 2 ... L Consider 2 cases: Case I If R is nonsingular then cri is not equal to zerofor i = 1, 2, 3 ... L. By letting cr.y.-p. = 0 :. y. = p.'/ cr. I I I I I I and W = VY Case II 14 IfR is singular, at least one of cri must be zero. For instance cri = o for i = k+1, k+2 ... L, then II R'Y-P'II2 = [( crlyl-p1') 2+ ( cr2y2-p2')2 ... + ( O'iYCPi') + P + p ] 1/2 k+1 ... L let Yi = pi'/ O'i for i = 1, 2, ... k and yi = 0 for i > k :. W = VY Whether R is singular or not, the solution W always exists. If one lets R = diag [ cr 1 1 cr2 1 .. cr x: -I] 1 < k < L W = VY = VR*P' = VR*UHP = R+P R+ = VR *UH is called the generalized inverse of R.

PAGE 26

15 There are several algorithms to calculate R +, and they fall 1nto two broad groups. The first group consists of full rank factorization and singular value decomposition (SVD); the iterative techniqQes are. in the second group. One of the commonly discussed iterative techniques in the second -group is due to Ben-Israel [5], this and its variants are not competitive for large matrices with SVD. The reader interested in the first group is referred to [ 6] for full rank factorization and to [7] for details on SVD. This group could be divided into two smaller ones. Full rank factorization and other methods based on some variations of Gauss-elimination, call these the "elimination methods" group. The second one consists of SVD only. If one suspects R is severly ill-conditioned and /or very high accuracy is needed, the best idea of all is to compute in exact arithmetic. This will provide R exactly if R has rational entries. Elimination methods are preferable with exact methods. 2-4 Computer simulation The program used in this chapter is written 1n BASIC language. A major part of the program: the subroutine Complex singular value decomposition was developed by P .A Businger and G.H. Golub (Communications of ACM Vol. 12, pp. 564-565, October 1696) in FORTRAN and was translated into BASIC. The input x(i) is generated as following: x(i) = 2*(RND-.5)*SQR(3*.1) where is multiplication, SQR() = ( )1 2 and RND is a function which generates random numbers between 0 and 1.

PAGE 27

16 x(i) is white noise with mean= 0 and variance= .1. The desired output d(i) is generated by applying x(i) to a MA model. The relationship between d(i) and x(i) is : d(i) = g(O)x(i) + g(1)x(i-1) + ... + g(p)x(i-p). Where g(O) = .5, g(1) = 1, g(2) = -.8 ... See the program Singular Value Decomposition in appendix B for values of g(i), i = 0, 1, 2, ... 20. These are the actual values of the weight vector W L of the system. The results in appendix A are the actual values of g(O), g(1), ... g(p) and their estimates for the cases : p = 6, 11, 20. Given input sequence x(i) and output sequence d(i), the program computes the autocorrelation matrix R and the correlation matrix P in the normal equation R W = P, then the generalized inverse of R (R+) is calculated, and finally the solution W IS: W=R+p Several observations should be pointed out: The estimates are fairly accurate even when the number of data points used is small. As the number of parameters (p) increases, so do the number of data to ensure accuracy. The values of singular values a i of R are machine dependent. Numeric values can be either integer, single-precision or double-precision. Single-precision numeric values are stored with 7 digits (although only 6 may be accurate) and double-precision numeric values are stored with 17 digits of precision and printed in as many as 16 digits.

PAGE 28

17 Therefore, if one uses single-precision and one of the singular values of matrix R happens to be less than 1 o8 then concluding R is singular is wrong. This could happen if double-precision is used and one of the singular values is less than 1 o-17 Assume that noise e(i) = 2*(RND-.5)*SQR(3*D) is added to the desired output d(i). The signal to noise ratio is defined as: Signal/noise = (variance of signal)/(variance of noise) where variance of signal d(i) = .29 (computed by using 100,000 data points ) and variance of noise e(i) is D By varying D, different signal to noise ratios will be obtained. The estimates are quite good when signal to noise ratio is high as expected. The data when signal/noise = 10, 103 105 are in appendix A. 2-5 Conclusion This chapter swnmarizes several various techniques to solve least squares problems (RW=P), or equivalently to find the best impulse response vector W of FIR filter so that sum of squared errors are minimized. Direct inversion, Gauss-elimination, iterative techniques and generalized inverse are briefly discussed. When matrix R in the normal equation RW=P is singular, only iterative and generalized inverse methods are applicable. There are several algorithms to compute generalized inverse of a matrix. SVD is one of them and is widely used. "elimination methods" are recommended when a matrix is ill-conditioned and /or high accuracy is needed.

PAGE 29

CHAPTER ill RECURSIVE LEAST SQUARES (RLS) 3-1 Introduction There are situations in which as new data are available, the impulse response vector W = W L needs to be updated accordingly (adaptive filters), but this becomes frustrated and time-consuming if one uses batch processing methods discussed in chapter II since R -I (direct inversion) or R+ (generalized inverse) ... has to be recalculated. This is one of the motivations for introducing recursive least squares algorithm (RLS). 3-2 Formulation of the problem The formulation of the problem for RLS is proceeded exactly in the same way in chapter II. Recall equation (2-2.1) RW = P redefined: k-1 R = L X(i)XT(i) = RK i=L-1 k-1 p = 1: X(i)d(i) i=L-1 Assume new data x(k) and d(k) now are available. How should one update W recursively ?

PAGE 30

19 k RK+1 = L X(i)XT(i) = RK+ X(k)XT(k) i=L-1 k PK+1 l: X(i)d(i) -P K + X(k)d(k) i=L-1 WK+1 -RK+1-1 PK+1 Employing the well known matrix inversion lemma, sometimes called "ABCD lemma": ( A + B C D t1 = A-1 -A-1 B [ D A -1 B + c-1 r1 D A-1 This identity is proved by multiplying the left with the right side and getting identity matrix I Applying "ABCD lemma" for RK+1 with the following associations: Then R -1 K+1 A = RK B = X(k) = R -1 K C -1 D = XT (k)

PAGE 31

20 Note R -lp. K K y(k) Let qk (scalar) e(k) d(k)-y(k) v After some algebraic manipulation and simplification: 3-3 Recursive least sguares algorithm CRLS): RLS consists of the following steps [8] : 1) Accept new data x(k), d(k). 2) Form X(k) by shifting x(k) into the information vector. 3) Compute the prior output (estimated output) y(k) = XT(k) WK 4) Compute the error e(k) e(k) = d(k)-y(k)

PAGE 32

21 5) Compute the filtered information vector ZK 6) Compute. the scalar qk 7) Compute the gain constant v 8) Compute the normalized filtered information vector ZK = 9) Update the weight vector 10) Update the inverse of autocorrelation matrix R -1 k+l 3-4 Some properties of RLS = 1) To use RLS, first of all, the autocorrelation matrix Rand the weight vector W must be initialized. For the weight vector, one could choose: W(O) = 0 For the autocorrelation matrix Rk' the initial value R0 can be computed in this way:

PAGE 33

k RK+1 L X(i)XT(i) i=L-1 k RK+1 = L X(i)XT(i) i=L-2 Ro = oi with I = identity matrix 0 -small positive constant R -1 s::-11 0 -u 22 + oi 2) The introduction of "forgetting factor" p into RLS ( 0 < p < 1) will put more weight on the recent data and less weight on the older ones. The motivation of using p comes from the cases where the data { X,d} changes its charateristic within record of data points (nonstationary process). The RLS in this case will be [9] : R -1 K+1 pk+1 = -] p [ (1/ p) d(k)X(k) + Pk]

PAGE 34

23 d(k)-y(k) = vvk + ----------zk p = 0 then every data is weighted equally. 3) From [10] one of the statistical properties of RLS is: b(k) 0 -R-1vv* k k number of iterations R-1 : inverse of R VV *, VV : weight vectors of the system and the model respectively. b(k) = E[VV(k)] VV *=the bias b(k).becomes smaller ask increases. \Vhen k approaches infinity then b(k) approaches zero. o should be small to reduce the bias. 3-5 Computer simulation From figure 2-2.1 page 7, let: w [wo"' WL-1*] = w1 L -the impulse response vector of the system VVL = [wo wl WL-1] = the impulse response vector of the model L -length of both VV L and VV L

PAGE 35

D R = number of data points or number of iterations sum of residuals The desired output d(i) is generated by applying the input x(i) = 2*(RND-.5)*SQR(3*.1) to the system The relationship between d(i) and x(i) is: d(i) = w0*x(i) + w1 x(i-1) + ... + wL_1 x(i-L+1) where w0 = g(O) = .5, w1 = g(1) = 1, w2 = g(2) = -.8 ... 24 See the program "recursive least squares" in appendix B for the values of g(i), i = 0, 1, ... 5 RLS is used in this chapter to find W L so that W L gets as close to WL* as possible. Figure 3-5.1 (page 25) shows the values of the estimates (w0 w1 wL_1 of the vector W) and the values of the parameters ( w0*, w 1 ... wL_1 of the vector W L ). The dotted lines are the estimates and the solid lines are the parameters. This is the case where L = 6 and D = 150. Figure 3-5.2 (page 26) is the plot of sum of residuals R vs number of iterations D ( D = 1000). Figures 3-5.3, 3-5.4, 3-5.5 (pages 27, 28, 29) show the estimates and the parameters when noise is added to the desired output.d(i) and the signal to noise ratio SIN= 10, 103 10s. L = 6, D = 1000.

PAGE 36

2 1 -1 -2 parameter values I \ I t r,. I L..J . I t LJ figure 3-5.1 : # of data points # of iterations 25 61 ,. 121 151 number of iterations recursive least squares = 150 # of parameters = 6 = 150 : signaJJnoise = infmite

PAGE 37

19I1 19I-6 19E11 19I-16 19I-21 R sum of residuals R -6 -11 -16 1 -21 1 LOC(R) 259 figure 3-5.2 : # of data points # of iterations 26 599 759 number of iterations recursive least squares = 1000 # of parameters = 6 = 1000 : signaVnoise = infinite R = sum of residuals

PAGE 38

27 parameter values -.ss -1.1 81 ,. 1111 r""' number of iterations .J figure 3-5.3 : # of data points # of iterations recursive least squares = 1000 : # of parameters = 6 = 1000 : signaJJnoise = 10

PAGE 39

28 parameter values 1 1 I-. .ss -.ss -1.1 1\ ___ ...__.,_._, __________________ _,_ ---I r figure 3-5.4 : # of data points # of iterations -........ 699 899 1999 number of iterations -recursive least squares = 1000 : # of parameters = 6 = 1000 : signaVnoise = 163

PAGE 40

29 parameter values 1.1 \ ----------------------.55 -.ss -1.1 1\_,_ ........ ______________ ,.,_ _____ 811 Hila number of iterations ,, _____________________________ ___ figure 3-5.5 : # of data points # of iterations recursive least squares = 1000 : # of parameters = 6 = 1000 : signaVnoise = 1 OS

PAGE 41

lQE1 liE3 liE5 19E7 19E-9 R -1 -3 -5 -7 -9 LOC 30 sum of residuals R signal/noise = 10 signal/noise = 103 signal/noise = 1 OS 1 258 751 1988 number of iterations figure 3-5.6 : recursive least squares # of data points = 1000 # of parameters = 6 # of iterations = 1000 signal/noise = 10 103 105 : R = sum of residuals

PAGE 42

31 Figure 3-5.6 (page 30) is the plot of R vs D ( D = 1000) for SIN= 10 103 10s. The data for these graphs are in A. 3-6 Conclusion RLS algorithm works quite well when no nmse is involved, gives fairly good estimates but needs many iterations. "Fast" RLS is one of the modifications of RLS to reduce the amount of computation.

PAGE 43

CHAPTERN 4-1 Introduction J Recall the performance function J in chapter II and ill: k L [e(i)]2 i=L-1 k = l: [d(i)-y(i)]2 i=L-1 J is the function of weight vector W only, its shape looks similar to a "parabola" in one dimensional problem (W = [ w 0]) and a "bowl" in L dimensions (W = [ w0 w1 ... wL_1 ]). Figure 4-1.1 (next page) is the "parabola" shape of J in one dimension, the minimum point of J corresponds to the best impulse response vector W = W optimal = W Recall the normal equation: RW = P In one dimension, the condition of the matrix R determines the shape of the "parabola" at the minimum point If R has full rank and well/ill-conditioned, then the "parabola" is sharp/flat at this point. Generalized inverse and Exhaustive search (ES) are probably the most suitable ways to cope with R being ill-conditioned. ES is quite simple and easy to understand.

PAGE 44

J ( performance furiction ) J (minimum) W* w figure 4-1.1 : The "parabola" shape of the performance function J vs W in one dimension. W* is the best impulse response vector . 33

PAGE 45

34 Although the purpose of RLS and ES is to find the optimal value of the weight vector W (W optimal ), their approaches are different. RLS bases on the concept of updating vector W (k -1) by just the right amount to generate W(k). The idea behind ES is to search the best impulse response vector W optimal on the performance function J. The following sections will examine exhaustive search technique and its properties. 4-2 Exhaustive search technigue CES) There are several ways to search W optimal on the performance function J: Gauss-Newton, gradient (results in least mean squares algorithm), and exhaustive search (ES) [11]. For simplicity, consider one dimensional case (figure 4-2 on next page). Let W = [ W 1 ] and the optimal weight vector W optimal = W"' = [ W 1 "']. Assume one starts with some initial value of W 1' i.e., W1(0). From W1(0) how should one proceed to the new value W1(1) in such a way that W1(1) gets closer to W1 "'. Let the value of J at W1(0) be R1 and W1(0) = S1 then: R1 = J[W1(0)] = J(S1 ) By increasing S1 by an amount two possibilities can happen: J (S1 ) > J (S1 + or J (S1 ) < J (S1 + a) If J (S1 ) > J (S1 +AS), let : R1 = J (S1 ) and R2 = J (S1 +AS) = J (S2 )

PAGE 46

0 J (performance function) W* 1 J .. mmunum figure 4-2: Schematic diagram of exhaustive search technique in one dimension 35

PAGE 47

36 This situation corresponds with J [W 1 (0)] being on the left side of the parabola representing the performance function J. Therefore, increasing S 1 will make J decrease, it is the hint giving one which direction to go. Next, increase S2 = S1 + 8S an amount 8S, namely: S3 = S2+8S If J (S3 ) < J (S2 ) then R1 = J (S2 ) and R2 = J (S3 ) and increase S3 i.e., S 4 = S3 + 8S evaluate R3 = J (S 4 ) Repeat this procedure until R1 > R2 and R3 > R2 Fitting a parabola R through these three points R1 and R3 : R = a+ bS + cS2 Three unknowns a, b, c are determined if R1 R2 and R3 are known. Take derivative of R with respect to S and set it equal to 0. dR = dS 2cS + b = 0 .. Smin will equal to the new value of W1(0): b 2c Note that the parabola fitting through R1 R2 R3 and the one representing J are not the same and neither are Smin and W 1 *.

PAGE 48

37 = dS = s2-sl = s3-s2 s2 = s1 +dS s3 = s 2 + .dS -s1 + 2dS R = a + bS + cS2 Rl = a + bS1 + cS 2 1 (Sl' Rl) (1) = a + bS 2 + cS 2 2 (S2' (2) -a + bS 3 + cS32 < s3, R3) (3) Use Cramer's rule to find b, c from (1), (2), (3) and Smin = (b/2c): 1 Rl S2 1 1 8,_2 1 b = 1 sl S2 1 1 s2 8,_2 1 s3 S2 3

PAGE 49

b 1 sl 1 s2 1 s3 c = 1 sl 1 s2 1 s3 c = s. = mm 1 1 1 1 1 1 RI S2 1 8S [ R1 -R2+ R3 ] sl S? s2 S2 2 s3 S2 3 Repeat the above procedure until : 38

PAGE 50

39 b) If J (S1 ) < J (S1 + let: In this case J [ W 1 (0) ] is on the right side of the parabola representing the performance function J. Increasing S will make J increase, going in other direction should decrease J. Next step is to decrease S1 an amount of S3 = S1 and let R3 = J (S3). From here on, the procedure is proceeded exactly the same as in case a). ES can be easily extended to L dimensional problem where: w = [ Wo w1 ... wL-1 l In this case each Wi, i= 0, 1, 2, ... L-1 is used to minimize J one at a time (W 0 is first, W 1 is next, ... and finally W L1). After this the process is repeated again and again . until J can not be reduced further or W optimal is achieved. Figure 4-2.1 (page 40) shows the estimates (dotted lines) of the weight vector of the model and the parameters (solid lines) of the weight vector of the system vs number of iterations I. Where: length of vector L = 6 number of data points D = 50 number of iterations I = 10

PAGE 51

40 parameter values 1 .s I 2 8 I \ \ number of iterations \ -.5 \ \ -1 figure 4-2.1 : exhaustive search # of data points = 50 : # of parameters = 6 #of iterations = 10 : stepsize = .1 signal/noise = infinite

PAGE 52

sum of residuals R liE1 -1 1QI 6 -6 liE11 -11 .. 181-16 -16 lQI21 -21 1 R LOC(R) 2 3 4 5 6 1 number of iterations figure 4-2.2 : exhaustive search 41 8 # of data points = 50 # of parameters = 6 # of iterations = 1 0 : stepsize = .1 li signaVnoise = infinite : R = sum of residuals

PAGE 53

42 parameter values 1 .5 I 5 7 ,-u number of iterations -.5 -1 figure 4-2.3 : exhaustive search # of data points = 50 # of parameters = 11 # of iterations = 10 : stepsize = .1 signaVnoise = infinite

PAGE 54

43 sum of residuals R 1911 -1 19I-6 -6 19E11 -11 lil-16 -16 19121 -21 1 2 3 4 5 7 8 19 R LOC(R) number of iterations figure 4-2.4 : exhaustive search # of data points = 50 # of parameters = 11 # of iterations = 10 : stepsize = .1 signaVnoise = infinite : R = sum of residuals

PAGE 55

44 Figure 4-2.2 (page_ 41): plot of sum of residuals R (or J) vs I. Where L = 6, D =50, I= 10. Figure 4-2.3 (page 42) and 4-2.4 (page 43): the estimates, the parameter values, R vs I. L= 11, D =50, I= 10. The data of graphs are in appendix A. One observation that should be pointed out is that the rate of convergence of exhaustive search is very fast. Mter only 5 iterations, the estimates are very good and the sum of residuals R is in the order of l0-9 r.m.s of R is in the order of l0-5 The generation of input x(i) and output d(i) was done exactly in the same way mentioned in chapter III. 4-3 Computation This section attempts to compare RLS and ES computationally. a) ES Let L number of parameters of the weight vector. number of data points used. mult. sub. multiplication subtraction Each y(i) ( filtered output) needs: L(mult.) + (L-1) (add.) LE [L(mult.) + (L-1) (add.)] Each value of R ( R1 R2 R3 ), where: add. : addition div. : division -----> each y(i) .-----> y(i)

PAGE 56

45 4: R (1/ Lr:) L [d(i) y(i)]2 requires: i=l -1) (sub.) + (mult.) + -1) (add.) + 1(div.) + Lr: [L(mult.) + (L-1) ( add.) ] or: LE (L+ 1) number of mult. and LE (L+ 1) -1 number of add. and sub. Let A : average number of times when the situation R1 > R2 and R3> R2 occurs. B : number of iterations for ES ceases to work (optimal values have been achieved). The total number of computations for using ES is: 3.(A).(B) [ LE (L+ 1)] number of mult. 3.(A).(B) [ Lr: (L+1) -1] number of add. and sub. b) RLS Going back to page 20 and 21, let:

PAGE 57

46 L number of parameters of the weight vector. number of data points used or number of iterations. step 4 and 7 require one sub., one add. and one div. steps 3, 6, 8, 9 needs 4L mult. and [(L-1)+(L-1)+L] = 3L-2 add. step 5 : L 2 mult. and L(L-1) add. step 10: L2 mult. and L2 sub. The total number of computations for one iteration: [3]+[4L+3L-2]+[2L2 +L2+ L2 -L] or [2L2 +4L]+[L2 + L2 -L+3L-2+3] [2L2 +4L] (mult) + [2L2 +2L+1] (add., and sub.) The total number of calculations for iterations: LR [2L 2 +4L] number of mult. LR[2L2 +2L+1] number of add., sub., div. As far as number of multiplications is concerned: RLS ES lr;[3.AB.L+3.AB]

PAGE 58

47 Most of the while AB > L. Example: assume RLS andES are used to estimate the parameters of the weight W L = [ W0 W 1* ... W L/l with L = 6. The vjues of the parameters are: W*4 4 --. RLS recursive least squares needs 1,000 iterations in order to achieve the following results: I :: : Sum of residu1 R = 2.417282E-14 w 2 --.8999999 W5 = .8999999 r.m.s of R 1.554761E-7 The number of multiplications for using RLS in this case is: LR[2.L.L+4.L] = 1000[(2)(6)(6)+4(6)] = 96,000 mult. ES : : ES needs 7 iterations (B=7) to achieve the above similar results,le average number of times when the situation R1> R2 I

PAGE 59

48 and R3 > R2 occurs is A= 12 (actually, A= 82/7). The number of data points used is = 30, the stepsize = 0.1. The following results are obtained by using ES: Wo .5000012 WI -.9999988 w2 = -.7999998 w3 -.1999998 w4 --.3999995 ws -.8999994 Sum of residuals R = 2.329294E-14 r.m.s of R -1.526203E-7 The number of multiplications in this case is: = 30[3(82)(6)+3(82)] = 51,660 mult. *) From this example, ES obviously requires less computation than RLS does. *) If adaptive stepsize is used in ES case, then the number of computations can be reduced .further. 4-4 Sample size Box and Jenkins [12] had suggested about 50 observations (data points) are an adequate sample size and is also the minimum required number to build a model. A too small sample size results in poor estimates. While larger one gives more accurate estimates at the

PAGE 60

49 cost of more computations. However, computer experiments had shown 50 data points are more than enough, for the cases of L = 6 and 21 (L : number of parameters) even 20 and 30 data points respectively are sufficient. There is no rule of thumb how big the sample size should be, as L increases so does the sample size. 4-5 Stepsize The speed of convergence and the accuracy depend very much on the stepsize Large makes estimates converge fast to optimal values but the accuracy is sacrified; small has opposite effect. The best approach to compromising these two ( speed of convergence and the accuracy) is to use adaptive stepsize. In the beginning of ES, large is used to ensure fast convergence, as the estimates get closer to the optimal values, small is used (one can choose new equal to one-half, one-tenth ... of the previous one) to achieve desired accuracy. When the matrix R in normal equation RW =Pis suspected to be ill-conditioned, small6S is highly recommended. Most of the results obtained in this chapter use 6S = 0.1 Again there is no specific rule as to what the proper .6S should be. 6S may be as large as 0.1, 1 or as small as lo-s -lo-ts (this depends on machine being used, single/double-precision).

PAGE 61

50 4-6 Input and desired outtmt It is fairly obvious that some conditions on the input sequence x(i) must be introduced in order to secure reasonable identification results. The input must have sufficient rich frequency content to excite all the modes of the system. The exhaustive search technique starts with some initial value ofR: 1 .k R L [e(i)]2 (k-L) i=L-1 1 k R :L [d(i) y(i)J2 (k-L) i=L-1 1 k R = :L [d(i)J2 (k-L) i=L-1 y(i) = 0 i = L-1, L, ... k This initial value of R depends on absolute value of mean (lmeanl) and absolute value of variance (lvariancel) of desired output signal d(i) which in some cases can be controlled by using proper input signal x(i) (which is chosen by the user). Small mean and variance of d(i) are desirable, give small initial value of R.

PAGE 62

51 4-7 Exhaustive search for noisy MA process One of the problems withES which limits its utility is ES's sensitivity to the additive noise ( noise can be added to the input, system or the output). Recall the normal equation: RW = P where k-1 k-1 R = l: X(i)XT(i) and P = l: X(i)d(i) i=L-1 i=L-1 When noise is added to the input x(i), the autocorrelation matrix R will change from R to 8R+R. oR is the result of input noise. Likewise, when noise is added to the output y(i), the correlation matrix P will change from P to 8P+P. 8P is the result of both input noise and output noise. Then the solution will change from W to oW+ W. Let c(R) = IIRII IIR-111 be the condition number

PAGE 63

52 where II II denotes any of the operator norms of R: II 111 II 112 .. Let R be nonsingular and W be the solution to normal equation RW=P. Then from [13] it had been shown that: 118WII 118PII 118RII < M. c(R). [ + ] (4-7.1) IIWII II Pll IIRII Although ( 4-7.1) gives only the upper bound of the ratio 118WII /IIWII it does explain several important points: 1) If 8R and 8P are large then it is reasonable to predict 118WII will be large. 2) Asume that there is no noise added to the input (which is the case for computer simulation in this section). Then 8R = 0, and: and 118WII 118PII < c(R) [ ] upper bound UB IIWII II Pll

PAGE 64

53 UB depends on two terms : c(R) and (118PII I IIPII). *) Assume 118PII I IIPII is a constant. The term c(R) is the condition number of matrix R, small or moderate value of c(R) implies that the equation RW =Pis well conditioned. A large c(R), however, does not imply the equation is ill-conditioned. (In numerical analysis, if "small" changes in the data lead to "large" changes in the solution, the problem is said to be ill conditioned; otherwise, it is said to be well-conditioned.) Although a large condition number c(R) does not indicate the solution W to equation RW = P is ill-conditioned for every P, it is true that W is ill-conditioned for some P. *) Assume c(R) is a constant The term 118PII in 118PII I IIPII corresponds to the additive noise. Large I small noise will lead to large I small 118PII which will make upper bound UB = c(R) ( 118PII I IIPII ) large I small and therefore there will be large I small change 118WII. This explains that as more noise is added to the output (signal to noise ratio decreases ) the estimates become worse. So the change 8W in solution W to normal equation RW = P depends not only on the condition of matrix R (ill/well-conditioned) but also on the "magnitude" of noise (variance of noise). For signal/noise ratio = 10, the estimates are quite different from the actual values of parameters ( figure 4-7.1 on page 54 5 The estimates are getting better ( the difference between the estimates and the parameters become smaller) as signal/noise ratio increases as indicated by figures 4-7 .2, 4-7.3 on pages 55, 56 Figure 4-7.4 is the plot of sum of residuals R vs number of iterations I for the cases S/N = 10, 103 and 105

PAGE 65

54 parameter values 1 1 I I ,1 --==-----------------It I 'l I II .SS l iL, 8 f ..._.,.' n I w a Y:,....-4--, --a a I ., l I number of iterations 1, "---\\ .SS \ I ..,. -1.1 figure 4-7.1 : exhaustive search # of data points = 40 # of parameters = 6 # of iterations = 20 : stepsize = .1 signal/noise = 10

PAGE 66

parameter values 1.1 /. I jl I/ I 55 ..... I{/ ...._. ,, I I 4 55 a-la-12-14-n-u 28 ""' number of iterations \. \-._._, ________ -.55 \ \ -1.1 figure 4-7.2 : # of data points # of iterations exhaustive search = 40 # of parameters = 6 = 20 : stepsize = .1 signaVnoise = l


PAGE 67

parameter values 1.11 I i i. ,I.,_,.-1' I' If .ss :j 56 ,1, .... 1,' , ..... -------------------11/ ... '' hi 'j 9 \'.. 4 ' I ., .... number of iterations ', ... \ -:----. __________________ -.55 \ \. I I \
PAGE 68

1iE1 -1 liE-9 _, R LOC(J) sum of residuals R 1 3 5 1 figure 4-7.4 : # of data points #of iterations signal/noise = 57 11 13 15 17 21 number of iterations exhaustive search = 40 # of parameters = 6 = 20 : stepsize = .1 10, 103 105 : R = sum of residuals

PAGE 69

58 4-8 Overestimate/underestimate the order of the system Before getting into the details of the consequences one will get if exhaustive search technique is used for the mode_! having the order which is greater or less than the order of the system, let's look at the modeling technique widely used in statistics, namely Box-Jenkins methods. The Box-Jenkins modeling procedure consists of three stages for finding a model fitting the available data. stage 1: identification. Choose one of the models among autoregressive (AR), moving average (MA), autoregressive moving average (ARMA) ... models. stage 2: estimation. Estimate the parameters of the model chosen at stage 1. stage 3: diagnostic checking. check the candidate for adequacy. If the model is satisfactory at stage 3, then the procedure stops, otherwise one has to go back to stage 1 or 2 over again until a "good" model is obtained. There are quite a few criterions to make sure a model is a "good" one at stage 3. One criteria is that the model must be able to predict the future value with good accuracy, the other is that a good model must contain the smallest number of parameters. Assume d(n), d(n-1 ), ... d(m) are available data and a moving average (MA) model of order L is used to fit the above

PAGE 70

59 data. The relationship between the weight vector W L = [ W 0 W 1 ... W L ] the input white noise x(k) and the filtered output y(k) (output from MA model) is: y(k) W0* x(k) + W1* x(k-1) ... + WL* x(k-L) The objective of Box-Jenkins technique is to find W0*, W/' ... W L so that y(k) and d(k) are as well matched as possible with the smallest value of L. For the details of how to find W 0 W 1 ... W L and the criterions to judge the "goodness" of these values, the interested readers are referred to [14]. One of the criteria is t-value ( this is an indication of how significantly one estimate is far from zero): The following example will clarify the meaning oft-value). Example: Take Wa* for instance. For a given data d(n), d(n-1), ... d(m) of some process, using Box-Jenkins technique, Wa"' is obtained. Given another set of data d((n'), d(n'-1), ... d(m') of the same process (one process has many realizations. d(n), d(n-1), ... d(m); d((n'), d(n'-1), ... d(m') are just two particular ones), another Wa* will be obtained. Wa* is a random variable, it has normal distribution with mean = r (for simplicity, let r = 0) and standard deviation S. At stage 2 (estimation), Wa* is estimated and its t-value is found (for example) to be 2. What does this This means Wa* is significantly different from r = 0. How is this so?

PAGE 71

60 A rule of thumb in statistics is that only 5% of the possible value W a would fall two or more times standard deviation S away from r = 0. So t = 2 means t = 2 times standard deviation S, and therefore W a is significantly different from r = 0. Finally, this means W a must be included in the MA model to ensure the "goodness" of the model. A model missing W a will lead to bad results such as bad prediction, d(k) and y(k) are not matched, and large sum of residuals. If by mistake, W a is not in the model, stage 3 will verify this and one will go back to stage 2 (or 1) to modify the model. In the Box-Jenkins technique, those parameters having small t-values will have absolute values close to 0, and they do not cons tribute heavily to the value of output y(k) and. hence can be neglected. A parameter is said to be significant in a MA model if it constributes heavily to the value of the output compared with other parameters. A small absolute value of a parameter does not mean that this parameter is not significant. So the term "significant" has relative meaning here. Now let's go back toES. a) Overestimate the order of the system Let L5 and LM be the orders of the system and the model respectively. wLs = [ Wo"' Wt ... wLs J weight vector of the system.

PAGE 72

WI.m -[ Wo W1 ... WLm ] weight vector of the model. 61 The desired output d(k) produced by the system and the filtered output y(k) produced by the model are: d(k) y(k) wox(k) +W1x(k-1) ... + Wox(k) + wl x(k-1) ... + WLm x(k-LM) Because all the parameters of W Ls are included in W Lm, by using exhaustive search, one can expect:_ w -----> w w -----> w w -----> w 0 0 1 1 Ls Ls and WLs+1' WLs+2' WI.m ----->0 and sum of residuals = l:[e(i)] 2 = l:[d(i)-y(i)] 2 is minimized. Example: (computer simulation) Let W* = 5 = 5 : the order of the system. [ -.5 1 -.8 .2 -.4 .9] weight vector of the system. 8 : the order of the model. weight vector of the model. The input white noise x(k) and the desired .output d(k) are generated as before in chapter ill. The number of data points = 37. The number of iterations = 10. Using ES, the estimates of the weight

PAGE 73

vector of the model W 8 : [ .5 .9999999 -.7999999 6.343689E-9 3.395458E-8 sum of residuals = 2.348792-11 d(k) and y(k) are well matched. b) Underestimate the order of the system 62 .2 -.4 .8999999 -1.069999-8 ] Because not all the coefficients of W Ls are included in W Lm' there will be some coefficients of WLs* left out By using ES, the results of its objectives (to minimize sum of residuals, to have the estimates of the model as close to the parameters of the system as possible, y(k) and d(k) are well matched) depend not on how many parameters are excluded but on how significant these excluding parameters are. The following examples will clarify this: Example1: L5 = 8 : the order of the system W* 8 = [ -.5 1 -.8 .2 -.4 .9 104 2. 104 -3.1o-5 J weight vector of the system. 5 : the order of the model. weight vector of the model. The input white noise x(k) and the desired output d(k) are generated as before in chapter III. The number of data points = 37. The number of iterations = 10. Using ES, the estimates of the weight vector of the model W 5 is:

PAGE 74

63 [ .4999712 1.000022 -.8000126 .1999798 -.399953 .9000398 ] sum of residuals R = 3.790102E-9 In this example, the last three parameters of the system:104 2. 104 -3.10-5 have very small absolute values, they are much less significant compare4 to other parameters. Therefore, excluding these last three does not lead to bad result at all. In fact, the sum of residuals is quite small (= 3.790102E-9), the estimates are close to the parameter values and finally d(k) and y(k) are well matched. Example 2: L5 -8 : the order of the system w 8 = [ -.5 1 -.8 .2 -.4 .9 .1 .1 -1 ] weight vector of the system. 7 : the order of the model. [Wo wl ... w7 1 weight vector of the model. The estimates of the weight vector of the model W 7 : [ .3050533 1.158459 -.8904336 .1458553 -.2210489 .8206268 -2.065472E-2 9.711713E-2] sum of residuals R = 7.387126E -2 In this example, W 8 has 8 parameters W 7 has 7 parameters. Although the model misses only one parameter, the results (sum of residuals is large the estimates of the model are not close to the parameters of the system, y(k) and d(k) are not matched) are bad.

PAGE 75

64 This is due to the fact that the excluding parameter ( -1 in this case) is quite significant. Its absolute value (I -11 = 1 ) is more or less equivalent to the absolute values of other parameters). It is noted that the model with a small number of parameters is preferred in practice. However, if this model leads to bad results then one may suspect that some significant coefficients were excluded and therefore increasing the order of the model may be helpful. 4-9 ES for Infinite impulse response CIIR) filters. Self adjusting or adaptive filters (FIR & IIR) have been successfully applied to a wide spectrum of problems, ranging from adaptive control systems, adaptive estimation, to adaptive signal processing. In a broad sense, most applications in the control and communication areas can be regarded as "signal processing". For the most part, signal processing applications. have relied heavily upon well known and well understood adaptive finite impulse response (FIR) filters. However, in real life, there are common situations where FIR filters are not practical, and lead to heavy computation. As the result, in recent years, research has been directed to extending the adaptive filters to more general, complicated yet more efficient infinite impulse response (IIR) filters. One of the advantages of IIR over FIR is the substantial decrease in computation. Exhaustive search has been shown to work successfully for FIR filter without additive noise or when signal to noise ratio is high. In this section, it is attempted to use ES for IIR filters. ES does not work well for IIR as opposed to FIR filters. Several observations deserve to be pointed out: *) In filtering problems, the objective is to minimize sum

PAGE 76

65 of squared errors L[e(i)]2 = L[d(i)-y(i)]2 The estimates of the weight vector of the model and the parameter values of the weight vector of the system may be quite different. These differences may be substantial but are tolerated as long as L[ e(i) ]2 is minimized. In identification problems, the goal is to obtain the estimates as close to the parameter values as possible. For FIR filters, the filtering problems and the dentification problems are achieved simultaneously. This is not so for IIR filters: minimizing sum of squared errors L[e(i)]2 does not make the estimates get close to the parameter values but the inverse is true. *) For MA model (FIR filters), the performance function J = L[e(i)]2 has parabola shape and has one and only one minimum point. On the other hand, for ARMA model (IIR filters), the performance function J has several minimum points. *) To use ES, one must initialize the estimates. Different sets of initial values can lead to different minimum points and in some cases may lead to neither minimizing L[e(i)]2 nor good estimates. Example: The input white noise x(k) = 2*(RND-.5)*SQR(3*1) and the desired output d(k) produced by the system are related as following: d(k) = -.2x(k-1) + .4x(k-2) .5x(k-3) .9d(k-1) + .6d(k-2) Likewise, the relationship between x(k) and the output y(k) produced by the model is:

PAGE 77

66 y(k) = W1x(k-1) + W2x(k-2) + W3x(k-3) + W4y(k-1) + W5y(k-2). Using ES with stepsize = .01, the number data points used= 50 and the initial values of the estimates are chosen to be: wl --.18 w2 = .42 w3 --.51 w4 = -.89 ws = .62 After 40 iterations the following results are obtained: wl --.2115997 w2 = .416828 w3 = -.5088732 w4 --.894435 ws = .5998373 sum of squared errors L[e(i)]2 = L[d(i)-y(i)]2 = 1.21141E-4 From this example, one can see that although the initial values are intentionally chosen to be close to the parameter values, after 40 iterations the estimates do not converge to the parameter values at all. The sum of the squared errors is not very small. Many iterations are needed 4-10 Conclusion For MA model (FIR filter), exhaustive search (ES) works quite well when there is no additive noise. In terms of computation and accuracy, ES is better than recursive least squares (RLS). Especially, when the autocorrelation matrix R in normal equation RW =Pis ill-conditined, ES gives more accurate estimates than RLS does. This is the main reason why ES came into being.

PAGE 78

67 When there is noise involved, none of RLS, ES, generalized inverse works better than the other two. For ARMA model, ES does not work well as explained in section 4-9.

PAGE 79

CHAPTER V DISCUSSION AND CONCLUSION The frrst step in using one of the techniques in chapters 2, 3, 4 is to assume a model. There are several kinds of models, but only MA and ARMA were used in this thesis. The second step is to estimate the order of the model using criterions such as FPE, AIC, CAT ... In identification problems, the objective is to obtain the estimates as accurately as possible. In filtering problems, one is concerned about minimizing the difference between the desired output ( produced by the system or the plant) and the model's output and can tolerate substantial parameter error. Techniques are classified into two broad groups: off-line and on-line. SVD andES belong to the former. RLS is in the latter group. There is great flexibility in selecting computational methods without any restriction on time, as a result the accuracy of the estimates is fairly high for off-line and preferred by statisticians. For engineers, both accuracy and time are crucially important and on-line techniques are preferred although they have less accuracy. In addition, in many situations one could not afford to waste time to collect all the data. In chapter II, several techniques : direct . Inversion Gauss-elimination iterative technique, singular value decomposition

PAGE 80

69 were briefly mentioned tp solve normal equation RW = P. When R is singular, all others fail except SVD and iterative techniques. Chapter III was about recursive least squares algorithm (RLS). RLS is based on the concept of starting with some initial values of the weight vector of the model and then using each input sample (input & desired output) to update the estimates so that the estimates get closer and finally converge to the parameter values of the weight vector of the system. RLS works pretty well when there is no additive noise, the estimates are quite good but requires many iterations (which means heavy computation). Chapter IV is about exhaustive search (ES). ES requires less computation and is more accurate than its counterpart RLS. Most of the properties of ES in chapter IV were not proved mathematically but obtained through computer simulation. Computer simulation is a very important and useful tool for investigating ES in particular and any other algorithms in signal processing. Howerver, a serious limitation is that it may not be conclusive. It is difficult to tell whether a simulation result has universal implication or merely reflects of the chosen data sequence. To obtain results of more general validity one must use mathematical analysis. When there is noise involved, none of SVD, RLS ES works better than the other two. With noise free data ES is the best in term of accuracy. All the techniques discussed in this thesis are applicable for MA models but they do not work for ARMA model.

PAGE 81

BIBLIOGRAPHY. (1] H. Wold, A study in the analysis of stationary time series, Almqvist and Wiksell, Uppsala 1938. [2] Simon Haykin, Adaptive filter theory, Prentice-Hall, 1986, pp. 67-71. [3] Steven M. Kay, Modern spectral estimation-theory and application, Prentice-Hall, 1989, p. 109. [ 4] Michael G. Larimore, John R. Treichler and C. Richard Johnson Jr, Theory and design of adaptive filters, John Wiley & Sons, Inc, 1987, pp. 43-44. [5]. A. Ben-Israel, A note on an iterative method for generalized inversions of matrices, Math. Comp. 20, 1966, pp. 439-440. [6] Campbell + Meyer, Generalized inverses of linear transformation, Pitman publising Limited, 1979, pp. 16-19. [7] see [6], pp 251-255. [8] see [4], pp 91-100. t9] see [4], pp 91-100. [10] see [2], p 392. [ 11] D. A. Ross Constrained optimization of correlation data, private communication, 1989. [12] Alan Pankratz, Forecasting with univariate Box Jenkins models John Wiley, 1983, p. 11. [13] Ben Noble and James W. Daniel, Applied linear algebra, Prentice-Hall, 1977, pp. 170-173.

PAGE 82

71 [14] see [12] [ 15] S. Lawrence Marple, Jr, Digital spectral analysis with applications, Prentice-Hall, 1987, pp. 94-104. [16] see [4], pp. 267-270.

PAGE 83

APPENDIX A This appendix contains the parameter values of the weight vector of the system, the estimates of the weight vector of the model, sum of residuals R, r.m.s of R ... of the graphs in chapters n, ill and IV. The programs are in appendix B. The computer used is Zenith.

PAGE 84

SINGULAR VALUK DiCOMPOSITION NUKBIR OJ PARAKITKRS : 6 KOMBIR OP DATA POINTS : 10 HU!BIR OF DATA POINTS : 20 TU ACTUAL VALOIS TBI ISTIUTIS TBI ACTUAL VALOIS TBI ISTI!&TIS OS 04999994 05 05000004 1 1 1 0 9999998 -08 07999995 -08 -07999998 0 2 0 2 0 2 01999997 -04 -04000004 -04 -03999995 0 9 09000002 0 9 08999994 HUKBIR Ol P&RA!ITIRS : 11 NUKBIR or DATA POINTS : 10 MUKBIB OF DATA POINTS : 20 TBK ACTUAL VALOIS THE ISTI!&TIS THI ACTUAL VALOIS THI ISTI!ATKS 05 05000003 05 o5 1 0 9999993 1 0 9999999 -08 07999996 -08 -07999991 02 02000003 02 01999995 -04 -04000023 -04 -03999993 09 09000001 09 0 8 .8000005 0 8 .7999998 .4 04000015 04 .3999998 -.3 03000021 -03 -03000001 -1 -.9999984 -1 -1 075 07499993 .75 07500001 HO!BIR OF PAR&KITIRS : 20 MOKBIR Or DATA POINTS : 10 MOKBIR or DATA POIRTS : 20 TBI ACTUAL VALOiS TBI ISTI!ATIS TBI ACTUAL VALOIS TJI ISTIK&TIS o 5 03176904 05 05000009 1 10041876 1 09999974 -08 -1.058452 -08 -08000009 .2 08182513 02 02000022 -04 -01677476 -04 -04000015 .9 1044248 09 08999991 08 1.257589 08 07999984 04 06428237 04 03999972 3 -02983337 3 -.299999 -1 -10186358 -1 -09999976 075 6808423 075 07500015 105 .7937741 1 5 105 101 .6679647 101 10100001 -06 -01443536 -06 .6000001 -10 2 07956138 -102 -10200001 -012 -501061331-02 012 01200018 1.55 09926724 1.55 10550001 -1.21 -1.109791 -1.21 -1.209999 -066 06846545 -066 -06600013 -109 -1.171808 -1.9 -10899997 73

PAGE 85

74 SINGULAR VALUE DiCOMPOSITIOH NOMBKH OF PARAMITKRS : 6 HOKBKR OF DATA POINTS : 40 SIGHAL/NOISi : 10 THE ACTUAL VALOIS 5 1 -.8 2 -.4 9 SIGHAL/HOISE : 10 3 THE iSTI!ATES .496573 1.035976 -.8325791 .1553249 -.3893638 .9987332 THi ACTUAL VALOIS THE iSTIKATIS .5 .4996572 1 1.003597 -.8 -.8032578 2 .1955324 -.4 -.3989365 .9 .9098734 SIGMAL/HOISi : 10 TBK ACTUAL VALOIS THi ISTI!AT!S 5 4999657 1 1.000359 8 .8003258 2 .1995532 -.4 -.3998937 9 9009871 SIGHAL/HOISi : 100 THI ACTUAL VALOIS .5 1 -.8 .2 -.4 .9 SIGHAL/HOISI : 10 THE KSTI!!TiS 4989162 1. 011376 -.8103024 .1858725 -.3966366 .9312219 THI ACTUAL VALOIS TBI ISTIMATIS .5 .4998915 1 1.001137 -.8 -.8010301 2 .1985872 -.4 -.3996637 .9 .903122 SIGHAL/HOISI : 10 THI ACTUAL VALOIS THI ESTIMATES .5 .4999891 1 1.000113 -.8 .800103 .2 .1998586 -.4 .3999664 .9 .900312

PAGE 86

75 RECORSIYI LIAST SQOABES ITKR G( Ol G( 1) G( 21 G( 3) G( 4) G( 51 .5 1 -.8 .2 -.4 9 10 7746811 1.130828 -1.128002 -3.5621721-02 -.471651 1.098035 50 .5452403 9960952 .8340062 .1867677 -. 4260443 .9096918 100 5151973 .9967139 -.8112345 .1996619 -.4098652 .9010181 200 .5037134 .9995578 -.8026518 .1991361 -.4027318 9003244 300 5011802 9997643 .8008996 .1999186 -.4008045 .9000525 400 .5004199 .9999334 -.8003286 .1999284 -.4003068 9000047 soo .5001542 .9999701 .8001208 600 .19997 46 -. 4001105 . 9000018 .5000513 .9999892 -. 8000393 700 .1999934 -. 4000412 .9000002 .5000183 9999971 .8000151 800 .1999946 -.4000134 .9000005 .5000066 .9999981 -.8000058 .1999989 -.4000034 .8999992 900 .5000021 .9999993 -.800002 .1999992 -.4000017 .8999996 1000 .500001 .9999999 .8000009 .1999996 -.4000007 .8999999 ITKR SOH OP RISIDOALS R RHS Of R 10 3 .0085341-03 5. 4850111-02 so 3.5459211-05 5.9547641-03 100 4.051348E-06 2.0127961-03 200 2.6224081-07 5.120946i:-04 300 2 .6893851-08 1.6399351-04 m 3.5868891-09 5.989065i-OS 500 4 .8223821-10 2 .1959921-05 600 5. 622787ill 7 4985241-06 700 7. 4078711-12 2. 7217 411-06 800 8. 7825141-13 9. 3115071-07 900 1.1551761-13 3.398788-07 1000 2 4172821-14 1.5547611-07

PAGE 87

76 RiCORSIVK LEAST SQUARES IUR 50! OF BKSIDDALS B RKS or B Gl 0) G( 1 l G( 21 G( 31 G( 4) G( Sl 1 -.8 .2 -.4 9 G( 61 G( 7J G( 8) G( 9) G(lOJ .8 .4 -.3 -1 .75 10 .3739014 .6114148 .3132842 -.7471896 -.5868756 .6026376 1. 949431 3.516864 -.1138547 -3.275992 -4.25797 -.3636949 0 100 1.3392871-04 1.1572761-02 5374023 .9385386 -. 7251513 .148172 -.399184 .95004 .7519163 .4286757 -.3196768 -.9856912 7417548 200 5.951337i-06 2.439537i-03 .50916 9874006 -.7856224 .1898938 -.3987489 .9085028 .7906353 U69814 -.3052762 -.9977764 .7502281 300 6.182849i-07 7. 86311i-04 .5026838 .9960766 -.7952621 .1966868 -.3996987 .9031326 .7970946 .4017215 -.3012052 -.999275 .7496806 400 7. 897627i-08 2.8102721-04 .9985184 -.7982588 .1987953 -.3997553 .9009021 7991148 .4005667 -.3005791 -. 9996179 .7498695 500 8.681652i-09 9.3175391-05 .5002956 .9995299 .7994212 .1995945 -.3999376 .9003593 .7996843 .4001393 -.3000817 -.9999771 7500116 700 1. 8348781-10 1.354577i-05 .5000549 .999935 -.7999253 .1999458 -.3999893 .9000444 7999479 .4000371 -.3000301 -.9999775 .7500006 800 l.6263U-11 4.03279i-06 .5000117 .9999835 -.7999765 .1999824 -.3999987 .9000178 .7999853 .4000045 -.3000032 -.9999984 .7500031 900 1. 937311-12 1.3918731-06 .5000042 .9999939 -.799992 .1999948 .3999994 .9000054 .7999946 .4000025 -.30000U -.9999996 7500013 1000 3. 948236E-13 6.28351:-07 .5000022 .999997 -.7999966 .1999977 -.3999992 .9000018 7999971 .4000009 -.3000009 -.9999987 .7499995

PAGE 88

77 RECURSIVE LEAST SQUARES I OF DATA POINTS = 1000 I OF ITERATIONS : 101)1) I OF PARAftETERS = b SIGNAL/NOISE RATIO = 10 ITER 61 Q) 6( 1) 6( 2) 6( 3) 6( 4) 6( 51 .5 1 -.8 ., -.4 .9 oL 10 7141346 1.623206 -1.253718 .3473605 -.8684159 1.321568 so .5554639 1.008998 -.8J5n7 .1796957 -.2713918 .8400678 100 .5628154 1. 085181 -.8333776 .1674355 -.3981438 .8587018 201) .56825 1.013933 -. 7777376 .2419255 -.3939645 .9182339 300 .4816526 1.047915 -.8252398 .1660199 -.3861295 .9101685 400 .4704792 .9298575 -.7594174 .1753052 -.4359998 .9638952 500 .4895214 .9352865 -.7741495 .1556807 -.3927307 .9340479 600 .4750061 .9602142 -.8096293 .1676619 -.4140025 .8831844 700 .4981464 .9094558 -.8344949 .1883573 -.4540443 .9572498 BOO .4886953 .9992925 -.7952107 .1426926 -.4640255 .9530052 900 .9905889 -.8569975 .1890078 -.446578 .9331438 1000 .5382441 1.036235 -.8014445 .1752481 -.4333977 .9247265 ITER SU" OF RESIDUALS R R"S OF R 10 .1036149 .3218926 50 2.259255E-02 .1503082 100 2.756636-02 .1660312 201) 2.617706E-02 .1617933 300 2.812892-02 .1677168 400 2.957574-02 .171976 500 2.878968-02 .1696752 600 2.763557-02 .1662395 700 2.909535-02 .1705736 BOO 2.908139-02 .1705327 900 2.B32117E-02 .168289 tOOO Z.B21182E-02 .16796l7

PAGE 89

78 RICURSIVI LIAST SQUARIS I OF DATA POINTS : 1000 I OF ITIRATIONS : 1000 I OF PARAMRTIRS : 6 SIGNAL/NOISI RATIO : 10"3 ITIR Gl 0) G( 1) G( 2) G( 3) G( 4) G( 5) .5 1 -.8 .2 -.4 .9 10 .6259846 1.680235 -1.456218 .5585368 -1.124058 1.375298 50 .5195635 1.044977 -.8165253 .1758852 -.3916199 .8932839 100 .5130924 1. 027508 -.8082988 .1858427 -.4041152 .8979336 200 .5094396 1.006385 -.7986541 .201(733 -.4014737 .9012273 300 .4988066 1.006244 -.8027991 .195801 -.3989227 .9011411 400 .4972594 .9934939 -.7960612 .1972776 -.4037133 .9063842 500 .4990446 .9937102 -.797448 .1954711 .3993407 .9033865 600 .4975378 .9960973 -.800975 .196721 4014246 .8983206 700 .4998269 .9909719 -.8034554 .1988193 -.4054132 .905725 800 .4988737 .9999376 -.7995225 .1942649 -.4064055 .9052991 900 _5004745 .9990618 -.8057001 .1986983 -.4046593 .9033137 1000 .503825 1. 003624 -.8001443 .1975238 -.4033402 .9024724 nn SUM Of RISIDOALS R RKS OF R 10 .111524 .3339521 50 4 .8058391-04 2.1922221-02 100 3.4965491-04 1.8699061-02 200 2.6208371-04 .016189 jOO 2. 819115i-04 1.679022!-02 400 2.9488251-04 1. 7172141-02 500 2 .8764851-04 .0169602 600 2.7630761-04 .0166225 700 2 .9091831-04 1. 7056331-02 BOO 2.908191-04 1. 7053421-02 900 2.8321251-04 1.6828921-02 1000 2.8211941-04 1. 6796411-02

PAGE 90

79 RECURSIVE LEAST SQUARES t OF DATA POINTS = 1000 t OF ITERATIONS = 1000 t OF PARAKETERS = 6 SIGNAL/NOISE RATIO = 10A5 lTER S( 0) 6( 1 I 6( 21 S( 31 S( 4) 6( 5) .5 .8 .2 -.4 .9 10 .6171702 1.685939 -1.476469 5796545 .149621 1.380671 50 5159735 1.048575 .814634 .1755042 -.4036427 .8986057 100 .5081202 1.02174 .8057909 -.4047122 .9018568 200 .5035586 -.80(17459 .1974281 -.4022246 300 .5005219 1.002077 -.8005549 .1987i91 -.4002021 .9002382 400 .4999376 .9998567 .7997253 .1994747 -.4005506 9006331 500 .4999968 .9995522 -.7997778 .1994501 -.4000018 .9003204 600 .4997913 .9996855 -.8001095 .199627 -.4001668 .8998341 700 .4999951 .9991236 -.8003518 .1998655 -.40055 .9005722 BOO .4998914 1.000002 -. 79995.J8 .1994222 -.4006435 .9005283 900 .5000495 .9999096 -.8005707 .1998874 -.4004673 9003305 1000 .5003832 1.000363 8000145 .1997514 -.4003345 .9002467 ITER SU" OF RESIDUALS RIIS OF R 10 .1138336 .337l924 5(1 3.753803-04 1.937473E-02 100 7.07027-05 8.408489E-03 200 7 .186342E-Ob 2.680H6E-03 300 3.24779-06 1.802163-03 400 2.906356-06 1.704B04E-03 500 2.857906-06 1.690534-03 600 2.759286-06 1. 66111E -03 700 2.905691-1)6 1.704609-03 BOO 2.908694-06 l.705489E-03 900 2.B32199E-06 1.6B2914E 1000 2.821322E-O& 1.679679E-03

PAGE 91

80 &IHAUSTIVK SEARCH or PARAHETXRS = 6 or DATA= so sTxPsrzx = .1 ITIR G(O) G(l) G(2) G(3J G(4l S(51 .5 1 .:8 .2 -.4 .9 1 .5059382 .9435526 -.7606144 2 .2665095 -.3501287 .8856738 .4898913 .9842024. -.8083669 3 .2013362 .-.396082 .9004196 .5007988 .9994779 -. 8007211 .1996734 -.3998517 .9000429 4 .500106 1.000037 -.8000151 5 .1999681 -.400004 .9000013 .5000042 1. 000005 -.7999982 6 .1999989 -.4000009 .8999999 4999999 -.7999998 7 .2000001 -.4 9 .5 -.8000001 .2 -.4 9 8 .5 -.a 2 -.4 .9 9 .4999999 1 -.8 10 .2 -.4 .9000001 5 -.8 .2 -.4 .9000001 ITIB SUI Of BISIDOALS R RIS Of B 1 9.8215951-05 9.9103961-03 2 3.7274331-06 1.9306561-03 3 1 .2270571-08 1.1077261-04 4 1.0987291-10 1.0482031-05 5 3.9942971-13 6.3200451-07 6 9.0767671-16 3.0127671-08 7 2.8151091-16 1.6778291-08 8 1.7437441-16 1.3205091-08 9 2 .9289071-16 1. 7114051-08 10 5.1209041-17 7.1560491-09

PAGE 92

81 iiB!OSTIVi SEARCH or PARA!iTXBS = 11 s or DATA= so STEPSIZI = .1 lTiR G(O) G(l) G(2) G(3) G(4) G(5) .5 1 -.8 .2 -.4 .9 G(6) G(7) G(8) G(9) G(lO) .8 .4 -.3 -1 .75 1 .5373538 .7671535 -J518795 .3710172 -.2474033 .8677823 .5753536 .2970887 -.274225 -.8567359 .6867448 2 .4582705 .9230392 -.7792879 .2486658 -.3506067 .9031748 .7671306 .370235 -.3036527 -.984106 .7401089 3 .4943438 9846756 -.802288 .2061994 -.3891598 9035449 .7969101 .3961909 -.3013491 -.9980882 7484755 4 5001471 .9975412 -.8012644 .2001024 -.3986763 .9007799 .7998453 .3997685 -. 3003469 -.9991742 .7497903 5 .5003152 .9997835 -. 8002174 .1998897 -.3999286 9001197 .8000241 .4000466 -.300061 -.999966 .7499876 6 .5000117 .999998 -.8000235 .1999727 -. 4000141 .9000104 .8000073 .4000138 -.3000088 -.9999961 .7500013 7 .5000103 1.000004 -.8000005 .1999964 -.4000042 .9000001 .8000013 .4000022 -. 3.000009 -.9999996 .7500006 8 .5000007 1.000001 -.7999998 .1999997 -.4000007 .8999999 .8000001 .4000003 -.3000001 -1 .7500001 9 .4999999 1 -.8 .2 -.4000001 9 .8 .4 -.3 -.9999999 .H00001 10 .5 1 -.8 .2 -.4 .9 .8 .4 -.3 -1 .75 ITIR SO! OJ BISIDU&LS R R!S or H 1 2. 048341-03 ."0452586 2 9.5155791-05 9.7547841-03 3 3.1143691-06 1.7816761-03 4 7.4418761-08 2.727981-04 5 1. 5282911-09 3.9093441-0S 6 4. 5890511-11 6.7742541-06 7 1.1259811-12 1. 0611231-06 8 1.5163771-14 1.23H131-07 9 3.4534881-16 1.8583561-08 10 3.1065431-16 1. 7625391-08

PAGE 93

82 EIHAUSTIYE SEARCH I OF DATA POINTS = 40 I OF ITERATIONS = 10 I OF PARAftETERS = 6 SIGNAL TO NOISE RATIO = 10 ITER 6(01 6(11 6(21 6(31 6(41 &!51 .5 -.8 .2 -.4 .9 1 .5791272 1.100057 -.9108996 -5.006526-02 -.1510451 ,9236923 2 .4647143 1.078307 -.7706338 4.498323E-02 -.3231962 .9823988 3 .5030513 1.025966 -.7840217 9.528696E-02 -.3584524 1.001091 4 .5203474 1.020781 -.7918397 .1086689 -.3681998 1.004547 5 .5236301 1.020338 -.7943437 .1121429 -.3702465 1.005102 6 .5242792 1.020436 -.7950426 .1128936 -.3706305 1.005157 7 .5243801 1.020503 -.7952063 .113036 -.3706873 1.005152 8 .5243895 1.020526 -.79524 .UJ0576 -.3706916 1.005147 9 .5243883 1.020532 -.795246 .1130596 -.37069011 1.005145 10 .5243871 1.020534 -.7952468 .1130592 -.3706898 1.005145 ITER SUft OF RESIDUALS R RRS OF R STEPS lZE = .1 1 1 2.1155537-02 .1629582 2 1.888654-02 .1374283 3 1.801856-02 .1342332 4 1.795079-02 .1339805 5 1. 794 738E -02 .13391178 6 1.794722E-02 .1339672 7 1.794722-02 .1339.72 8 1.794721-02 .1339672 9 1.794722-02 .1339672 10 1. 794721-02 .1339672

PAGE 94

83 EXHAUSTIVE SEARCH I OF DATA POINTS = 40 I OF ITERATIONS = 10 I OF PARA,.ETERS = 6 SIGNAL TO NOISE RATIO = 10A3 ITER G!OI G(11 6(2) 6(31 6(4) 6(51 .s 1 -.8 .2 -.4 .9 .5476148 1.067056 -.8826056 2.413719-02 -.1925286 .8378168 2 .447068 1.050989 -.7726919 .1272363 -.3537032 .8907127 3 .4834038 1.006088 -.7886431 .1750563 -.3862042 .9071142 4 .4989363 1.002076 -.7963551 .1873996 -.39492 .9100473 5 .5018057 1.001835 -.7987092 .190512'l -.3967018 .9104'J32 6 .5023555 1.001958 -.7993452 .1911683 -.3970246 .'ll05294 7 .502436 1.002025 -. 7994901 .1912885 -.397069 .'l10522 8 .5024418 1.002046 -. 7995193 .191J056 -.3970712 .9105169 9 .5024401 1.002052 7995241 .1913066 -.3970699 .910515 10 .5024391 1.002053 -.7995248 .1913062 -.3970692 .9105146 ITER ITER SU" OF RESIDUALS R R"S OF R 1 7.26021-03 8.520686-02 2 9.711315E-04 3.1162'l8E-02 l 2.380349-04 1.542838-02 4 1.823075E-04 1.350213-02 5 1.796037-04 1.340163-02 6 l.794774E-04 1.339692-02 7 1. 794 722E -04 1.JJ9672E-02 8 l.794721E-04 1.339672-02 9 1. 794 722E -04 1. 339673-02 10 1. 794719-04 1.3396 71E -02

PAGE 95

84 EXHAUSTIVE SEARCH I OF DATA POINTS = 40 I OF ITERATIONS = 10 t OF PARA"ETERS = 6 SIGNAL TO NOISE RATIO = ITER 6(0) G(l) 6(2) 6(3) 6(41 6(5) .s 1 -.8 .2 -.4 .9 1 .5444685 1.063756 -.8797762 3.155838E-02 -.1966778 .8292298 2 .4453036 1.048258 -.7728978 .1354619 -.3567541 .8815441 3 4814392 1. 0041 -.7891053 .1830333 -.3889794 .8977166 4 .4967953 1.000206 -.7968065 .1952727 -.397592 .9005974 5 .4996233 .9999848 -.7991458 .1983499 -.3993473 .9010324 6 .5001632 1.000ll -. 7997753 .1989958 -.3996641 .9010668 7 .5002416 1.000177 -.7999186 .1991137 -.3997072 .901059 B 5002471 1.000198 -.7999472 .1991304 -.3997092 .9010538 9 .5002453 1.000204 -.799952 .1991313 -.3997079 .9010521 10 .5002443 1.000205 -.7999525 .1991309 -.3997072 .9010516 ITER SU" OF RESIDUALS R R"S OF R STEPSIZE = .1 1 b.946154E-03 B.334359E-02 2 7. 797317E -04 2.792368E-02 3 5.916066E-05 7.691597E-OJ 4 4.56192E-06 2.135865E-03 5 1.922627E-06 1. J8b58BE -OJ b 1.799b4BE-06 1.34151E-03 7 1. 794914E-06 1.J39744E-OJ B 1.7'l4741E-Ob 1.JJ967'lE-OJ 9 1. 79473'lE -06 l.JJ9679E-03 10 1. 794714E -06 1.3J9669E-03

PAGE 96

APPENDIXB PROGRAM LISTINGS All the programs in this appendix are written in GW-BASIC language. program "Singular value decomposition" is from [15] it was written in FORTRAN and translated into GW-BASIC. *program "Recursive least squares is from [16], it was also written in FORTRAN and translated into GW -BASIC. the major and main part of program "Exhaustive search algorithm" : subroutine parameters search" was developed by professor Douglas 'A. Ross [ 11].

PAGE 97

10 . 20 '11111111111111111 SINGULAR VALUE DECDftPOSITION 111111111111111111111 30 . 40 'DEFDBL O,M,E,R,T,Y,U,O,P,A,S,D,F,G,H,Z,l,C,V,B SO ft=30 : L=20: ftftft=ft-L 60 Dlft llfti,Difti16ILI,YIL,1l 70 FOR TO ft-1 80 l(K)=21(RND-.SIISGR(Jt.1l 90 NEXT K 100 610 1=.5 : 6(1 1=1 110 6(4 1=-.4 : 6(5 1=.9 120 6(8 1=-.3 : 6(9 )=-1 130 61121=1.1 : 6(13)=-.6 140 6(161=1.55 : 6(17)=-1.21 150 FOR K=L-1 TO ft-1 :6(2 ) = -.8 :Sib l =.8 :6(10) =.75 :6(141 =-1.2 :6(19) =-.66 :6(J 1=.2 :6(7 )=.4 :6(11)=1.5 :6(15) :-.12 :6( 19)=--1. 9 160 DIKI= 6(0)1l(K)+6(1)11(K-11+6(211l(K-2)+6(3)11(K-3)+6141ti(K-41 170 DIKI= DIKI+61511l(K-51+6161tX(K-61+G(7111(K-71+61BIII(K-81 180 Dill= DIKI+6(9)tl(K-91+6(1011IIK-101+6(11)11(K-111+6(12111(K-121 190 DIKI= DIKI+6(131111K-131+6(14)1l(K-14)+6(1SIII(K-15)+6(1611l(K-161 200 DIKI= 210 NEXT K 220 '111111111111111111111 COftPUTE R & P 11111111111111111111111111111111 230 Dift P(L-1111,RIL-1,L-11,AIL,L),ASIL,LI 240 KK=ft-1 250 FOR I=O TO L-1 260 T=O 270 FOR J=L-1 TO KK 280 T=T+D(J)Il(J-11 290 NEXT J JOO PII11l=T : Y(I+1,1)=P(I,1) 310 NEXT I 320 FOR I=O TO L-1 lJO FOR J=l TO L-1 340 T=O 350 FOR Jl=tl-11-1 TO IKK-11 l60 T=T+l(Jl)ll(J1-J+Il l70 NEXT Jl 380 RII,JI=T : R!J,II=RII,Jl J90 AII+1,J+1l=RII1J) : A(J+l11+1)= A(I+11J+ll 400 NEll J 410 NEll I 420 11111111111111&1&1 PRINT P AND R 11111111111111&111111111111&111111 430 PRINT 1 THE ELEftENTS DF (L-1)11 ftATAil PARE (l ELEftENTS): 440 PRINT THE ELEftENT DF ftATRil p STARTS (0,1) I 450 FOR 1=0 TO L-1 460 PRINT Ptl,11 470 PRINT 480 NEXT I 490 PRINT THE ELEftENTS OF ftATRil (L-1)1(L-1) RARE (lll ELEftENTS): 86

PAGE 98

500 PRINT THE ELEftENT OF ftATRII R STARTS (010) 510 FOR 1=0 TO L-1 520 FOR J=O TO L-1 530 PRINT R(I1Jl1 540 NEIT J 550 PRINT 560 NEXT I 570 .ENTER A1ft1N1NU,NV1S1U1V11P 580 aaaaatllllllllllt SINGULAR VALUE DECOftPOSITION lllllllllltlllllllllll 590 'llllflllflllllllllllllll A=UISIY(H) ttlllllllllllllllllllllllllllllll 600 . A="'" ftATRII I U="'" I S=ftiN I V= NIH 610 ft=L : N=L : NU=ft : NV=N : IP=O 620 "ftAX=" : NftAI=N 630 Dlft U(ft1ft)1V(N1N)1UH(ft1")1VH(N1N)15"1"1N)11Sft(N1ft) 640 Dl" SINJ,B(l00)1C(100)1T(100) 650 ETA= 1.2-07 : TOL= 2.4E-l2 660 NP=N+IP :Nl=N+l 670 FOR 1=1 TO ft 680 FOR J=1 TO N 690 A5(11J)= A(I,JI 700 NEXT J 710 NEll I 720 730 740 750 760 770 780 'lllllllllllltllllllllll HOUSEHOLD REDUCTION tllllllllllllllllllllllll Clli=O K=l Kl=K+l 'lllllllllllllllll EllftiNATION OF A(l1k)1I=K+l1 1 111111111111111 Z=O FOR I=K TO ft 790 Z=Z+A(I1K)A2 BOO NEXT I 810 9(1)=0 820 IF Z<=TOL GOTO 1060 BJO Z=SDRill 840 BIKl=Z 850 W=AB5(A(k1K)) 860 0=1 870 IF W>
PAGE 99

990 NEIT J 1000 '8111tllllllltllllllll PHASE TRANSFORftATION tllllllttlllllllltltllltl 1010 0=-A(K,K)/ABS(A(K,KII 1020 FOR J=K1 TO NP 1030 A(K,J)=QIA(K,J) 1040 NEXT J 1050 'tllltllttlttlllll ELI"INATION OF A(K.J)1J=K+21 ,,, 1N lllltlttlllltl 1060 IF K=N 60TO 1370 1070 l=O 1080 FOR J=K1 TO N 1090 1100 NEXT J 1110 C(ll 1=0 1120 IF 1<= TOL 60TO 1340 1130 Z=SORIZI 1140 C(Ul=Z 1150 M=ABS(A(K,K1ll 11&0 0=1 1170 IF W>EPS THEN EPS=S(K)+T(K) 1420 NEH K 1430 EPS=EPStETA 1440 '11111111111111111111 INITIALIZATION OF U AND V llllllllllllllttlttllltt 1450 IF NU=O 60TO 1520 14&0 FOR J=l TO NU 1470 FOR 1=1 TO 88

PAGE 100

1480 U(I,J)=O 1490 NEXT I 1500 U(J,Jl=1 1510 NEXT J 1520 IF NY=O 6DTO 1600 1530 FOR J=1 TO NY 1540 FOR 1=1 TO N 1SSO YII,J)=O 1560 NEIT I 1570 YIJ,J1=1 1580 NEXT J 1590 '111111111111111111111 DR DIASONALIZATION 1111111111111111111111111111 1600 FOR KK=l TO N lla10 K=NHK 1620 'llllllllltllllllllllll TEST FOR SPILT 111111111111111111111111111111 1630 FOR LL=1 TO K 1640 L=K+1-LL 1650 IF ABS(TILll<=EPS THEN GOTO 1970 1660 IF A8S(S(L-1ll<=EPS GOTD 1690 1670 NEIT LL 1690 '11111111111111111111111 CANCELATION OF BILl lflllllfflfllllllllllllll 1690 CS=O 1700 SN=1 1710 ll=L-1 1720 FOR I=L TO K 1730 F :oSNU II I 1740 TII)=CSIT(I) 1750 IF ABSIF)<=EPS SOTO 1970 1760 H=SIIl 1770 W=SDR(FA2+HA2) 1780 5(1)=16 1790 CS=H/11 1900 SN=-F/N 1810 IF NU=O SOTO 1990 1920 FOR J=1 TO N 1830 I=U(J,lll 1840 Y=UIJ,I) 1850 UIJ,Ll)=IICS+YISN 1860 U(J,I)=YICS-IISN 1970 NEXT J 1880 IF NP=N GOTO 1950 1890 FOR J=Nl TO NP 1900 O=Aill,J) 1910 R=AII,J) 1920 AIL1,J)=OICS+RISN 1930 AII,J)=RICS-OISN 1940 NEXT J 1950 NEIT I 1960 '1111111111 TEST FOR CONVERGENCE llllllllllllllllfllllllllllllllllfllll 89

PAGE 101

1970 II=S(q 1980 IF L=K SOTO 2580 1990 '11111111111111111111111 ORIGIN SHIFT llttllttlll11tlltlllllllllllttll 2000 l=Sill 2010 Y=S(K-11 2020 G=T( l-1 I 2030 H=T(K) 2040 F=((Y-II)I(V+II)+(6-H)t(6+H)l/(21HIY) 2050 2060 IF F
PAGE 102

2460 FOR J=N1 TO NP 24 70 Q=A ( 1-1, J ) 2480 R=AII, J) 2490 AII-11J)=QtCSRISN 2500. A(I,Jl=RICS-GISN 2510 NEIT J 2520 NEXT I 2530 l(Ll=O 2540 T(Kl=F 2550 SIKl=l 2560 GOlD 1630 2570 'lllllllllllllllllllllll CONVERGENCE lllllllllllllllllllllllllllllllll 2580 IF W>=O GDTO 2640 2590 S!Kl=-11 2600 IF NY=O SOlO 2640 2610 FOR J=l TO N 2620 VIJ,K)=-VIJ,Kl 2630 NEXT J 2640 NEXT U 2650 '1111111111111111111 SORT SINGULAR VALUES llllllllllllllllllllllllllll 2660 FOR k=l TO N 2670 6=-l 2680 J=K 2690 FOR l=l TO N 2700 IF 5111<=6 GOTO 2730 2710 6=5(1) 2720 J=l 2730 NEXT I 2740 IF J=l 60TO 2950 2750 S(J)=SI ll 2760 SIKl=& 2770 IF NV=O 60TO 2830 2780 FDA 1=1 TO N 2790 O=VII I J l 2800 VII,J.l=Y( I ,ll 2810 YII,Kl=O 2820 NEXT I 2830 IF NU=O SOlO 2890 2840 FDA 1=1 TO N 2850 Q=U!I,Jl 2860 UII,Jl=Uti,Kl 2870 U(I1K)=Q 2880 NEXT I 2890 IF N=NP 60TO 2950 2900 FDA I=Nl TD NP 2910 Q=A(J11) 2920 AIJ,II= A(K,II 2930 A( K, ll=Q 2940 NEXT I 91

PAGE 103

2950 NEXT l 2960 tttttlllltta BACK TRANSFORftATIDN lltlllllltttlllltlltlltttlllllllllll 2970 IF NU=O GOTO 3160 2980 FOR KK=l TO N 2990 K=N1-U 3000 IF BIKI=O GOTO 3150 3010 0=-AIK,K)/ABSIAIK,Kll 3020 FOR J=1 TO NU 3030 U(K,J)=OIU(K,Jl 3040 NEXT J 3050 FOR J=1 TO NU 3060 0=0 3070 FOR I =K TO II. 3080 Q=O+A(I,K)tU(I,JJ 3090 NEXT I 3100 0=0/ ABS(A(l,KiiBil)) 3110 FOR I=K TO II 3120 U(I,J)=U(I,J)-OIA(I,Kl 3130 NEXT I 3140 NEXT J 3150 NEXT U 3160 IF NV=O GOTO 3510 3170 IF N<2 60TO 3510 3180 FOR KK=2 TO N 3190 K=N1-U 3200 U=K+1 3210 IF C(K1J=O GDTO 3360 3220 O=-AIK,K11/A8S(AIK,K1ll 3230 FOR J=l TO NY 3240 V(K11J)=OIV(K1,J) 3250 NEXT J 3260 FOR J=l TO NV 3270 0=0 3280 FOR l=l1 TO N 3290 O=O+A(K,l)IV(l,Jl 3300 NEXT I 3310 O=O/ABSIAIK,K1)tCIK1ll 3320 FOR I=Kl TO N 3330 YII,Jl=V(I,JJ-DIA(K,II 3340 NEXT I 3350 NEXT J 33110 NEIT KK 3370 PRINT"USE SINGULAR VALUE DECOIIPOSITION TO SOLVE:" 3380 PRINT "R(LILI I M(Lil) = P(llll IIATRII EQUATION" 3390 PRINT "LET R(LIL)=AS(LtL) P(Lil) = Y(Lil)" 3400 PRINT"USE SINGULAR VALUE DECOftPDSITION TO SOLVE:" 3410 PRINT "ASilll) I W(lll) = Y(Lil) ftATRII EQUATION" 3420 PRINT AS=UI SIIIV(Hl l4l0 PRINT U(H)IASIY=SII 92

PAGE 104

3440 PRINT W=YIIS"IUIHIIY 3450 PRINT U"'" UIHER"ITIANl-"'"" 34&0 PRINT AS"IN" 3470 PRINT YNIN, YIHEAftiTIANI-NIN" 3480 PRINT S"-"IN, IS" -Nift" 3490 PRINT THE SINGULAR VALUE DECOftPOSITION OF AS KIN ftATRII" 3500 PRINT I U(HERftlTIANIIASIY=S"" 3510 '111111111111111111 PRINT SINGULAR VALUES 1111111111111111111111111111 3520 PRINT "THE SINGULAR VALUES ARE:" 3530 FOR 1=1 TO N 3540 PRINT Sill, 3550 NEll I 3560 PRINT 3570 '11111111111111111111 "ATRIX UIHEA"ITIANI 11111111111111111111111111111 3580 FOR 1=1 TO 3590 FOR J=l TO ft 3600 UHII,JJ=UIJ,I) 3610 NEXT J 3620 NEXT 1 3630 '11111111111111111111 PRINT THE "ATRIX U lttllllllllllllllllllllllllll 3640 PRINT "THE ELE"ENTS OF "'" ftATRIX U ARE:" 3650 FOR l=l TO 3660 FOR J=1 TO 3670 PRINT U(I,J), 3680 NEXT J 3690 PRINT 3700 NEXT I 3710 '11111111111111111111 PAINT THE "ATRII UIHERftlTIANI 1111111111111111111 3720 PRINT "THE ELE"ENTS OF "'" "ATAII UIHER"ITIANI ARE:" 3730 FOR 1=1 TO 3740 FOR J=l TO 3750 PRINT UH(I,Jl, 3760 NEXT J 3770 PRINT 3780 NEXT I 3790 PRINT THE KATAIX AS llltlallllllllllaaaalllllllll 3800 PRINT "THE ELEftENTS OF KIN ftATRIX AS ARE:" 3810 FOR 1=1 TO K 3820 FOR J=1 TO N 3830 PRINT ASII,Jl, 3840 NEXT J 3850 PRINT 3860 NEXT I 3870 '11111111111111111111 KATRII VIHER"ITIAN) tllllllllllllllllllltltllllltl 3880 FDA TO N 3890 FOR J=l TO N 3900 YHI!,J)=YIJ,Il 3910 NEIT J l920 NEXT I 93

PAGE 105

3930 'lllllllltlllllllllll PRINT THE IIATRIX V lllllllltllitlllllllllllllltll 3940 PRINT 'THE ELEIIENTS OF NIN IIATRII Y ARE:" 3950 FOR 1=1 TO N 3960 FOR J=1 TO N 3970 PRINT V(I,JI, 3990 NEIT J 3990 PRINT 4000 NEXT I 4010 'llllttllllllllllllll PRINT THE ftATRII VIHERftiTIANI 1111111111111111111 4020 PRINT 'THE ELEftENTS OF NIN IIATRII V(HERIIITIANJ ARE:' 4030 FOR 1=1 TO N 4040 FOR J=1 TO N 4050 PRINT VH(I,JI, 4060 NEIT J 4070 PRINT 4090 NEXT I 4090 'llllllllllltllllllllllllllll IIATRII Sft llllllllllllllllllllltllllllll 4100 FOR 1=1 TO ft 4110 FOR J=1 TO N 4120 SIIII,J I =0 4130 IF I=J THEN SII(I,JJ=S(I) 4140 NEIT J NEXT I 4160 '11111111111111111111 PRINT THE 511 11111111111111111111181111111 4170 PRINT "THE ELEIIENTS OF ftiN ftATRIX Sll ARE:" 4190 FOR I=1 TO II 4190 FOR J=1 TO N 4200 PRINT SII(I,JI, 4210 NEXT J 4220 PRINT 4230 NEXT I 4240 '181111181811111111111111111 IIATRIX 1511 1111111111111111111111111811111 4250 FOR 1=1 TO N 4260 FOR J=1 TO II 4270 1511(1,JI=O 4290 IF I=J AND 5(11> .0001 THEN IS111l,JJ=1/S(II 4290 NEXT J 4300 NEXT I 4310 IIIIUUIIIUIIUIU PRINT THE ftATRII 1511 UIUUUIIIIIUIUUIIUIUI 4320 PRINT 'THE ELEftENTS OF Nlll IIATRIX 1511 ARE:' 4330 FOR I=1 TO N 4340 FOR J=l TO II 43SO PRINT ISft(I,JI, 4360 NEXT J 4370 PRINT 4380 NEXT I 4390 Dill VIIN,II), YIU(N,III, US(II,N), USYIII,NJ,W(II,11 4400 'Dlft VI(S,Sl, VIU(S,Sl, US(S,Sl, USV(S,Sl,W(S,ll 4410 PRINT'USE SINGULAR VALUE DECOftPOSITION TO SOLVE:' 94

PAGE 106

4420 PRINT "RIULI I II(LI1) = Plllll IIATRIX EQUATION" 4430 PRINT "LET R(LIL)=AS(LIL) I P(Lill = Y!Lil)" 4440 PRINT"USE SINGULAR VALUE DECOIIPOSITION TO SOLVE:" 4450 PRINT "AS(LIL) I ll(llll = Y(Lil) IIATRIX EQUATION" 4460 PRINT AS=UI SftiV(H) 4470 PRINT U(H)IAStY=SII 4480 PRINT II=YfiSIIIU(HlfY 4490 PRINT U-111111 UIHERIIITIAN)-11111" 4500 PRINT ASIIIN" 4510 PRINT Y-NIN, YIHEAIIITIAN)-NIN" 4520 PRINT I Sll-IUN, ISII -NIII" 45JO FOR I=l TO N 4540 FOR J=l TO II 4550 YIII,J)=O 4560 FOR K=l TO N 4570 Vl(l,J)=YI(I,J)+Y(l,K)IISIIIK,Jl 4580 NEXT r 4590 NEXT J 4600 NEXT I 4ol0 FOR 1=1 TO N 4620 FOR J=l TO II 4630 YIU(I,J )=0 4640 FOR K=1 TO II 4650 YIUII,Jl=VIUII,J)+VI(l,KIIUH(K,Jl 4660 NEXT k 4670 NEXT J 4690 NEXT I 4690 FOR 1=1 TO N 4700 FOR J=l TO 1 4710 ltii,JI=O 4720 FOR K=l TO II 4730 llll,J)=II(l,J)+VIUII,K)IYIK,JI 4740 NEXT K NEll J 4760 NEXT I 4770 'lllfflllllllllllfllf PAINT THE IIATRII Yf!SIIIUIHI 1111111111111111111 4780 PAINT "THE ELEIIENTS OF Nlll IIATAII YIISIIIU(H) ARE:" 4790 FOR 1=1 TO N 4900 FOR J=1 TO II 4810 PRINT YIU!I,J), 4820 NEXT. J 4830 PRINT 4840 NEJT I 4850 'IIIIIUUUIIU CHECK AS =UI SillY IUIIUUIIIUUIIIIUUUIIUUUI 4860 PRINT CHECK AS =UI SillY 4870 FOR 1=1 TO II 4890 FOR J=l TO N 4890 US( I,J l=O 4900 FOR K=l TO II 95

PAGE 107

4910 USII,Jl=US(I,Jl+UII,llf Sft(K,Jl 4920 NEll K 4930 NEXT J 4940 NUT I 4950 FOR 1=1 TO ft FOR J=1 TO N 4970 USVII,J)=O 4980 FOR K=1 TO N 4990 USV(I,Jl=USVII,J)+US(I,KIIYHIK,JI 5000 NEll K 5010 NEXT J 5020 NEXT I SOlO aalllllllltlllllllll PRINT THE ftATAII AS=UI SftiY(H) 11111111111111111111 5040 PRINT "THE ELEftENTS OF ftiN ftATRII AS=UI SftiYIHJ ARE: 5050 FOR I=l TO 50&0 FOR J=l TO N 5070 PRINT USY(I,JJ, 5080 NEll J 5090 PAINT 5100 NEXT I 5110 PRINT COftPUTE N=YIISfttU(H)IY 5120 FOR I=1 TO N 5110 W(l ;1)=0 5140. FDA K=1 TOft 5150 W(l,ll= W!I,ll+YIUII,Kll Y(K,ll 51&0 NEIT K 5170 NEIT I 5180 LPRINT TAB!Ol;"NUftBER OF DATA POINTS= ";ftftft 5190 LPRINT TABIOJ;"THE ACTUAL YALUES";TA8(20l;"THE ESTI"ATES" 5200 FOR 1=1 TO N 5210 LPRINT TAB!OJ;6(1-1J;TABI20l;ii(I,ll 5220 NEXT I 5230 END 96

PAGE 108

10 'lllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllll 20 'lllllllllllltltfllf RECURSIVE lEAST SQUARES tllllllltltllltllllll 30 ltllllllllllltlllllllllllllllllllllllllllllllllll KEY OFF:CLS 50 'DEFDBl J,M,G,R,D,C 60 'liN-REAl INPUT SAftPlE AT TlftE K 70 'YDUT-AEAl OUTPUT SAftPlE AT TlftE K 80 'D-DESIRED OUTPUT SAftPLE AT TlftE K 90 'RHO-REAl EXPONENTIAL FORGETTING FACTOR 100 'M-ARRAY CONTAINING MEI&HT VECTOR AT TlftE K 110 'NORD-NURSER OF TAPS IN THE FILTER 120 'ITDL-ARRAY FOR STORAGE OF PAST DATA SAftPLES 130 'IPTR-POINTER FOR DELAY LINE ARRAY 140 'RINV-REAL ARRAY FOR STORAGE OF INVERSE CORRELATION ftATRil 150 60SU8 250 : RE" DI"ENSIONS AND INITIALIZATION 160 PRINT PLOT ESTiftSTES OR RESIDUALS DR OUTPUT E/R/0 ' 170 INPUT AS. 180 IF AS="O" SOTO 230 190 IF AS="E" SOTO 210 200 IF AS="R" SOTO 220 210 GOSUB 1430 : SOTO 230 220 GOSUB 1780 230 SOSUB 430 : REft RECURSIVE LEAST SOUARES ALGORITHft END 250 REft llltlltlttll DlftENSIONS AND INITIALIZATION ttlttlltllllltllll 260 11=0 : 12=0 270 ft=956 :C=S :l=S :16PTR=L :ITER=O :EO=O :RH0=.99 280 Dift Xlfti,ITDLIL+11,Nil+11,6(25 l,Z(l+l),RINV((Lt11A21,Difti,C!fti,Y1(L+11 290 FOR 1=0 TO l 300 XTDL (I 1=0 310 N!II=O : Y1(11=0 320 RINV(II(l+11+11=5000 no NEXT I FOR 12 = o TO l J50 'RP(0112J=O : EP!O,I21 = 0 360 NEIT 12 370 6(01= .5 :6(1)= 1 :G(21= -.8 :6(31= .2 :6(4)= J80 6!51= .9 390 LPRINT TAB!OI;"ITER' 400 lPRINT TA8(01;"6( O)'jTA8(141;'6( 11';TA8(281;"6( 2J";TA8(42J; "&( 31"; TAB!561 ;"6( 41";TA8(701 ;"G! 5)" 410 lPRINT TAB(OI; 61 OJ ;TA8(14J; 61 11 ;TA8128J; 61 21 ;TA8142J; 6( 31 ; TA8(56) ; G( 41 ;TA8(701 i 6( 51 420 RETURN 430 REft llllltltltlllf RECURSIVE lEAST SOUARES ALGDRITHft llfllllllflfft 440 'Ill INPUT' DESIRED OUTPUT Ill 450 REft D!K) = G(O)Il(KI+ G(L)ti(K-Ll 460 FOR k=O TO ft-1 97

PAGE 109

470 I!Kl=2t!RND-.5)t5DR!3t.01) 480 NEXT K 490 IIN=X(IGPTR) 500 ACCUII=O 510 LPTA=ISPTR 520 FOR 1=0 TO L 530 ACCUII=ACCUII+SIII&l(LPTRl 540 LPTR=LPTR-1 550 NEXT I 5b0 16PTA=I6PTA+1 570 IF ISPTA -1 = 11-1 GOTD 240 : 11-1-!Ll = I OF ITERATIONS OF RLS 580 D=ACCUII 590 'Ill ESTIIIATE OUTPUT Y(Kl Itt bOO ITOL(IPTR)=l!N blO ACCUII=O &20 LPTR=IPTR 630 FOR 1=0 TO L b40 ACCUII=ACCUII+W(I)IITDL!LPTRI &SO LPTR=LPTR+l bbO IF LPTR >L THEN LPTR=O 670 NEXT I 680 YOUT=ACCUII 690 'Ill COIIPUTE ERROR &It 700 E =D-YOUT 710 'Ill COIIPUTE FILTERED INFORIIATION VECTOR Z ttl 720 FOR 1=0 TO l 730 LPTR=IPTR 740 ACCUII=O 750 FOR J=O TO l 7b0 ACCUII =ACCUII+RINV(JI(L+1)+1)11TDLILPTR) 770 LPTR=LPTR+l 780 IF LPTR >L THEN LPTR=O 790 NEXT J Z!I)=ACCUII 810 NEIT I 820 'til COIIPUTE O=I(TRANSitZ til 830 ACCUII=O 840 LPTR=IPTR 850 FOR 1=0 TO l 860 ACCUII=ACCUII+ITDLILPTR)IZIII 870 LPTR=LPTR+l 880 IF LPTR > L THEN lPTA=O 890 NEll I 900 O=ACCUII 910 aaa COI!PUTE GAIN CONSTANT tat 920 Y=l/IAHO+Q) 930 'Ill UPDATE WEIGHTING COEFFICIENTS ttl 940 FOR 1=0 TO L 950 W(II=W!II+EIVIZ(II 98

PAGE 110

960 NEXT I 970 'Ill UPDATE R INVERSE ftATRIX Ill 980 FOR 1=0 TO L 990 FOR J=I TO L 1000 JPTR= JIIL+1)+I 1010 KPTR= II(L+1)+J 1020 RINV(JPTRI=IRINV(JPTRJ-ZIIIIZ(JJIVI/RHO 1030 RINYIKPTRJ=RINV(JPTRI 1040 NEXT J 1050 NEXT I 1060 IPTR=IPTR-1 1070 IF IPTR < 0 THEN IPTR=L 1080 ITER = ITER+1:60SUB 1290 :ITER = ITER -1 1090 IF AI="E" 6010 1120 1100 IF AS="R" GOTO 1130 1110 ITER=ITER+1 :12=ITER : 60SUB 1150 : 60TO 490 1120 ITER=ITER+1 :12=ITER : &OSUB 1690 : &OTO 490 1130 ITER=ITER+1 :16=1TER : 60SUB 1890 : 60TO 490 1140 REft 111111111111111111 PRINT RESULTS 111111111111111111111111111111 1150 CLS 1160 LPRINT 1170 LPRINT TABIOI; ITER 1180 LPRINT TABIO); Ml OJ ;TABI141; Ml 11 ;TAB12BI; M( 2) ;TA8142); Ml 3) ; TABI56J ; WI 41 ;TA8(70) i M( 5J 1190 LOCATE 1 ,1 :PRINT "6(0 1=";610 ):LOCATE 2 ,!:PRINT Ml 0 I 1200 LOCATE 1 ,25:PRINT "611 1=";6(1 ):LOCATE 2 ,25:PRINT Ml 1 I 1210 LOCATE 1 ,SO:PRINT "6( 21=";61 2J:LOCATE 2 ,50:PRINT WI 2 I 1220 LOCATE 3 ,1 :PRINT "6(3 1=";613 ):LOCATE 4 ,1:PRINT Ml J I 1230 LOCATE 3 ,25:PRINT "614 1=";61 4J:LDCATE 4 ,25:PRINT Nl 4 I 1240 LOCATE 3 ,SO:PRINT "6( 51=";615 ):LOCATE 4 ,50:PRINT MIS I 1250 LOCATE 18,25:PRINT "ITER =";ITER 1260 LOCATE 19,25:PRINT "SUft Of RESIDUALS R : ;R 1270 LOCATE 20,25:PRINT "RftS RESIDUALS = SORIRI = ";SDRIRI 1280 RETURN 1290 REft 11111111&11111111111111 COftPUTE RESIDUALS 111111111111111111111 1300 R=O : 01=0 : U2=l 1310 FOR I=L TO l+ITER 1320 03=02 1330 Dlll=O : Clll=O 1340 FOR J=O TO L 1350 Dill= DIIJ+ 61JI11(03J 1360 Clll= CIIJ+ M(JJIXIOJJ 1370 03=03-1 :NEIT J 1380 02=02+1 1390 1400 NEXT I 1410 R=R/IITER+ll 99

PAGE 111

1420 RETURN 1430 REft lttttlllltllttttltt l-Y PLANE FOR ESTiftATES ltltltttttltllttltt 1440 W=2 1450 CLS : SCREEN 2 1460 FOR Y=O TO 160 STEP 20:LINE (38,Y 1(40,Y I : NEIT Y 1470 FOR Y=39 TO 589 STEP 110:LINE (Y,79J-(Y1811 : NEIT Y 1480 FOR 1=1 TO S :LOCATE 111141I+1 :PRINT (ft-1-LIII/S: NEXT I 1490 LOCATE 1,1 : PRINT W : LOCATE 20,1 :PRINT -M 1500 LOCATE 5,1 : PRINT W/2 : LOCATE 16,1 :PRINT -W/2 1510 LINE (39,0)-(39,160) 1520 LINE (39,80 1,80 1530 FOR 1=0 TO L 1540 X=6(1) 1550 60SUB 1580 1560 NEXT I 1570 RETURN 1580 IF 1<0 60TO 1640 1590 Z= 80 (Xt199)/(M+M+MI39/801 1600 FOR J=l9t1000/600 TO 600 STEP 25 1610 PSETIJ,ZI 1620 NEIT J 1630 &OTO 1680 1640 l=(W+ABS(I))t199/(W+W+Mtl9/80) 1650 FOR K=l911000/600 TO 600 STEP 25 1660 PSETIK,ZI 1670 NEXT I 1680 RETURN 1690 REft tttlttttltttllttllllt PLOT THE ESTiftATES ttttttltllttttttttltlt 1700 VIEW(39,0)-(589,160) 1710 NINDOW(O,N)-(ft-L1-M) 1720 LOCATE 22,70 : PRINT "ITER=";ITER 1730 FOR l=O TO L 1740 LINE (11,Y1(0))-(121M(01) :LINE. IIl,Y1(1))-(12,Milll 1750 NEIT I 1760 FOR 1=0 TO L : Y1(l)=Nill : NEIT I : X1=12 1770 RETURN 1780 REft ltttttttttttlttttt X-V PLANE FOR RESIDUALS ttttllttltlltltltt 1790 CLS : SCREEN 2 1800 LINE 1119,8 )-(119,136 I 18IO LINE (1I9,136 1-15201136 I 1820 FOR Y=8 TO 136 STEP 8 :LINE 111B,Y )-(120,Y I : NEIT Y 1830 FOR Y=119 TO 520 STEP (520-119)/IC-11 :LINE (Y,1l7 )(Y1135): NEIT Y 1840 LOCATE I9,I5:PRINT !:LOCATE 19,65: PRINT ft-1-L 1850 A= -9 :8= -1 :NIN=10A(A) : ftAI= 10AIBI:K=IABSIAJ-ABS(8JJ/16 1860 FOR I=O TO 16 STEP 4 :LOCATE 1+2,1 :H=III+1 :PRINT "10E-"(HJ :NEXT I 1870 FOR 1=0 TO 16 STEP 4 :LOCATE 1+2,11 :H=Itl+l :PRINT -(HI :NEXT I 1880 RETURN 1890 REft tltttttttllltltttltlltt PLOT RESIDUALS ttllllllttttllttttlttlt 1900 A= -9 :8= -1 100

PAGE 112

1910 'R=L061Alll06110l 1920 LOCATE 20,3 : PRINT 'R':LOCATE 20,9:PRINT "L06(AI' 1930 LOCATE 20,20: PAINT 'SU" OF RESIDUALS A = ;R 1940 LOCATE 21,20: PRINT 'R"S OF R = ';SOR(RI 1950 LOCATE 22,20: PAINT'LOS!R) = '; L06!AJ/L06(101 19b0 LOCATE 23,20: PRINT'ITER = ';ITER 1970 Ab=L06(Al/L06(101 1980 VIEW1119,B )-(520,136) 1990 WINDOW !1, B )-("-1-L, A I 2000 IF 15=0 6010 2020 2010 LINE(151R5)-(Ib,Abl 2020 15=1b : RS=Rb 2030 RETURN 101

PAGE 113

10 20 '1111111111111111111111 EXHAUSTfYE SEARCH AL60RITH" 11111111111111 30 40 KEY OFF:CLS SO 'DEFDBL X,W,6,R,C,D 60 60SUB 180 : REft DlftENSIONS AND INITIALIZATION 70 PRINT"PLDT ESTI"ATES OR RESIDUALS OUTPUT E/R/0 SO INPUT AS 90 IF At="O" SOTO 130 100 IF AS="E" SOTO 120 110 GOSUS 1310:6010 130 120 GDSUB 940 130 GOTO 160 140 GOSUB 460 : AE" EXHAUSTIVE SEARCH 150 END 160 GOSUB 880 : REft Ill GENERATED INPUTS AND OUTPUTS FOR ES 170 SOTO 140 180 HEft 111111111111 DiftENSIONS AND INITIALIZATION lllltllllllllllll 190 Xl=O : 12=0 :15=0 :R5=0 200 "=15 :C=50 :L=S :ITER=O :'1 OF DATA USED= "-L 210 Olft lift), WIL+l),6125 )1Difti,CI"I,Y1(l+ll 220 FOR 1=0 TO l 230 Wlll=O :Y1(11=0 240 NEIT I 250 6(0)= .s :6111= 1 :6(2)= -.8 :6(3)= .2 :6141= -.4 260 6(5)= .9 270 LPRINT TA8(30J;"ES" ;TA8142);"1 OF DATA = 10" 280 LPRINT TABIOI;"ITER";TA8(14);"SU" OF RESIDUALS R";TA8(421;"RftS OF R" ;TABI 56l;"STEPSIZE = .01" 290 LPRINT TABIOI;"SIOI";TA8(141;"6(1)" ;TABI2Bl;"GI21";TAB(421;"6131";TABI56) ;"614l";TA81701;"6151" 300 LPRINT TABI 01; 6(0) ;TABI14J; 6111 ;TA8128l; 6121 ;TA81421; 613) ;TABI56 I; 6141 ;TABI70l; 6(5) 310 RETURN 320 RE" lllllllttltlllllll PRINT RESULTS 1111111111111111111111111111 330 CLS 340 LPRINT TA8101; ITER ;TAB!l41; R ;TA8(421; SORIA) 350 LPRINT TAB( 01; WIOl ;TA8(14J; Will ;TABI281; Wl2l ;TA81421; Will ;TABI56 I; W(4) ;TA81701; WISI 360 LOCATE 5 ,1 :PRINT "610)=";610l:LOCATE 6,l:PRINT WIOl 370 LOCATE 5 ,2S:PRINT "6(11=";611l:LOCATE 6,2S:PRINT Will JBO LOCATE 5 ,SO:PRINT "6121=";612J:LOCATE 6150:PRINT Wl21 390 LOCATE 7 ,1 :PRINT "6(l1=";61ll:LOCATE 811 :PRINT Will 400 LOCATE 7 12S:PRINT "6(41=";6(41:LOCATE 8,25:PRINT W(4) 410 LOCATE 7 ,SO:PRINT "6151=";61SI:LOCATE B,SO:PRINT Wl51 420 LOCATE 18,25:PRINT "ITER =";ITER 430 LOCATE 19,25:PRINT "SUft Of RESIDUAlS R=" ;R 440 LOCATE 20,2S:PRINT "RftS RESIDUALS = SORIRI = "; SOAIRI 450 RETURN 102

PAGE 114

460 RE" ltlltlllllllltlltt EXHAUSTIVE SEARCH tllllllllllllltllllllllll 470 60SUB 740 480 Z=2 : 'Z=INT((ITERI/2)+1 490 IF Z<=15 THEN 60TO 510 500 SL=lOA(-151 : 60TO 520 510 SL=lOA(-ZI 520 FOR N=O TO L 530 60SUB 620 540 NEXT 550 IF AS="O" 60TO 590 560 IF AS="R" 60TO 580 570 ITER=JTER+l : X2=ITER : &OSUB 1200 : GOTO 600 580 ITER=ITER+l : 16=iTER : 60SU8 1430 : GOTO 600 590 ITER=ITER+l : 6DSUB 330 bOO IF ITER= C THEN 60TD 150 :C=IITERATIONS OF EIHAUSTIVE SEARCH b10 &DTO 480 620 REft lllfllllltlllllll SUB PARA"ETERS SEARCH lllllllllllllllltlltll 630 RE" PARA"ETER SEARCH 640 Rl=R:W1=WINI b50 W(NI=W(NJ+Sl:60SUB 740:R2=R:W2=WINI 660 IF R2>Rl THEN SL=-Sl:A=R2:R2=R1&Rl=A:A=W2:W2=W1:W1=A:W(NJ=W(NI+SL 670 W(NI=W(NJ+SL:SOSUB 740:R3=R:W3=W(NI 680 PRINT "Rl=";Rl,"R2=";R2,"R3=";Rl 690 IF R1>R2 AND R2>R3 THEN R1=R2 :R2=Rl:W1=W2:M2=WJ:60TO 670 700 W(N)=,51((Wl+W2111Rl-R2)-(W2+WJ)I(R2-Rlll/1Rl-21R2+Rll 710 PRINT Z,G(NI,M(N),Sl 720 GOSU8 740 730 RETURN 740 REft ltltltllfllllllll CO"PUTE RESIDUALS lllltlttttllllllllllllllll 750 R=O :IPTR=O : 16PTR=L 760 FOR I=L TO "-1 770 LPTR=I6PTA 780 Dlfl=O : C(ll=O 790 FOR J=O TO L BOO Dill= Dill+ &IJIIIILPTA) 810 C(fl= Clll+ W(J)IIILPTAI 820 LPTR=LPTR-1 :NEIT J BlO 16PTA=I6PTR+1 840 R=R+IDIII-C(I))A2 850 NEXT I 860 R=R/(" -LI 870 RETURN 880 'llllltttlttltlllllltlllll INPUTS FOR ES lllllllllllllllttlltttllt 890 RE" Dlkl = G(OIIIIkl+.,, &lllii(K-Ll 900 FOR K=O TO "-1 910 I(K1=21(RND-.5)tSUR(li.Oll 920 NEIT K 930 RETURN 940 RE" ttttttltltltttt 1-Y PLANE FOR ESTI"ATES lltltlttlttlltltttttll 103

PAGE 115

950 M=1 960 CLS : SCREEN 2 970 FOR Y=O TO 160 STEP 20:LINE (38,Y J-(40,Y I : NEIT Y 980 FOR Y=39 TO 589 STEP 1589-391/C :LINE IY,79J-(Y,811 : NEIT Y 990 LOCATE 11,73 :PRINT C 1000 LOCATE 11,1 : PRINT 0 :LOCATE 1,1 : PRINT M : LOCATE 20,1 :PRINT -M 1010 LOCATE 5,1 : PRINT M/2 : LOCATE 16,1 :PRINT -M/2 1020 LINE 139,0)-(39,160) 1030 LINE (39,80 1-1589180 1040 FOR 1=0 TO L 1050 1=6111 1060 GOSUB 1090 1070 NEll I lOBO RETURN 1090 IF 1<0 GOTO 1150 1100 Z= 80 (lt199)/(M+M+Mtl9/80) 1110 FOR J=39t1000/600 TO 600 STEP 25 1120 PSET(J,ZI 1130 NEXT J 1140 GOTO 1190 1150 Z=IM+ABSilllt199/(M+M+M139/801 1160 FOR K=3911000/600 TO 600 STEP 25 1170 PSET(K,Zl 1180 NEXT K 1190 RETURN 1200 RE" tltltltllllllllllllt PLOT ESTI"ATES 11111111111111111111111111 1210 M=l 1220 YIEM139,0)-(589,160) 1230 MINDOM(O,MI-IC,-Ml 1240 LOCATE 22,70 : PRINT "ITER=";ITER 1250 FOR 1=0 TO L 1260 LINE ll1,Y110))-(l2,W(Oll :LINE 111,Y11111-(l2,MIIIl 1270 NEIT I 1280 FOR 1=0 TO L : Y1(1l=M(Il : NEIT I : 11=12 1290 RETURN 1300 &OSUB 1310 ll10 REft llllllttlttlltlll 1-Y PLANE FOR RESIDUALS 1111111111111111111 1320 CLS : SCREEN 2 1330 LINE 1119,8 1-1119,136 I 1340 LINE 1119,136 J-(520,136 1350 FOR Y=B TO 136 STEP 8 :LINE 1118,Y I-1120,Y I : NEXT Y 1360 FOR Y=119 TO 520 STEP 1520-119)/(C-11 :LINE (Y,137 )-(Y11351: NEIT Y 1370 LOCATE 19,15: PRINT 1 1180 LOCATE 19,65: PRINT C 1390 A= -21 :8= -1 :ftiN=10A(AI : ftAI= lOA(B):l=(ABS(AI-ABSIBII/16 1400 FOR 1=0 TO 16 STEP 4 :LOCATE 1+2,1 :H=KII+l :PRINT "10E-"IHI :NEXT I 1410 FOR 1=0 TO 16 STEP 4 :LOCATE 1+2,11 :H=KII+l :PRINT -(H) :NEXT I 1420 RETURN 1430 REn 11111111111111111111111 PLOT RESIDUALS llttlltllllllllllllltl 1M

PAGE 116

1440 A=-21 : B=-1 1450 LOCATE 2013 :PRINT ;A' :LOCATE 20,9 :PRINT 'LOG!RI' 1460 LOCATE 20,20:PRINT 'SU" OF RESIDUALS'R =';R 1470 LOCATE 21120:PRINT 'R"S OF R = SQR(RI = ';SQR(Rl 1480 LOCATE 22,20: PRINT 'L06(Rl ='; iLD61Rl/L06(101l 1490 LOCATE 23,20:PRINT 'ITER=';ITER 1500 R6=L06(R)/L06(10) 1510 VIEW(l19,B l-1520,136) 1520 WINDOW 11, A l-IC, B l 1530 IF XS=O GOTO 1SSO 1540 LINEIXS,R5)-(16,R6) 1550 15=16 : R5=R6 1560 RETURN 105



PAGE 1

PHYSICIANS' OPINIONS ABOUT PATIENT INVOLVEMENT IN I HEALTH AND MEDICI CARE DECISIONS AND TELEPHONE-BASED DECISION SUPPORT Jerry Wjyne Howell I B.A, University of Northern Colorado, 1977 MS., University of at Boulder, 1986 A thesis submitted to the I University of Golorado at Denver I in partial fulfillment of the requiremets for the degree of I Doctor of Philosophy Health and Behavioral Sciences 1J 9 .. B}.

PAGE 2

1999 by Jerry Wayne Howell All right reserved.

PAGE 3

This thesis for the Doctor of Philosophy degrjby Jerry Wayne Howell has been approved by Deborah S. Main Wendyhl.lynch Date

PAGE 4

Howell, Jerry Wayne (Ph.D., Health and Behavioral Sciences) Physicians' Opinions About Patient InJolvement in Health and Medical Care Decisions and Telephone-Based Decitn Support Thesis directed by Professor Deborah S. Main ABSTRACT Survey research (n = 914) and key interviews (n = 21) were used to assess physicians' opinions about: I) cobsumers/patients being more informed and taking a more active partnership role in Ihedical care decisions, 2) the incorporation of telephone-based decision support (TBDS) into care management, and 3) the effect a more active consumer/patient partnership and TBDS will have on health outcomes, patient satisfaction, doctor-patient clinical efPciencies, utilization, costs, and malpractice liability. The percentage of physicians agreeing that patients want more information about health and medibl issues and more active involvement in medical decision making was 900/o and s;s%, respectively. The percentage of physicians agreeing that physician and patient together should make decisions about minor and major medical conditions wasi56% and 68%, respectively, with 66% agreeing that they would like their patients to take a greater role in self-managing chronic conditions. Physicians believed that a more informed patient would result in better medical decisions leading to imprJved health outcomes, patient satisfaction, doctor-patient relationships, utilization efficiencies, and a potential reduction in medical liability. However, physicians in their preference for involving patients in decision making and acknowledged that patients have varying preferences for involvement. They believed that the knost prudent clinical approach is for the physician to be sensitive to each indivicWal patient's preference for information and involvement. Opinions varied regarding !the purposes, potential impacts, and reasons for why health plans had implemented TBDS services. Logistical regression analyses showed that primary care physicians werb more likely than referral specialists to: agree that their patients should take a grJater role in treating minor conditions; agree that their patients should take a greater rJte in self-managing chronic conditions; be knowledgeable about TBDS; have a positive opinion about TBDS; believe that iv

PAGE 5

the health plans implemented TBDS to respond to interests of the business community, to more efficiently care for capitated patients, improve quality of care, and member satisfaction; believe that would decrease the number of medically inappropriate visits and unnecessary office calls. Explanations for these findings, implications for working wit 1 physicians, and recommendations for future research are discussed. This abstract accurately represents the content of the candidate's thesis. I recommend its publication. v

PAGE 6

DEDI
PAGE 7

Dr. Main for her unwavering '-'V1Ji1.1U'!;i11'-" in me, her boundless enthusiasm, her sage leadership, mentoring and in directing this thesis. Dr. Main, Dr. Iverson, Dr. Dr. Lynch, and Dr. Stulp for working together as a multidisciplinary in sharing their wealth of knowledge and experience to create and nurture 1 learning environment that should characterize all dissertation eXJ>eliJen<:es. Joe Mc<;Jloin for his patience and cal assistance in data management and statistical analysis. Lu Sandoval for her tenacity and in transcribing the key informant interviews. Sherry Holcomb for her and attention to detail in survey administration and data cotllectlon. David Nash, M.D., Zach M.D. and Mark Earnest, M.D., for their critical review of the qualitative Barb and Don Day for providing "home away from home." Pepper Hatch for his leadership "flying high cover" and providing consistent support and 1 to complete this work. All the physician subjects who l!fl!lCI
PAGE 8

co CHAPTER 1. IN'I'R.ODUCTION...... . . . . . . . . . . . . . . . . . . . . . . . . . . 1 Specific Aims ......... l .............................................. 2 I Background and ..................................... 3 Overview ofResearcH Design and Setting ....................... 8 I Study Design ..................................................... 9 Study Site .......... j ............................................ 11 I Telephone-Based Decision Support Services .............. 12 Key Questions ..... j ............................................ 16 I Reconciling Quantitative and Qualitative Analysis ............ J ............................................ 18 Organization of the DJssertation ................................. 18 2. REVIEW OF LITERATURE ........................................... 20 Geographical Variatio sand Appropriate Use ofHealth Care .... 1 ............................................ 20 Demand-Side lnlerviions ...................................... 24 Perceived Need and Self-Care .............................. 24 Chronic Disease MLgement ............................... 27 Patient Preference id Shared Decisions ................... 29 V111

PAGE 9

Patient Preference for Information and Involvement in Decision Making ............................................... 32 Physician Perception lbout Patient Desire for Infonnation and Involvement .................................... 52 Doctor-Patient ComJunication and Relationships: Challenges/Barriers abd Effects on Outcomes . . . . . . . 54 Doctor-Patient coLunications: Challenges and Barriers ........ l ............................................ 55 ............. 65 Models for Physician-Patient Communication and Relationships. J ............................................ 71 Diffusion of Innovatijns: Constructs, Components, and Determinants ..... 11 ............................................ 79 Components ofDitsion ..................................... 79 Adoption Rate Determinants ................................. 82 Diffusion oflnnovltion in Health Care ..................... 85 Summary and ImplicJions for Research ..................... 117 3. QUANTITATIVE METHODS AND RESULTS ................... 119 Introduction ............ j . . . . . . . . . . . . . . . . . . . . 119 Methods ........................................................... 120 Subjects ........................................................ 120 I Physician Opinion Survey (POS) .......................... 120 lX

PAGE 10

POS Administration and Data Collection Protocol ............ l ........................................... 122 Analysis and Statistill Methods .............................. 123 Sample Descripti+ ......................................... 123 Description of Variables ..................................... 123 Selection of StatiJcal Tests ................................ 126 Results ............................................................. 129 Sample Description .......................................... 130 I Description of Variables ..................................... 132 Logistic Regressiot ......................................... 133 4. QUALITATIVE METHODS AND RESULTS ..................... 153 Introduction ............ l .......................................... 153 Methods ................ J .......................................... 154 Sample Selection .. 11 .......................................... 154 Physician Opinion Interview (POI) Guide ................ 156 Interview Structure and Administration ................... 157 Analyses ........................................................... 158 Transcription, Cod1g, Segmentation. . . . . . . . . . . 159 I Data Analysis and
PAGE 11

Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 163 Subjects ........................................................ 163 Opinions About Patient Involvement in Medical ............................................ 165 Opinions About Tilephone-Based Decision Support .l ........................................... 214 Closing Questions ........................................... 242 Synopsis: Overarching Themes ............................... 246 Theme 1: PhysiciaJs Support Idea of More Informed Adive Patients... . . . . . . . . . . . . 246 I Theme 2: Improved Health Outcomes: Chronic and Major Dlness j...... . . . . . . . . . . . . . . . . .. 247 Theme 3: Improved Patient Satisfaction (Patient Preference j. . . . . . . . . . . . . . . . . . . . 249 Theme 4: Doctor-Patient Relationship (Patient Preference)......................................... 250 Theme 5: Clinical Efficiencies for Minor and Chronic Illnesses....................................... 252 I Theme 6: Reduced Health Care Costs for Minor and Chronic Condi,.ms...... ... ......... ... ...... ... ...... 255 Theme 7: Reduced Potential for Medical Liability.. . . . . . . . . . . . . . . . . . . . . . . . . . . 256 Theme 8: Self-Efficacy Improved with More Informed Active Pahent .................................... 257 I Summary. Themes 1-8 ..................................... 257 X1

PAGE 12

Theme 9: Variance in Physician and Patient Preferences ...................................................... 258 Theme 10: TBDS -Uncertain or No Impact on Physicians' Practices and Patients .............................. 261 Theme 11: TBDS -Uack of Support by Practicing Physicians Will Diminish Its Value ............... 262 5. DISCUSSION ............................................................ 266 Key Quantitative and Qualitative Results .................... 266 Physicians' Opinions about Shared Decision Making ............. l ........................................... 267 Physicians' Opinijns about TelephoneBased Decision .................................... 277 Commentary on Jplications for Working with Physician Organizations ............................... 291 Limitations ........................................................ 304 Recommendations for Future Research ....................... 314 Conclusions ....................................................... 319 APPENDIX A PHYSICIAN OPINION SURVEY (POS) ........................... 321 I B. PHYSICIAN OPINION INTEVIEW (POI) ........................ 328 Xll

PAGE 13

C. QUALITATIVE ANAL VIIS CODEBOOK ........................ 334 D. VARIABLES WITH MISSING VALUES .......................... 340 REFERENCES ..................................................................... 343 xiii

PAGE 14

CHAPTER 1 I INTRODUCTION During the last decade ofU.S.1io-cultural history, the consumer has played an increasingly greater role in shaping the marketplace. The baby-boomer generation that has shaped virtually every other sectlr of the U.S. economy is emerging as the "new consumer'' for the healthcare markltplace as well. This new consumer is generally more informed; more derpandink about access, convenience, and service quality; generally more assertive in the d,ctor -patient relationship. In essence, they want to be more self-sufficient and independent. For this new consumer to be effective in shaping the healthcare they will need to technological tools and processes that can provide infoLtion and that can improve their ffi . h d I .lhi e ect1veness m t e octor-pat1ent re at10ns p. Consequently, this new era of conLerism in healthcare has stimulated an interest in the concept of demand manageLent and has spawned a rapidly growing increase in consumer health management id decision support services. Demand management has been defined as the use decision and self-management support systems to enable and encourage consumJs to make appropriate use of medical care (Vickery & Lynch, 1995). One of the conLmer health management services that has emerged has been telephone-based decisiol support (TBDS). Through a 1-800 1

PAGE 15

number, such services provide 24-hour per day access to specially trained nurse counselors. Using on-line decision supplrt protocols or algorithms, the nurse counselors provide decision support, selftgement support, psychosocial support, education and infonnation to consumers/patients on virtually any health care topic. The market potential for these types of sekces has been estimated to be somewhere I between 100-200 million lives (Equitable Securities Research, 1996). Even though the opinion of the may stands as a critical variable in determining the ultimate impact of more Jronned, actively involved patients and mos, reviews of published literature revLed no systematic assessment of physicians' viewpoints about these issues. This research will advance our understanding about how physicians view enhanced participation of patients in medical I decision-making and the integration of TBDS services into care management. This infonnation will provide key insights on thl implications for working with physicians in optimizing the integration of these comp\onents into the delivery of health care. Specific Aims A multi-method approach was used to adJess four central issues: The specific aims of this research were: 2

PAGE 16

Aim #1: To ascertain primary care and specialty physicians' opinions about more informed, actively involved patienls as partners in health and medical care decisions. Aim #2: To assess primary care and specialty physicians' opinions about the impact they believe consumers/patients bling more informed and taking a more active partnership role in health and medicali decisions will have on: the quality ofthe doctor-patient relationship, clinical efficiencies, health outcomes, patient satisfaction, health care utilization, health care costs, L malpractice liability. Aim# 3: To assess primary care Ld specialty physicians' opinions about the incorporation of TBDS services into care management. Aim# 4: To assess primary care and specialty physicians' opinions about the impact they believe a TBDS service will hlve on: the quality of the doctor-patient relationship, clinical efficiencies, health otmes, patient satisfuction, health care utilization, health care costs, and malpractice liability. Background and Significance Proposals and strategies fur reform (generally cost contairunent) have been derived chiefly from a supply-si(ie economic paradigm. Supply-side I strategies have historically emphasized global budgets, consumer tax incentives, cost\ shifting to consumers, capitated payments for physicians, technology management, 3

PAGE 17

hospital bed capacity control, use of gatekeepers, utilization review, and other forms of "managed care." In contrast, little eLphasis has been given to integrating the I biological sciences with demand-side social, cultural, psychological, and health risk factors which influence health care decisiln making, health care behaviors, and health I care utilization that ultimately impact health outcomes, patient satisfaction and health care costs. Recently, however, the concept oi "demand management" has received significant national interest with the incorporation of demand-side strategies into managed care delivery systems. Demand Lanagement has been defined as the use of decision and self-management support syslems to enable and encourage consumers to I make appropriate use of medical care (Vickery & Lynch, 1995). I The supposition behind demand-side strategies is that health care decisions and health care seeking behaviors are influenc1 by a complex set of psychological, social, and culturallilctors which are more relatJ to perceived need and patient preference rather than to morbidity (Vickery & LyncJ 1995). It appears that the access an individual has to infonnation and resourceJ influences perceived need and patient preference for services which in tum influelces behaviors associated with health and I medical decisions (Berkanovic Telesky, & Reeder, 1981; Bradley, Zia, & Hamilton, I 1996; Dentzer, 1991; Greenfield, Kaplan, W\ are, Yano & Frank, 1988; Levine & Britten, 1973; Lorig, Mazonson & Holm 1993; Mayo, Richman & Harris, 1990; 4

PAGE 18

Vickery, Kalmer, Lowry, Constantine, Wright & Loren, 1983). That is, a well-infonned patient will make ruJ1 nt decisions than a less-informed patient and will subsequently utilize health care resources differently. I This recent interest in demand-side strategies has subsequently spawned a rapidly growing increase in consumer he1th management and decision support I services. One such service has been TBJS. Through a 1-800 number, such services provide per day access to specially trained nurse counselors. With the aid of sophisticated on-line decision support pJtocols or algorithms, the nurse counselors provide decision support, self managemelt support, information, and psychosocial I fc ().I .. d.. d" support to consumers patients or: a tnflge or pre-VISit ects10ns regar mg acute minor illnesses and injuries (e.g., infectiol, headache, back pain, etc.); (b) shared d h h a1 h d I h d d. ectsions mt e t care proVl ers concermng maJor t erapeutic an Iagnosttc procedures (e.g., surgery, hospitalization, chemotherapy); c) chronic disease management (e.g., diabetes, asthma, hypertension); and d) lifestyle management (e.g. .gh lini. I a1 . ) exercise, stress, wei t management, c c preventive sernces There are over 60 million Amerijs enrolled in some form of managed care. Capitation has increased in prevalence as 1echanism for reimbursement. For managed I care to be ultimately successful, however, it must mature beyond strictly a mechanism for cost containment to one focused on mLng and managing value. Market dynamics suggest that employers and goveLent as the ultimate purchasers of care 5

PAGE 19

and patients as the ultimate consumers of care are looking beyond just cost as a basis by which they will select health care seJces. Even though cost will remain a primary consideration, they are looking to other ijdicators of value and quality such as patient satisfilction, access, convenience, and outcomes. With the advent of a more informed and assertive consumer, the health care industry is ripe for consumer health and dlcision support services. The market I potential for TBDS has been estimated to be at least I 00 million lives and as many as 200 million (Equitable Securities ResearcJ 1996). However, despite the recent growth of TBDS services, to date we knor very little about their impact on the quality of the doctor-patient relationship, clinical efficiencies, health outcomes, patient satisfaction, health care utilization, health jare costs, and malpractice liability. As these types of services gain visibility and bLme more widely incorporated into health care delivery systems as a "demand-side" stategy, an increasingly sophisticated American population of consumers, patientl, physicians, purchasers of medical care I (government and business), and insurers will pose questions about the medical efficacy and efficiency of such services. \ Physicians, in particular, are in clinically appropriate questions. Do consumers/patients want more information about health and medical care decisions? I Do consumers/patients want increased involvement in health and medical care decisions? Will TBDS services improve the\ quantity and quality of information 6

PAGE 20

desired by consumers/patients? If so, what impact will such services have on: the quality of the doctor-patient clinical efficiencies, health outcomes, patient satislilction, health care utilizatiot health care costs, and malpractice liability? Consumers and patients are desiring more infonnation related to health and medical care decisions. In 1988, GreenfiJld, Kaplan, Ware, Yano, and Frank prophetically stated, "We are entering an era of unprecedented sophistication and knowledge among American patients. . In the everyday practice of medicine, will 1 fh .\.aldi will" kind mvo vement o t e patient represent a cntlc sruptlon or 1t promote a new of relationship that enhances the roles of lhysician and patient in generating improved I outcomes" (p. 457)? Anecdotal reports TBDS decision support services have indicated that they support the doctor-patient relationship, enhance patient satisfaction and save money without sacrificing qualiJ of care (Barnett, 1995). Even though the opinion of the stands as a critical variable in detennining the ultimate impact such decilon support interventions will have on physician-patient relationships, clinical ,lencies, patient outcomes, medical care utilization, and costs, reviews of literature revealed no systematic assessment of physicians' viewpoints about these types of services. Nor has anyone tried to ascertain the implications of their opinions on the implementation, utilization, and ultimate impact of such services on health care quality, cost, and access factors. 7

PAGE 21

Consumer health management s pport services, including TBDS, have experienced rapid growth (Barnett, 199b; Equitable Securities Research, 1996). However, there is a dearth of informatiol regarding the opinion of physicians about such services and the effect physicians' might have on the ultimate effectiveness of these services. This reselch is one of the first to: (a) Advance our understanding about how physicians vi1 enhanced participation of patients in medical decision-making and the integration ofTBDS services into care management and (b) provide key insights on the implications fir working with physicians in optimizing the integration of such services in the of health care. Overview of Research Design and Setting' This research has been designed tJ be primarily descriptive and hypothesis generating. A multi-method approach uslg both quantitative (survey research) and qualitative (key infonnant discussions) melhods has been utilized to address the specific aims of this study. This section ptvides an overview of the study design and study site as well as a description of the telephone-based decision support services that were implemented within the community lhere the physician subjects for this study were practicing. Key questions explored l this study are presented. The reconciliation of quantitative and qualitatije data are discussed. 8

PAGE 22

Study Design A multi-method approach was chosen to enhance methodological rigor. Patton (1990) has emphasized the value lftriangulation or the use of multiple I methods in the research and evaluation of a given phenomena. Triangulation or multiple methods can involve the utilizatiln of different kinds of quantitative and qualitative data and methods. Therefore, both quantitative and qualitative approaches were employed concurrently to assess provider perceptions and opinions. Survey research and key infonnanl interviews were used to ascertain physicians' opinions about: 1. The increased involvement of as partners in health and medical care decisions. 2. The impact they believe consumers/patients being more infonned and taking a more active partnershib role in health and medical care decisions will have on: the quality of the relationship, clinical efficiencies, health outcomes, pltient satisfaction, health care utilization, health care costs, and liability 3. The integration ofiDDS servit into care management and the impact they believe a TBDS service will have on: the quality of the doctor-patient 9

PAGE 23

relationship, clinical efficiencies, health outcomes, patient satisfaction, health care utilization, health care costs, and malpractice liability. Table 1.1 denotes the specific questions from the Physician Opinion St (Appendix A) and the Physician Opinion Interview (Appendix B) used to address each of these issues. T bl 1 1 Ph . 0 . S 6 k nfl a e .. l}'Sictan pwon urvey an: ey_ 1 ormant mterview questiOns. I Issue Physician Opinion Survey Key Informant I Intenriew Patient involvement I B1, B2, B3, C2 1, 2, 4, 5, 6, 13 Doctor-patient D4b 4, 5,6,9 relationship Clinical efficiencies D4a \ 4, 5, 6, 9 Health outcomes D3, D4b \ 3, 4, 5, 6, 9 Patient satisfaction D3, D4b I 4, 5, 6, 9 Patient retention D4b I Continuity of care D4b I Health care costs B2,D3 \ 4, 5, 6, 9 Malpractice liability D3, D4a \ 4, 5, 6, 9 Physician D4a reimbursement Utilization and :82, D4a I 4, 5, 6, 9 appropriateness TBDS Service Dl, D2, D3, Elf E2 7, 8, 9, 10, 11, 12, 14 10

PAGE 24

Study Site The study site was the communi of Rochester, New York (NY). Located in upstate NY on Lake Ontario, Rochester las a largely stable population base ( + 3% from 1980-1996) of just over one millioJ with a demographic composition ( 16% minority) comparable to similar sized urbL areas in the Northeast United States. It has an export-oriented, high technology manjfacturing-based economy. Unemployment rate is usually a percentage point or more lower than the U.S. average with the cost of living slightly above the national average. Prior to this study, the Rochester system received recognition as a "model" for health care delivery. Rochester's success is attributed primarily to strong regional planning and cooperation that focused primarily on supply side management initiatives including: community rating and continuot open enrollment, near universal coverage, I technology assessment and control, limitei hospital bed capacity, inter-hospital cooperation, and a moderate managed care presence. Health insurance and medical d d aril b I d h d d care were provt e pnm y y two camls an t err respective m epen ent practice associations (IPAs): (a) Blue Cross and Blue Shield of the Rochester Area (BCBSRA) and the Rochester Community IPA (RCIPA) which owned 73% of market share, and (b) Preferred Care, an independent HMO, and the Rochester IP A (RIP A) which owned 13% of market share at the tile of this study. There were 11

PAGE 25

approximately 2100 total physicians com rising both RCIP A and RIP A with 95% of the total physicians providing services both BCBSRA and Preferred Care. Approximately 80,000 insured fi es from seven major employer groups and two major health plans in Rochester weJ offered a TBDS service. This service had just been implemented at the time of this ldy. Available 24 bows/day and toll-free, the telephone-based nurse counseling 1ce was staffed by nurse counselors trained in assisting, educating, and supporting peeple faced with choices about health and medical care. The counselors were requiJd to be registered nurses with a minimum of five years of current clinical experience jd a minimum of a bachelor of science degree . mnursmg. Nurses who staffed the TBDS service received 240 hours of structured education and training. The training focu{ed on imeractive learning, designed to familiarize them with concepts of commumcation, communication skills, and applications of these skills in interactions tth callers. Training emphasized nondirective, non-prescriptive techniques of communication as well as patient-centered rather than provider-centered communica+ns. Second, the nurse counselors were trained in psychosocial assessment and self+-care alternatives, and how to apply them to 12 I

PAGE 26

the phone counseling process. Third, they received training on how to use computer hardware and software which comprised re service's infunnation system. The training curriculum for the nurse counselors was comprised of three phases. Phase I consisted of 160 hours initial training in communications, infunnation systems, and protocols for tching medical infOrmation. Phase ll embodied an intensive 80-hour mentorship with a senior nurse counselor which included performance audits and feedbacJ. Phase Ill incorporated another 8-10 weeks of modified mentorsbip with a senior ni counselor, also with performance audits and feedback. Fallowing this formal, initi81 training, nurse counselors received 60 hours of continuing education per year on information systems, medical and health topic reviews, and research protocols; participated in weekly case review meetings; d k 1 . .. \ ed an too part m quarter y m-servtce trammg sesstons 10cus on customer servtce. The primary objectives of the nursl counselors were to provide non-directive, . d . If. I "al I non-prescnpttve ectston support, se -management support, soct support, stmp e information requests, and referrals to local agencies over the phone on virtually any health and medical issue: (a) triage or pre-provider decisions concerning acute minor illness or injuries (e.g., headache, back pJ sprained ankle); (b) major medical decisions shared with health care provi, (e.g. surgery, chemotherapy, advanced imaging, hospitalization); c) chronic disease management (e.g. diabetes, asthma, 13 I

PAGE 27

hypertension); (d) lifestyle changes to reduce risk of disease and injury (e.g. body fat loss, exercise, stress management). In addition to the TBDS service, hll recipients of the service received the fOllowing member conununication matJals: (a) an introductory packet (letter of I introduction, brochure, wallet cards, refrigerator magnet, initial quarterly newsletter) regarding the telephone counseling servJe; (b) a copy of the medical self-care reference guide, Take Care if Yourself; l) eight postcard reminders to use the service; and (d) a quarterly newsletter. In Februlry, May, August, and November, subjects received the quarterly newsletter. The quLerly newsletters emphasized specific, I usable information for personal health and medical care decisions regarding preprovider triage, major chronic diLe management, and healthy lifestyles. Each month in between newsletter issues, subjects received one of the reminder postcards promoting the benefits of calling the service for a wide range of health and medical issues. The self-care reference Jde featured clinical algorithms for over I 00 common medical problems for adults and ihildren. Promotional and educational materials were supplemented with comprlhensive and coordinated employer and insurer communications through e-mail, vJice-mail, employee letters, and newsletters explaining and promoting the service as pi of the benefit plan being offered to them.

PAGE 28

Physicians in the community were provided information about the TBDS service as one component of a comprehJsive and coordinated community-wide communications plan developed and impllmented by executive leaders from the . 'd 1 d I . F' h part1c1patmg proVI er, emp oyer, an msurer orgaruzat10ns. 1ve to SIX mont s pnor to the intervention, the governing boards ls well as the various standing committees of the two participating IP As were provided with information about the community-wide demand management initiative and the 1DS service. Three months prior to the intervention, every physician in each ofthe IPAs (RCIPA and RIPA) received a letter from their respective elected exJcutives explaining the service, asking for their support and inviting them to participlte in a conununity physician "town meeting" planned for three weeks prior tore implementation of' the TBDS. This letter was accompanied by a roster of the physician's participating patients. Two months prior to the intervention, each of4e lP As as weB as the county medical society included an article about the commbruty-wide demand management initiative I and the TBDS service in their respective prcian newsletters. During this same time frame, a press conference describing the oor ectives of the initiative, the community aspect of the project, and the TBDS servic was held. A subsequent press release was distributed to local and national media. Three weeks prior to the start date for the TBDS service, all physicians were provided the opportunity to attend one of community physicians' "town 15 I

PAGE 29

meetings" jointly sponsored by the co, medical society and the two IP As. During these town meetings, infurmation rel demand management, implementation of the mos service, evaluation of the servir and employer perspectives were shared by a panel of physician, employer, insurer, and vendor executive leaders. Physicians attending the meetings were afforded the \oppor1unity to express their views and to have questions or concerns addressed. Tio weeks prior to the intervention, a physician advisory panel comprised of phtsician executive and clinical leaders the IP As and the county mekcal society were provided with the opportunity to "test" the TBDS service. Key Questions Important questions explored in this study included: 1 Wh h b I ab at are p ys1c1ans oplDlons a out patients wantmg more uuormat10n out health and medical issues? 2. What are physicians' opinions about patients wanting more involvement and taking a more active role in the health and mebical care decision-making process? 2a. Who should primarily make decisiols about treatment for minor and major health issues? 2b. What are physicians' opinions about patients taking a greater role in the self-management of chronic disease? 16 I

PAGE 30

3. What are the opinions of physicians about the impact they believe a more I informed, active consumer/patient will have on: the quality of the doctor-patient relationship, clinical efficiencies, heallh outcomes, patient satisfaction, health care utilization, health care costs, and +ce liability? 4. What are the opinions of physicians ibout the incorporation of telephone based decision support (TBDS) for into the delivery of health care services? 5. What are the opinions of physicians about the impact they believe the I incorporation of TBDS into the health care delivery process will have on: the I quality of the doctor-patient relationship, clinical efficiencies, health outcomes, patient satisfaction, health care utilizJion, health care costs, and malpractice liability? 6. Do physicians' opinions differ by age, gender, specialty, practice structure, visit volume, and reimbursement structurej 7. What are the primary reasons physicians believed that the health plans implemented TBDS services? 8. Which independent variables (age, gender, specialty, practice structure, visit volume, and% of patients capitated) lethe most significant predictors of physicians' opinions?

PAGE 31

Reconciling Quantitative and Qualitative alysis The final step integrated the results of both quantitative and qualitative analyses to summarize physicians' opiniols about these key issues. Methods triangulation is often accompanied by the risk of divergent quantitative and qualitative data. The subsequent challenge is developing a single, well-integrated analysis and interpretation. For this study, steps wer1 taken to minimize this risk. The questions for the Physician Opinion Interviews (qualitative data) were derived directly from the Physician Opinion Survey (quantitative Ja). That is, the interview questions obtained physicians' opinions on the issues as the survey, but were formatted to derive more in-depth, information-rich understanding of that issue through the interview process. Nonetheless, where diJcordance between quantitative and qualitative data occurred, the quantitative were used as foundation for interpretation wbile the qualitative data Jvided additional explanatory power. Organization of le Dissertation This chapter introduces the signifiL of the study, states the specific aims, provides an overview of the research and outlines the organization of what is I to follow. Chapter 2 presents a review of iterature pertinent to health care utilization, demand-side strategies for improving health care, physician and patient 18 \

PAGE 32

perspectives about more informed and aiively involved patients as partners in health and medical issues, and a review of factors that influence the diffusion of new innovations in health care. Chapters 314 are dedicated, respectively, to quantitative and qualitative methods, analyses, and results. Chapter 5 presents a discussion of the integrated quantitative qualitative results including limitations, implications, recommendations for futuJ research, and conclusions. 19 I

PAGE 33

CJIMlTER2 I REVIEW Oi LITERATURE A review of literature pertinent t0 this study is presented in this chapter. The following topics are discussed: (a) health \care utilization, (b) demand-side strategies for improving health care, (c) physician r patient perspectives about more informed and actively involved patients as partners r health and medical issues, and (d) a review of factors that influence the diffusion of new innovations. Geographical Variations and Appropriate Use ofHealth Care In the current debate on health i reform, three very salient and inextricably linked questions central to health policy and strategy have been posited: (a) "Will we I need to ration effective care?'' ( Brook & Lohr, 1986); (b)" Do the geographical variations in the use of services have any jlinical significance?" (Brook, Lohr, Chassin, Kosecoff; Fink & Solomon, 198t; and c) "Is it prudent to focus exclusively on how to manage the supply of medical services, or does a new concept demand management -deseJVe serious considefati,n as well?" (Vickery & Lynch, 1995). Defining, measuring and classifying appropriateness has been recently delineated by Brook, Kamberg, Mayer-oles, Beers, et al. (1990): 20

PAGE 34

A service is provided when its benefits (expected positive exceed its costs (expected negative .... Based on literature review, we havb delineated three approaches by which to judge whethJr a service is appropriate. .. Benefit-risk approach .. .I a service is judged to be appropriate only if the heflth benefits to the patient exceed the negative health consequences to the patient by a wide enough margin so that procedure is worth doing .... Benefit-cost approach .. l extends the benefit-risk approach by considering the resourbe cost of providing the service .... Implicit approach ... for judging appropriateness may have been used, but clear, explicit definitions are specified .... Each of these approaches ... can be used to classifY a service as being misused, or overused. The underuse of a procedUr-e, office visit, or hospital stay is the failure to use these when they are indicated .... Misuse implies that are incorrectly used . . Overuse is the use of a office visit, or hospital Admission that is not indiJated (p 3). What have we learned from healtJ services research about the appropriateness I of medical care utilization? Vickery and Fries (1994) have pointed out that up to 700/o of all physician visits for new problems jy be medically unnecessary. Referencing a significant and growing body of evidence medical care utilization research, nationally prominent health services scientists, Brook and Lobr, estimated that, ... perhaps one-third of the resources devoted to health care today are being spent on ineffective or unproductive care" (1986, p. 6). After conducting a literature review regarding the appropriateness of acute medical care for the elderly, Brook et a1 (1990) stated that, "Vutually every included in this review fuund at least double-digit levels of inappropriate care.]. erhaps as much as one-fifth to one-quarter 21

PAGE 35

of acute hospital services or procedures felt to be used fur equivocal or inappropriate reasons and two-fifths to on.e-half of the medications studied were overused in outpatients" (p. 225). In addihon to these studies of overuse, they noted that the few studies that examined undenlse or misuse of services also documented the of these phenomena. Furthermore, they noted that for the services examined in these appropriateness studies, very little is knoi about the effects of such utilization on health outcomes. This can be viewed pessimistically as an indictment of the current U.S. health system or optimistically as I a window of opportunity. Even while awaiting more data on appropriateness, mhch work needs to be done to decrease the extent in the U.S. health care system. Depending on one's beliefs, that work could take the form ... at the clulicallevel ofbetter education or dissemination ofinformatit to patients ... (1990, p. 240). Eddy (1990, 1991, 1993) has tended to support this perspective by pointing out that resolving our societal conflict with health Le costs will require connecting value to cost and addressing unchecked demand 1 volume and intensity of services by involving patients as partners with physicians in clinical decision making. Vickery and Lynch (1995) have p+sed one such opportunity fur addressing volume and intensity of service utilization through the concept of "demand management." They have defined demand Lanagement as, "The use of decision and self-management support systems to enablj and encourage consumers to make 22

PAGE 36

appropriate use of medical care" (p. 4-5). Vickery and Lynch (1995) pointed out that the supposition behind demand manag4ent is that health care decisions are influenced by a complex set of psychological, soj and cultural variables, most of which are unrelated to morbidity (nature and of the illness). Similar to Brook & et al, (1990), I they have defined appropriate decision making as that which optimizes benefits rhile minimizing risk. However, Vickery and Lynch ( 1995) have provided an intriguing, expanded, and intuitively sound psychosocial-cultural dimension to apprJpriateness. They have indicated that two components of demand, perceived need ld patient preference, are central to demand management and hold considerable for enhancing more appropriate utilization of health and medical care ,ces. For example, in regards to perceived need, they have pointed to research by social scientists such as Berkanovic ( 1981) who has estimated that 42% of the variance in decisions to seek health care can be accounted for by psychosocial variables ( ersonal beliefs, social networks etc.) with morbidity explaining only 12%. Recent evidence has indicated tha an individual's access to information and resources is an important variable in healt care decisions and subsequent health care utilization. That is, a well-informed pati,t will make different decisions than a less informed patient. Since several of the moJ popular contemporary telephone-based decision support services generally includ pre-provider triage (self-care and perceived 23

PAGE 37

need), shared decisions for major medi issues (patient preference), and chronic disease management; the empirical basis for each of these decision-support components will be reviewed. "Demand-side Interventions This section will review the literatbe pertinent to various types of "demand side" interventions. Particular highlight 111 be given to the impact of these interventions on health outcomes, patient satisfaction, health care utilization, health care costs, cost-benefit, and clinical efficielcies. Perceived Need and Self-care Vickery and Lynch (1995) have hypothesized that self-care interventions directly influence an individual's perceptioJ of the need for health care by addressing the individual's capabilities to assess and m\anage medical problems outside of the formal medical care system. As noted by Jickery and Lynch (1995), there is a considerable amount of evidence supportink the effectiveness of self-care interventions in decreasing demand for medical care. Vickery, Kalmer, Lowry et al. (1983) conducted a prospective, randomized, controlled trial of self-care educational intelentions within a health maintenance organization to determine the effect of se}f.lare written communication materials on 24

PAGE 38

ambulatory health care utilization. Statistically significant decreases of 17% and 35% were found fur total medical visits and tor illness visits, respectively. They also estimated that these decreases in health ea.re utilization resulted in a savings of$2.50 to $3.50 for each dollar invested in the interventions. Lorig, Kraines, Brown and RicJ.dson (1985) conducted a study on the impact of a workplace health education progrJ on reducing medically unnecessary outpatient visits. The purpose of the was to evaluate a health education program's ability to help employees makl more informed decisions about when to seek professional medical care at the physicil' office or the emergency room. The main finding of this study was that the self-helJ intervention decreased self-reported physician visits (17% or 2.0 visits/househlld/year) in first-dollar-fee-for-service benefit plans. However, in contrast to the et al. (1983) study, no reduction was found among HMO members. The reselchers also examined whether or not these decreases in visits resUlted in poor medici\ outcomes. They tested this indirectly by examining increased hospitalization and o tpatient visit rates during treatment periods I of the study. They reported that, "Neither of these results was observed although the power for detecting such changes in this .Ldy is not great ... In future investigations of self-care educational programs, more alention must be given to the long-term health benefits ... (Lorig et al. 1985, p. 1053 ). 25

PAGE 39

Utilizing a case-control pre-post iudy design, Levine and Britten (1973) studied the impact of self-administering blood products in 45 hemophiliac patients selected from a cohort of patients with 1genital coagulation disorders routinely followed through the New England Medic::al Center in Boston. The intervention consisted of theoretical instruction for palients and family members on hemophilia treatment and complications, instruction t sterile technique and venipuncture as well as suggested dosages for minor and hemorrhage, and two to four supervised infusions. The self-therapy intervention resulted in a reduction of 74%, 89%, 76%, and 45% for absenteeism, hospital days, otpatient visits and health care costs, respectively, for the 45 patients. J Most intriguing were the subjecti l evaluations completed by study participants. Levine and Britten ( 1973) r,orted that, "All4 5 patients indicated their complete satisfaction and felt that home therapy had greatly improved their lives. I None wanted to change the program .... Numerous patients volunteered the information that they were enjoying their rllationships with the physician and hospital staff for the first time. Several attributed j\. s to their own participation in health care and their decreased dependency needs" (pp. 198-199). This patient feedback tended to suggest that overall health status and qukty of life improved significantly as a result of patient participation in self care. HoweJer, except for the limited morbidity,

PAGE 40

mortality, and this informal patient satisfaction survey information, no formal evaluations of patient-assessed function1 status or health outcomes were conducted. Chronic Disease Management Greenfield, Kaplan, Ware, Yano, and Frank (1988) studied the effects on blood sugar control and quality of life in diabet cs who participated in medical decision I making. The intervention consisted of a 20-minute coaching session while the patients were waiting to see their physicians. Thj components of this session included: a) following an algorithm to review with thJ patient the last recorded medical visit, b) I assisting the patient in identifying medii decisions most likely to arise during the visit, c) encouraging the patient to focus en treatment decisions that could be affected by lifestyle and/or preferences, d) rehearsJng negotiation skills, e) encouraging patients to ask questions of their physician, f) revi1wing obstacles that can impede negotiations with their physician, and g) rehearsing h1 to overcome these obstacles. Glycosylated hemoglobin (HbAI c)l values for the experimental group decreased from 10.6 +/-2.1% at pre-intervention to 1 +/-1.9% (p<0.01) after the intervention. There was no change in (HbAic) values for the control group from pre-intervention (10.3 +/2.00/o) to post-intervention (10.6 +/-2.22%). These researchers reported that the experimental patients were approximately twice as effective as controls during post-intervention in the elicitation of infolation from physicians, reported 27

PAGE 41

significantly fewer functional limitations : obility, role, physical, lost work days, self care), and assessed their health (overall perception of health, health concern, number I of current health problems) more favorably. Mayo, Richman, and Harris ( detennined the effect of an outpatient program designed to reduce re-admissions and hospital days for asthmatic exacerbations among adults with asthma. The intervention was classified as an intensive outpatient treatment program wr.ch consisted of a vigorous medical regimen combined with an educational program wruch emphasized teaching patients aggressive self-management strategies in case of asthma exacerbation. The patient I education program did not involve any written, lecture, or audiovisual materials. Patient education was completed during rve contacts between the patient and physician. Hospital readmission rates and hospital day rates per patient per month I were reduced by 70% and 64%, In commenting on the fact that wrihen materials, lectures, and support groups I were not used in this study, Mayo et al. (1190) reported that," ... it rapidly became evident that the face-to-face interaction of lhe medical care provider and patient I resulted in the most meaningful transfer of information . .. Preliminary results suggest that a nurse practitioner is as effjve as a physician in reducing readmission rates among difficult adult asthmatics" (pp. 870-871). 28

PAGE 42

Lorig, Mazonson, and Holman ( 993) studied the effects of a four-year Arthritis Self-Management Program (ASMP) on pain, physical disability, physician I visits, and cost-benefit in osteoarthritis (IDA) and rheumatoid arthritis (RA) patients. Studies were also conducted to ascertain how the ASMP achieved the measured effects. Results indicated that change scor from baseline through four years fur the two ASMP groups were persistent and statistically significant (p
PAGE 43

versus elective surgery were more risk -a ersive than physicians. When BPH patients (n=284) at Kaiser Pennanente HMO in denver viewed an interactive video describing the benefits and risks of watchful waiting versus elective surgery for BPH, all but 30 chose watchful waiting over surgery. Tjs patient preference for watchful waiting over elective surgery approximated a 44% reduction in the rate of prostate surgery I among HMO members in the lint year. jimilarly, surgical rates initially plummeted 60% at Group Health ofPuget Sound jo in Seattle. Lennan, Brody, Caputo, Smith, Lazaro, and Wolfson (1990) constructed a Perceptions of Involvement in Care Scale (PICS), a self-report questionnaire that assessed patients' perceptions of doctor and patient behaviors that occur during a routine medical visit. Specifically, this mslment was designed to ascertain patient perceptions about doctor filcilitation of involvement, level of information exchange, and patient participation in decision making. Subjects used in the development of this instrument were pty HMO patients attending a primary care office at an academic medical center. The! subjects were predominantly female (62%) and black (800/o) with a mean age of39 years. A major finding of this study was tLt physician facilitation of patient 1 d ctual . I I d . akin "gnifi 1 mvo vement an a patient mvo vement m ectston m g were st cant y related to patients' satisfaction with care.ldditionally, physician facilitation of patient involvement and infonnation exchange wet consistently and significantly associated 30

PAGE 44

with patients' perceptions of post-visit c anges in their understanding, reassurance, perceived control over their illness, and lxpectations for functional improvement. Also, patients who reported that they wJe more involved in decision making with the doctor were more satisfied with their phJsicians' technical competence. These findings Ieinfurced previous research (WLa., hrui, Baniatua, Carter & Lippincott, 1984) which demonstrated tll important role physician facilitative behavior can and should play in producin improved therapeutic benefits for patients. An intriguing study by Brody, Miher, Lerman, Smith, and Caputo (1989) examined the relationships between patiets' perceptions about the roles they played during medical visits and their subsequenJ attitudes about their illnesses, treatments, I and self-rated improvements. This study involved 117 adult patients from a general internal medicine faculty practice which adult primary care to a predominantly HMO population. The meL age of the patients was 32 years; 62% were female, 85% were minorities, 92% HMO insurance, and a presented with symptoms suggestive of relatively minor illnesses (sore throat, vaginal discharge, musculoskeletal pain) at the Je of the study. Questionnaires assessing the roles batients wanted to play during the medical visit, the role they later perceived they plated, illness outcomes, patient attitudes about their illnesses, and patient satisfaction wi, their physician were administered befure the medical visit and then again one day and one week after the visit. Their physicians 31

PAGE 45

also completed a questionnaire which provided infonnation about the seriousness of the problem and expected alleviation of atient discomfort and dysfunction one week later. Forty-seven percent of the patijs reported playing an active role wbile 53% of the patients reported playing a passiv, role. One week following the medical visit, active patients reported statistically (p = 0.04) less discomfort, greater all f d I h. a1 edical di. evtatton o symptoms, an more 1mprorement m t err gener m con t10n than did passive patients. One day post-VIsit, active patients reported less concern with their illnesses (p = 0.04), a greater sense if control of the illness (p = 0.04), and more satisfaction (p = 0.02) with their physi,s than did passive patients. The researchers concluded that their findings were consistent with others who have demonstrated the impact that increasing a patient's participltion in care can have on improved disease controL higher quality oflife, and a more latisfactory doctor-patient relationship. They speculated that patients' increased slnse of control and self-efficacy may be possible explanations for their findings. Patient Preference for Information and Involvement in Decision Making The traditional doctor-patient relatonship has been characterized by a Parsonian, paternalistic approach the patient has played a predominantly passive, dependent role and the physician r asswned an autonomous, professionally 32

PAGE 46

dominant role in the health and medical care decision-making process. However, a variety of social, political, and cultural +s in combination with technological advances in communication and management point toward a changing doctor -patient relationship characterized r increasing consumer/patient participation in decisions affecting both health and medical care services. Although the previously referenf studies have provided evidence that the inclusion of patients in the health and medical care decision-making process can result in more appropriate use of health care returces, decreased health care costs, increased patient satisfaction, improved control of disease, and better clinical and functional outcomes; it is also important tl address the fundamental questions related to the consumer/patient's role in clinical Jecision-making. That is, to what extent do consumers/patients prefer to be more infoLed and what type of information is desired I about health and medical care decisions? fo what extent do consumers and patients want to be more actively involved in what type of health and medical care decisions once they have been afforded specific infoFation? And to what extent should physicians encourage active and collaborahve patient participation in health and medical care decisions? Cassileth, Zupkis, Sutton-Smith, ,d March ( 1980) examined the degree to which 256 cancer patients desired to be more informed and to participate in decisions regarding their care. Using an lnformatioJ Styles Questionnaire, patient preferences 33

PAGE 47

regarding involvement in medical decisiTs were measured using two alternatives one indicating a preference fur their Jr to make decisions about their medical care, the other indicating a preference to participate in decisions about their medical care. Similarly, preferences for quantity and q1ity of information regarding diagnosis and treatment were ascertained. l The most notable findings from t s study were: 1. Patients who desired detailed Jrormation about their treatment in I comparison to those who didn't seek sue! information were younger, white, better educated, and more recently diagnosed. 2. Those patients who preferred to participate in treatment decisions were younger and better educated compared tj those who prefurred to leave treatment decisions up to their doctor. 3. Even though age was the only factor that distinguished those patients who wanted more infOrmation and active involr in their own care from those who preferred a minimum of information and active involvement, a majority of patients in I each age group preferred active participation and detailed information. 4. There was a strong correlation Jetween preferences for maximum =-c. d h d I d uuormat1on an t e esrre to partiCipate m ec1s1ons. 34

PAGE 48

5. Because not all patients desired an active involvement in the medical decision-making process, physicians sholld be attentive to patient preferences for such roles. Stroll, Lo, and Charles (1984) examined to what extent 210 hypertensive patients from a conununity hospital c, an HMO, and a Veterans Administration outpatient clinic preferred more infonnJ' on about hypertension and preferred to participate in medical decisions. Cone rntly, the patients' physicians were asked to estimate their patients' preferences for more infonnation about hypertension and preference to participate in medical decisibns. The researchers also studied whether or not patient demographic and other health lcare related fucto111 affucted patient preferences. Similar to other studies, sJn et a!. utilized five different preference alternatives encompassing "active", "conaLrative", and "passive" decision making styles to ascertain how decisions are made and patient preference for how decisions should be made. 1 The major finding of this study w that clinicians underestimate patients' desire for more infonnation and discussioL about therapy but overestimate patients' desires to participate in decisions. That is\ 41% of the patients preferred more infonnation about their illness, but clinicians underestimated patient desires for discussion about therapy in 29% of cases; and while 53% of patients preferred to 35

PAGE 49

participate in treatment decisions, clinicians estimated that their patients desired to participate in decision making in 78% ofle cases. In terms of how decisions are us1y made, 63% of patients that the clinician usually makes the decision whill clinicians reported such decision making in only 20"/o of the cases. Additionally, 80't of clinicians reported that the patient participates to any extent in decisions while only 3 7% of patients reported that they d akin I participate m eas1on m g to any extent. In terms of how decisions should L made, 47% of the patients preferred that the clinician make the decisiols without patient participation. Approximately one-third of the patients Jmcated that the physician should make the decision while strongly considering the palient's opinion. Less than one quarter (19%) of the patients indicated a preference to Le equally with the physician in making decisions and 3% indicated a preference ft making decisions themselves. In contrast, physicians believed that patients wanted ti participate in decision making in 78% of the cases. In only 22% of the cases did the physician believe the patient wanted the clinician to make the decision without the batient. Because of possible physicianpatient discordance about desires fur mJ1 information and preferences fur medical decision-making, and, given the fact that enhanced patient participation may result in improved outcomes, this study acknowledied that physicians may want to find explicit 36

PAGE 50

ways to discuss these issues with patient in order to facilitate communication and decision making. Using a study population of312 ,atients randomly selected from a hospital based primary care clinic, Ende, Kazis, ASh, and Moskowitz (1989) developed the informed and their desire to make medic decisions. The API consisted of two scales --one to assess information seeking and one to assess decision making. For each of the two scales, totals were linearly adjustjd to range from 0 to I 00, where 0 corresponded to complete lack of desire rl. information seeking or for decision making and 100 corresponded to the strojgest possible desire for each of these dimensions. On the 0 to 100 scales for information seeking and preference for decision making, the mean scores were 79.5 and 33.2, respectively. There were no significant correlations between patients' desires for iDformation and their desires for decision making or with patient satisfaction. with ather studies, they found stronger preferences for active involvement in deciston making in younger, more educated, higher income individuals, and less preferelce for involvement in decision making among the more severely ill. Interestingly, \after adjusting for sociodemographic fuctors and severity of illness in a regressiol only 19% of the variance in decision making preferences could be explled. The authors concluded that patients 37

PAGE 51

had a strong desire to be well informed, ut, in support of other studies, relatively I weak preferences for participating in medical decisions. The findings of this study tended fo support the view that patients prefer a doctor-patient relationship whereby the doctor keeps the patient informed and ed d . akin bl d th fun engag m ects1on-m g processes, ut oes not reqwre e patient to ctlon as the principal decision maker. The pointed out that a great deal of the individual preference among patients for information and decision-making roles may be more related to ingrained specific culJral and ethical variables rather than the sociodemographic factors measured in thil study. To this point, they, like Stroll et al. (1984}, acknowledgld that physicians should have explicit discussions with individual patients regar g information and decision-making preferences. Sutherland, Llewellyn-Thomas, Lockwood, Tritchler, and Till (1989) used a convenience sample of 52 patients (35 woLen and 17 men) with a mean age of 48.5 years who were undergoing treatment for lancer (radiation therapy or chemotherapy) to study their desire for intbrmation and +erences for participation in treatment decisions. Preferences for intbrmation wT assessed with two instruments the Health Opinion Survey (HOS) and the Information Seeking Questionnaire (ISQ). P c. c. . . dl.. ed . re1erence 10r participatiOn m treatment cisJons was assess usmg a quest10nnarre adapted from Stroll et al. (1984). 38

PAGE 52

The results showed that although most patients indicated that they actively sought information, 63% of the them felt the physician should take the primary responsibility in decision making, 27% fit decision making should be an equally shared process, and 10% felt they should a major role. The authors pointed out that this discrepancy between patient preferences for information compared to their preference fur a limited role in decision .L may be best explained by the patient's desire ... to satisfy an aspect of 'psychblogical autonomy' that does not necessarily include participation in decision making'' 262). Like others, they suggested that I clinicians should be sensitive to situation-dependent preferences for information and d akin h f di d a1 I ec1s1on-m g on t e part o m VI u Another study assessed patient desire for information and involvement in decision making using the Autonomy Preflrence Index (API) and the Health Opinion Survey (HOS). The subjects were 167 unilersity-based primary care outpatient I patients with benign prostatic hyperplasia, back pain, or mild hypertension (Nease and Brooks, I 995). Assessments from both insLents indicated that patients' desire fur information exceeded that for involvement in decision making. Consistent with the findings by Ende et al. (1989), the API decision-making score in this study decreased as the seriousness of the clinical condition tcreased. That is, as the seriousness of the hypothetical clinical condition increased, patients' desire for involvement in decision making decreased. 39

PAGE 53

Consistent with previous studies, rhis research pointed out that there is substantial variation among patients in th ir desire for information and for their involvement in medical decisions. That is,l although sociodemographic factors (younger age, female gender, higher eduJI lion) may be associated with greater desires for information and decision making, a 1 ;ge part of the variance for patient responses to these two aspects of medical care remals unexplained. Similar to others (En de et al. 1989; Sutherland et al. 1989; Strull et 1. 1984; Cassileth et al. 1980), these researchers pointed out that even though desire for information is high, the substantial variation among individual patilnts for involvement in decision making should serve as a caution against rigid polibes and recommendations about the degree to which patients should be involved in mtcal decisions. Beisecker and Beisecker (1990) studied 106 rehabilitation medicine patients (42 men and 64 women ranging in age frol17-85 years) to determine information seeking behaviors and decision-making prJerences using the Desire for Medical I Information and the Locus of Authority in Medical Decision-making Scales. The results indicated that patients desired infoiation about a wide range of medical topics, but when communicating with their they did not engage in many information-seeking behaviors. FurthermJe, patients believed that medical decision making authority should rest primarily with the physician rather than with the patient. Finally, it was noted that patient information-seeking behaviors were more closely 40

PAGE 54

associated with situational variables such as length of interaction, diagnosis, and reason for the visit than with patient atti des or sociodemographic characteristics. The researchers surmised that inflrmation-seeking behavior was attenuated in this study because the subjects were primkruy patients with chronic diseases and disabilities. Like others, they also emphjized that situational factors (diagnosis, presence ofa companion, reason for visit surrounding the doctor -patient relationship may better explain patient infirmation seeking behaviors and desires to be involved in medical decision-making than sociodemographic variables. They suggested that future research examine thl impact of situational variables on information seeking behaviors and preferebce for involvement in medical decision-making. Bradley, Zia, and Hamilton ( 1996) evaluated patient preferences about involvement in medical decision making and factors that might influence these I preferences in a predominantly white female population. They used a survey instrument that included I 0 specific case depicting different acute and chronic medical situations that varied in severity. For each scenario, patients were asked to select one of seven responses varied levels of physician/patient decision making. These responses ranged lorn very passive ("I prefer that my doctor tell me what to do") to collaborative ("I that my doctor and I make the decision together") to very active ("I prefer that I 1e the decision without any information or 41

PAGE 55

recommendation from the doctor") prefe ences for involvement in decision making by the patient. The survey responses in this stud 1 indicated that patients filvored the physician as decision maker. However, the prefer,ces for involvement in decision making varied by scenario. The authors acknowledged that patient decision making preferences were scenario specific and p4bably illness specific. Patients wanted less control in the hypertension scenario and Jore control in the cancer scenario. Consistent with other studies ( Cassileth 1 al. 1980; Strull et al .. 1984; Ende et al., 1989), increased age and lower educational levels were related to increased emphasis on the physician as primary decision mak:J whereas lower age and increased educational levels were related to increase6 patient control. Similar to others (Nease and Brooks, 1995; Ende et al. 1989; Suthjrland et al. 1989; Stroll, 1984; Cassileth et I al. 1980), this study suggested that: a) preference for patient involvement in decision-making depends upon the situation and b) bhysicians must Jearn to be both facilitators of decisions as well as decision makers upon the patient and the situation. Hack, Degner, and Dyck (1994) eJammed the quantitative and qualitative relationships between preferences for medical information and preferences for involvement in making treatment decisions in 35 women (aged 32-83 years) with stage I and II breast cancer who were 2-6 months post-diagnosis. Using a card sort technique developed by Degner and Sloan 1992), they explored patient preferences 42

PAGE 56

for specific medical infonnation and preferences for involvement in making treatment decisions. This card sort technique pro1ded patients with five different degrees of preferred involvement in medical decisiols ranging from active ("I prefer to make the final selection about which treannent I +ve") to collaborative ("! prefer that my doctor and I share responsibility for which treatment is best for me") to passive ("I prefer to leave all decisions my treatment to my doctor''). They also developed a card sort measure to delennine patient preferences for diagnostic, prognostic, and treatment-related infonnltion desired by the patient. The information preference card sort procedure addressed the specific manner by which patients want physicians to provide them with information (quantity, media, type of descriptive terminology to be used). Overall, the findings suggested that patients who desire an active role in choosing their treatment also desired morl detailed infonnation about their medical care. That is, those patients who preferre1 a more active preference for involvement in decision making also preferred to receit more detailed infonnation regarding their care while those patients who preferred a more passive style preferred either a brief explanation or did not want any infonnatiln at all. However, consistent with other studies regarding patient variability (Ende L al. 1989; Sutherland et al. 1989; Strull et al. 1984; Cassileth et al. 1980), Hack et aljpointed out that there were nine passive patients who wanted detailed information egarding their treatment. 4B

PAGE 57

Second, all of the patients who preferred active involvement in decision making and most of the passive patients wanted tir physicians to provide them with all treatment alternatives and the degree of rii associated with each alternative. Active and collaborative style patients wanted complete information about all major alternatives while the passive patients pjerred to hear about the best single treatment or hear nothing at all. Third, education was the only sociodemographic variable significantly related to decision-making role preference. ThoJ patients who preferred an active decisionmaking role were more likely to have gradLted from high school whereas the nonhigh school graduates preferred a more pre role. In contrast to other studies, age (Cassileth et al. 1980; Strull et al. I984; Ende et al., 1989), and illness severity (Cassileth et al. I 980; Ende et al. I 989; Bjadley et al. I 989), were not significantly related to decision-making role prererencel. Fourth, the qualitative analyses disJiosed that: a) patients actively engaged in decision making wished to play such a rol, because it affords them a greater degree of control over "body, health, and life;" b) the more passive patients wanted to be kept infonned, but at the same time it was impolm "to place faith and trust in their physicians because only physicians possess the medical knowledge and expertise necessary to formulate treatment decisions'[ (p. 286); c) active patients were more like to believe their physicians wanted them to De more active in medical decisions; d) 44

PAGE 58

being pressured to play a more active role in decision-making processes was more anxiety-producing for the passive patien s; and e) both active and passive patients style and circumstances. To this point, Hack et al. noted t t several patients emphasized that physicians needed to be sensitive to and respect the information and decision-making preferences of each individual patient. Preferences for information and decision-making may differ depending on milestones related to sevedty of illness, elapsed time since diagnosis, and treatment progression. Similar to other Ldies, the younger and better educated patients expressed a greater desire for ilvement. Using the same card sort procedures, Davison, Degner, and Morgan (1995) assessed information and decision makinJ preferences of 57 men (mean age 71 years) with prostate cancer who were 0-6 montJs post-diagnosis. The majority of the men in this study were married (74%) and had lels than a high school education (54%). Twenty-nine (51%) had received no +t. Consistent with another recent study by Degner and Sloan (1992), a majority of men (58%) in this study preferred a ,ve treatment decision-making role, while 23% and 19% preferred a collaborative anCI an active role, respectively. However a high proportion of men wanted more inflation regarding likelihood of cure, types of treatment available, and disease progressiol. 45

PAGE 59

Speculations as to why the majori of men preferred to delegate treatment decisions to their doctor included: 1) old r married male patients tend to involve their wives in medical care decisions and subse uently tend to relegate to them a decision-making role; 2) older patients may be less likely to believe they should challenge the authoritarian role of the physician; and 3) older patients may actually prefer a more authoritarian model in the doctor-patient because they have a high level of trust in their personal physician; 4) there ry be a fear of assuming responsibility for decisions that may lead to poor outcomes;\ and 5) they may believe they lack the knowledge and information necessary to make a good decision. As has been true of most of the stuhies reviewed in this section, Davison and colleagues (1995) point out that informatiJn seeking and decision-making preferences are determined by a variety of sociodemo+pbic, intra-personal, and situational factors. Thus, clinicians should be explicitly sensitive to individual differences in regard to needs for information and involvJment in medical decisions. Davison and associates (1995) also sounded an approje call fur the critical and urgent need for research in determining the role thlt wives .and the family unit play in health and medical care decisions. Beaver, Luker, Owens, Leinster, Degner, and Sloan (1996) also used the Degner and Sloan (1992) card sort procedu\e to examine the degree of control desired in treatment decision making in 150 women (mean age 54.8 years) newly diagnosed 46

PAGE 60

(2.5 weeks) with breast cancer. Their r aim was to ascertain whether these women preferred to play an active (patient-controlled), collaborative Gointly\ controlled between patient and or passive (provider-controlled) role in decisions regarding breast cancer treatment. For a majority (66%) of the newly diagnosed women, lumpectomy was theJ.\lanned treatment with mastectomy the planned treatment for the remaining 34 o. Two hundred women with benign breast disease served as a comparison group. I The most significant finding of thil study was that a majority of the newly diagnosed women preferred a passive roll, leaving treatment decisions to their doctor; whereas the women with benign breast dilease (in the comparison group) preferred a I more collaborative role. For the newly diagnosed group, the variables found to be weakly correlated with decision-making pteference were age, with older women preferring a more passive role; education, lower levels of education associated with preference for a more passive role; arld social class, with lower social classes preferring a more passive role. However, regression analysis, age and level of education were not predictive of decision laking preferences. Even though social class was predictive of decision making pr1\ferences (p = 0.04), it's influence was relatively weak accounting for only 2% of he variance. In the benign group, age and social class were weakly correlated with prlferences for decision making. A,family 47

PAGE 61

history of breast cancer and marital status were not associated with decision making preferences in either group. Beaver and colleagues speculated about the explanations for these findings. Using supportive information from other cancer studies and preferences for decision making, it was argued that potentially life threatening illnesses may make patients more passive. They also noted that even tough a majority of the women with newly diagnosed breast cancer preferred a more passive role in decision making, there were still many who wanted some fum of involrent. They cite that a lack of appropriate information could also be one of the reasons that women felt uncomfortable in playing an active role in decision making. Similar to previous research cited in this review, they emphasized once again that the most prudent clinical approach is )or the clinician to be sensitive to and take into account each individual patient's prefJences for decision making involvement. They reiterated that there seems to be no sLple predictive formula for determining I patient's preferences for decision-making. o assist clinicians with such assessments, they suggested that the card sort procedure could easily be integrated as a screening tool into the everyday practice of medicine. The authors indicated that future studies were underway to determine the effect of"tune since diagnosis" on decision-making preferences. 48

PAGE 62

A particularly interesting and noteworthy study was conducted by Thompson, Pitts, and Schwankovsky ( 1993) the situational and demographic influences I on the preference for involvement in medical decision-making. Participants for the study were 459 randomly selected memt of an HMO wbo were predominantly white (87%), female (68%), low to middle income (49%), married (65%), had some college (63%), with a mean age of 54 rel.. To assess patient preference for decision making they used the Ende et al.(1989) ktonomy Preference Index (API) and the I Desire for Involvement Questionnaire (DIQ) they developed specifically for this study. Thompson et al. (1993) the significant difference between the two instruments. The API is constructed to ask about patient preference in regard to medical situations that generally fall wiJ the province of a medical professional. The DIQ is designed to ascertain desire fJ input into situations where medical expertise is not required. Four vignettes +varying severity (knee injury, high blood pressure, cancerous growth, and stress heddaches) each requiring two treatment I choices with varying side effects, trade-off! between quality and length of life, and effects on patient's lifestyle were used to preferences. As hypothesized, one situational faLor was found to be strongly related to patient desire for involvement. Respondentl expressed a greater desire for involvement in decisions that did not require medical 4ertise (DIQ) and significantly lower preferences for involvement in decisions that required medical expertise (API). With 49

PAGE 63

this finding, Thompson and colleagues suggested that the conclusions reached in other studies (Strull et al. 1984; Ende et al., 19k9; Sutherland et al. 1989; Beisecker and Beisecker, 1990) about patients not wantLg to be involved in medical decisions are misleading. Additionally, in an analysis to see if experience with a condition impacted the desire for involvement, they found that thLe who had hypertension were less likely to I want decisional involvement than those who did not have hypertension. They speculated that experience with a chronic bondition and all of its associated difficulties may lessen the desire for resplbility. From a clinical practice viewpoint, Thompson and associates (1993) pointed out the documented advantages ofhaving patients participate in'medical decisions while at the same time conceding that clinibians need to recognize that there are individual patient differences in decision-mlg preferences. They urge clinicians to be sensitive to these differences and subseJuently to the types of decisions for which patient involvement is most likely to be and beneficial. In summary, it is clear that patients and desire information about the diagnostic, prognostic, and therapeutic aspects of medical decisions. What isn't so clear is to what extent patients want to be rvely involved in medical decisions. The literature reviewed here has presented conflicting findings. The evidence has suggested that there are certain and situational factors which have 50

PAGE 64

been consistently lissociated with decisiormaking preferences: I) age, with younger individuals preferring a more active role; 12) education, with more highly educated individuals preferring a more active gender, with men preferring a more passive role than women; 4) marital stai with married men preferring a passive role; 5) prognosis, with more seriously ill patients preferring a more passive role. Two critical variables that need further studyJ \ elude time since diagnosis and the role that wives and the family unit play in health d medical care decisions. It is clear from this literature reviel that information seeking and decisionmaking preferences are determined by a Lety of sociodemographic, intrapersonal, and situational factors. The most prudent llinical approach is for the physician to be sensitive to and take into account each inkvidual patient's preferences for information and for involvement in decision-making. Finally, the intriguing work by Th,mpson et al. (1993) indicates that there may be some significant methodological issues isociated with how we are assessing patient preference for involvement in medical decision-making. That is, respondents in their study expressed a greater desire for iJvolvement in decisions that did not require medical expertise as measured by the D, fur Infurmation Questionnaire (DIQ) and significantly lower preferences for involvement in decisions that required medical expertise (API). Like the API, the Locus Jf Authority Scale utilized in the Beisecker and Beisecker (1990) study also required to answer many of the 51

PAGE 65

questions. What information should be covered in the patient's history? Whether there is a need for medication? Which drug to hse? What dosage to use? Whether to change medications? Whether to have the patienl undergo diagnostic tests? Whether to treat the patient in the hospital or on an basis? Thompson et al. 's ( 1993) opinion that misleading conclusions have been relched in other studies (Strull et al. 1984; Ende et al., 1989; Sutherland et al. 1989; rsecker and Beisecker, 1990) about patients not wanting to be involved in medical decisions warrants important consideration. Physician Perception about Patient Desire for Infonnation and Involvement There is a substantial body of evidLce regarding patient preference for information and medical decision making. However, there seems to be a coinciding scarcity of information regarding physician perception about how active patients should be in seeking information or in in health and medical care decisions. The purpose of this section is to review some of the more pertinent physician infonnation that may provide further insight on the matter. I As noted previously, Strull et al. (1984) examined to what extent 210 hypertensive patients from a community hobital clinic, a free-standing, HMO and a V Ad lini. I ed b eterans nurustratton outpattent c c preJ.err more uuormatlon a out hypertension and preferred to participate in medical decisions. Concurrently, they 52

PAGE 66

detennined how well clinicians estimated their patients' preferences about these aspects of medical care and whether or not patient demographic and other health care related fuctors affected patient In terms of how decisions are usually made, 63% of patients reported that the biinician usually makes the decision while clinicians reported such decision making l only 20'/o of the cases. Additionally, 80% of clinicians reported that the patient pjcipates to any extent in decisions while only 37% of patients reported that the patient in decision making to any extent. In tenns of how decisions should le made 47% of the patients preferred that the clinician make the therapeutic decisij without patient participation. Approx imately one-third of the patients indicated lat the physician should make the decision I while strongly considering the patient's opron. Less than one quarter (19%) of the patients indicated a preference to share eq11ally with the physician in making de-cisions and 3% indicated a preference for making themselves. In contrast, physicians believed that patients wanted to participate in decision making in 78% of the cases. In only 22% of the cases did physician believe the patient wanted the clinician to make the decision without the P,atient. Because there was relatively little demographic infonnation characterizing thl physicians in this study, vei)Ilittle else about physician perceptions regarding patilt information seeking and involvement in I decision-making can be extrapolated from this information. 53

PAGE 67

Doctor-Patient Communi tion & Relationships: Challenges/Barriers ahd Effect on Outcomes One of the very appropriate and rd concerns that physicians have when health plans begin to consider the imple,entation of24-hour telephone-based nurse counseling (TBDS) services is, "What r will this service have on the doctor patient relationship?" Preserving the physician-patient relationship in an era of managed care has become a central concJm for most physicians (Emanuel and Dubler, I 995). In studying the perceptions and olunons that physicians have about contemporary TBDS services, it's impera!ive that we have a fundamental understanding of the communication dynLcs, challenges, and success factors which comprise the doctor-patient relationship. Traditionally, the doctor-patient rrionship has been predominantly a paternalistic one in which the patient has assumed a very passive, dependent, and cooperative/acquiescent role while the phtcian has assumed a very active/directive, autonomous, and professionally dominant/controlling role. Recently, however, ethical, social, economic, and legal concert in combination with cultural shifts and advancements in communications and management technology have begun yet another era in the "social transformation of American medicine" (Starr, 1982). This era is advocating a more informed coJsumer of health and medical care; more patient .autonomy and self-determination in medical decisions; and a mutual, 54

PAGE 68

Collaborative partnership between physicians and patients in communicating with one another and in subsequently making h..k and medical care decisions. In their review of the literature ol doctor -patient coliiiiiWlication, Ong, DeHaes, Hoos, and Larnmes (1995;p. ,3) stated that, "ColiiiiiWlication can be seen as the main ingredient in medical care." In their treatise, they examined this issue in three domains: purposes of medical colunication, communicative behaviors, and the influence of communicative behavior\ on patient outcomes. This will serve as a useful framework for our review. Doctor-Patient Communications: Challenges/Barriers Ong et al. ( 1995) identified three different purposes of communication between doctor and patient: I) to develoi an inter-personal relationship, 2) to exchange information, and 3) to make diisions. Since perspectives on decision making have been previously reviewed in Ws report, attention will be focused information exchange processes. The development of interpersonal relations and the information exchange process bas included the actions giving, receiving, inteipreting, and evaluating verbal and non-verbal informati n by both patients and doctors. Physicians and patients seek and need information for obvious reasons. Physicians need different 55

PAGE 69

types of information for diagnostic, prognostic, and treatment planning purposes. As reviewed previously, patients have a stro g desire and preference to know as much about their medical situation as possible Ong et al. (1995; p. 904) have captured the patient's need for information by stating, "From the patient's point of view, two needs have to be met when visiting the doctor: the need to know and understand (to know what is the matter, where the pain comes from) and 'the need to feel known and understood' (to know the doctor accepts him and takes him seriously)." Speedling and Rose (1985; p. 116) have supported his perspective by stating that, "The I physicians' expertise in curing the illness is rarely seen to count more than his or her facility to care for the patient's psychosobal needs." Several factors have contributed tl documented discordance in information exchange between patients and physicians Observational research (Beclanan and 1984) has indicated that the physician interrupts the patient's beginning narrative after only 18 seconds; and the t complaint presented by the patient may not be the most critical one! It has been noted that 60-80% of medical diagnoses and I treatment decisions can be made without the need for expensive diagnostics just on the basis of a complete history and listening tj what the patient has to say (Hampton, Harrison, Mitchell, Prichard, and Seymour 1975; Sandler, 1980). Sir William Osler, the father of modem internal medicine, astUtely implored physicians to listen -for the I patient will reveal the diagnosis (Osler, 18 9). 56

PAGE 70

In support of the previous DiMatteo (1994) outlined fuur barriers or limitations to physician-pati,t communications: 1) physician use of medical jargon resulting in patient lack ofunderstanding and misunderstanding; 2) lack of active listening to patients by physii; 3) the desire the patient has for infonnation is usually not fulfilled by the provider; and 4) societal expectations and the I influence of medical tradition make it difficult for patients to effectively assert their needs for infonnation and participation 1 decision-making. Facilitating patients in telling their story about their illness expenence and then actively listening to what they have to say are critical success fu1 optimizing the doctor-patient relationship. As noted previously, patients alm1st always want as much infonnation as possible about their situation. However, physicians have tended to I underestimate this desire for information itrull et al. 1984), they may have subsequently provided insufficient infonnation to the patient to fully satisfy his/her needs. Ong et al. (1995), like DiMatteo (111994), have pointed out that another contributor to patient dissatisfaction with he exchange of infonnation is rooted in a lack of concordance between patient perclption and physician perception about the !l'!!ll of infonnation exchanged. often view medical infonnation as that which is objective (diagnostic test results, prognosis, type of disease, treatment procedures) whereas patients define medit infonnation in terms of its personal relevance to their illness experience (what IS the likelihood of cure? what is the effect Sil

PAGE 71

of treatment on my job? social life? farnilr? sexual activity? risk of disease to my family?). To this point, F rederikson ( has emphasized that a more effective strategy for physician consultation with is one which attends to the "concepts, perceptions, and views of the patient" raL than providing vast amounts of standardized information. Utilizing her mbdel of information exchange for medical consultations, she studied tasks in a self-selected sample of 3 5 general practice doctor-patient dyads. The consuhation evaluation lent was a questionnaire which asked the patient to rate the level ofintportance ( esr desirable, unimportant) for the following information exchange tasks: exploring patient ideas, dealing with patient concerns, exploring patient expectations diagnosis and treatment, discussing the effects of problems, explaining the etiology, explaining treatment, discussing side effects, and outlining a prognosis. The fobowing outcomes were also rated by the patients: satisfaction, patient understandink of health problems, doctor's response to h f I f :ffi ed, t e pattent, perceptions o appropnateness o treatment o er treatment mtenttons, development of the relationship between Joctor and patient, and change in concern and information provision. The information exchange tasks most strongly correlated with patient satisfaction were: dealing with patient conlems, exploring patient expectations, 58

PAGE 72

exploring ideas, and discussing elects of problems. Fredriksen (1995) noted that the quality of information provided y physicians to patients is more important than quantity and that physicians should arget information toward the specific individual needs of the patient. von Friederichs-Fitzwater, Flynn, and Williams (1991) analyzed 30 transcripts of conversations between p,cians and patients in fuur different settings (AIDs clinic, hospice program, family practice clinic, hospital room/doctor's office). The purpose of this study was to identifY Llational control patterns in the . b h . d I Th h h commurucattons etween p ystcians an patients. e pnmary emp as1s was on ow (relational dimension) physicians commuJcate with patients rather than on what (the dim ")h"" .\d .. content enston p ys1c1ans commurucate to patients. A total of 5,968 utterances were ctded fur furmat (assertion, question, talkover, noncomplete and other) across nine relational categories (support, nonsupport, extension, answer, instruction, order, disclnfirmation, topic change, and initiation. ")C .. b J\ .. d. d tenrunation ommurucations etween p ystcians an patients were concentrate into three of the nine possible relational ca egories: extension, support, and answer with predominantly symmetrical exchangeJ The researchers concluded that the interactions were primarily doctor-<:enterj or disease-centered as opposed to patient centered as the physicians frequently chanJed the topic, asked more questions and talked more than the patient. 59

PAGE 73

Further analysis also revealed thjt there was a tendency on the part of patients toward a more domineering behavior. However, further analysis of the conversations reflected a ... lack of the dialectic labt ofunderstanding, oflistening and interpreting, of clarifying and acknowledging what has been said, and responding" (p. 30) by the physicians. Additionally, whe1l the patients in this study raised an emotional issue, physicians tended to respond with question about physical symptoms or change the subject, therefore, controlling le meaning of the illness. The findings of this study support1 previous relational research and the frequent complaint by patients that they l not pennitted to tell their stories. The authors suggested that physicians could djvelop a more patient -centered partnership model of care by allowing patients to talk bout their illness experience; providing I more empathy and direct support; and facilitating patient communication around topics which are particularly meaningful to the 1ient such as feelings and expectations. Suggestions for further research in relational communications included: the significance of gender on relational and the influence of I physician gender on patient credibility; the influence of physician attitudes about dh h h I 1 hi d 1 certam toptcs; an ow t e p ystctan-patleDt re at10ns p eve ops over tune. Lerman, Daly, Walsh, et al. (1993) ldied 97 patients (mean age of 55 years) I with Stage I or Stage II breast cancer who had all undergone primary surgical treatment and who were scheduled to recei e postoperative radiation or 60

PAGE 74

chemotherapy. This study evaluated patient-provider communications in relation to patient coping styles and psychological Patient-provider communicationiwere evaluated using the three subscales from the Medical Interaction factor of th Cancer Rehabilitation Evaluation System (CARES): 1) Problems Obtaining InfoJation from Medical Team; 2) Control of Medical Team; and 3) Difficulty Commu1cating with Medical Team. The percentage of patients reporting one, two or three, fJur to six, and seven to ten communication problems were 84%, 39"/o, 35%, and 10)., respectively. The most commonly reported problems were difficulty understanding pJysicians (49.5%), difficulty expressing feelings (46.3%), desire for more controllver the medical team (45.3%),, and difficulty asking physicians questions (42.6%). ThJse patients who were more distressed, who were less optimistic and who felt more hj less about their disease were most likely to report communications problems. Even though it was not possible to determine the direction of causal relationships, Lerman et al. ( 1993) speculated that one mechanism that might be responsible fur the relationship between J unications problems and psychological distress was that patients who were able to ask questions and voice their needs and concerns were more likely to elicit informa ion and receive personalized advice, which reduced uncertainty and provided reassuralce. They suggested that both patient oriented interventions and physician-orient1 interventions which enhance physician-61

PAGE 75

patient communications may be of value in improving patients' adjustments to treatment. Patient-oriented interventions would include those intetventions that enhance patient skills in asking questions rd expressing concerns. Finally, they recommended that future research inci prospective randOOJized trials which evaluate the impact of inteiVentions to inlrease patient and/or physician communication skills. Not only are patients frustrated 1h physician-patient communications, it also appears that physicians may be just as frulated. Levinson, Stiles, Inui, and Engle ( 1993) undertook a study of the nature ol "frustrating" visits from the practicing physicians' perspective. This study exami ed feelings of frustration with patient visits in 1,076 practicing physicians who were aJending a voluntary workshop on physicianpatient communications. l Analysis from the 25-item questio aire developed specifically for this study revealed several key findings: I. Thirty six percent reported fruTtion in up to one quarter of patient visits. 2. The perceived cause of the frustirating visits was most frequently attributed to the patient in more than 50% of the frustrating visits. 3. The most significant patient-attributed communication problems were: lack f b h .\ . h o trust or agreement y t e pattent, patient presentmg Wit too many problems, lack of patient adherelce, lack ofunderstanding on the part of 62

PAGE 76

both patient and physician, r or controlling patient, patient had special problems (alcohol, dnigs, chronic pain). Another communication problem by physicians in ''fiustrating visits" was a feeling of distress on their part during or rrer seeing the patient. For each of the seven communicalion problems identified by the physicians, Levinson et al. (1993) discussed how could be addressed by developing communication skills that are oriented to lard a more -centered, collaborative, mutual participation model of clinical1. Chan!cteristics of this model incloded: use of open-ended questions, allowing the patient to talk uninterrupted at the outset, active listening, repeating and summarizinl the patient's problems, avoiding the use of medical jargon, and involving the patient 1 decision-making about treatment options. They particularly noted the danger of the tendency for physicians to blame the patient; and, concurrently emphasized the impojce of discussion and education for h nhan. h I p ystctans m e cmg p ystctan-pattent commurucattans. Wbite, Levinson and Roter (1994) Lovided insight into one of the most I common challenges encountered by practicing physicians and one of the "frustrations" identified by physicians in the Levinson et (1993) study: the "oh by the way ... phenomenon which occurs at the end ofthJ medical visit. The objectives of the White I et al. ( 1994) study were to define and describe the physician-patient communications during the closing phase of the medical visiJ and to identifY the types of 63

PAGE 77

communication during the visit that migh have contributed to the introduction of a new problem at the close of the visit. Audiotapes of 88 outpatient visits with 20 primary care physicians were analyzed for communication patterns using the Rater Interactional Analysis System. I The patients were predominantly white, highly educated (80% had some college education), females (61%). I' Closure was initiated by physicianj 86% of the time and physicians asked whether or not there were any more quesJons 25% of the time. In 21% of the closures, patients presented new problems not previously discussed. Both physician and patient behaviors throughout the visit were associated with the presenting of new problems during closure. The specific ph1sician behaviors associated with the reduction in the occurrence of new problems at closure were: orienting the patient to the flow of the visit, giving patients more about the treatment plan, and asking about patient beliefs. The specific pltient behaviors associated with reduced occurrence of new problems at closure w4: talking more about their therapy and sharing emotions and feelings. the findings of this study are closely related to the observational studies by B+ and Frankel (1984) demonstrating that patients are interrupted on average 18 seconds into their opening narrative; and, the first problem presented may not be the \rirnary problem. 64

PAGE 78

Similar to other suggestions and odels for a more participatory, collaborative, patient-centered approach to medical cJe, White et al. (1994) proposed a set of communication skills to use before and luring closure which may be helpful to physicians in decreasing the frequency oi new problems presented at closure. Before closure: orient patient to the flow of the visit, address emotions and psychological issues, explore patient beliefS, check for Lerstanding, ask if the patient has additional questions. During closure: orient the patilnt to the closure, provide a summary of the visit, complete a final check fur undemalmng, clarify the plan, provide encouragement I and reassurance, and consolidate rappo,caring. Finally, a noteworthy practical ooservation was made regarding the concern with time pressures in the everyday practJce of medicine. These researchers pointed out that instead of asking open-ended qulstions and whether or not the patient had additional questions during closure, it J be more efficient to ask these questions during the visit and before closure begins. Implicitly embedded in this observation is the possibility that by using the aforementioned communication skills, overall clinical efficiencies in terms of total encounter Je may be achieved as well. Doctor-Patient Communications: Effects on Outcomes We have entered into an era of in health care where the ultimate purchasers of care (government and emplbyers) are placing emphasis on value 65

PAGE 79

enhancement. The notion of value enhan ment recognizes that the cost of delivering medical care and the associated resource can be reduced even as quality is improved. Because quality and value enhbcement may become central to provider differentiation and competitive advantage in the market place, the importance of providers understanding the impact of phlsician-patient communications on quality, value enhancement, and outcomes is paramount. Different dimensions have been ujd to define patient outcomes: satisfaction, clinical indicators, compliance/adherence, fuality oflife, health status, patient-derived measures of functional status, etc. The puwose of this section is to review some of h fd . .1 d" t e unpacts o octor-pattent commurucattons on vanous outcome tmenstons. In their review of the literature, et al. (1995) noted some of the more salient findings regarding the relationship patient satisfaction, compliance, and halh h.. I .. e t outcomes to p ystetan-patient commurucations: Patients are frequently dissatisfied \. th the infonnation they receive and physicians tend to underestimate plent desire for information. T -C. b "d I "gnifi 1 I ed .uuormation-gtvmg y proVl ers to patients was s1 cant y re at to pattent satisfaction. Time spent in discussing preventive care was positively associated with patient satisfaction. More time spent in the medical interview was positively associated with patient satisfaction. A more dominant and controlling physician communication style was associated with less patient satisfactibn. 66

PAGE 80

Non-verbal behaviors such as freguent eye contact, appearing interested etc were positively associated with satisfaction. Positive associations have been nLed between patient-centered care behaviors and patient satisfaction, complianbe, resolution of patient concerns. C li hasb d I h :-: d omp ance een assoCiate Wit more uuormatton-gtvmg an postttve talk. Patient-centered behaviors were related to higher self-reported compliance More patient controlling less physician controlling behaviors and more information-giving by physiJians in response to information-seeking by patients were related to improved patient health status-clinically, functionally, and subjectively. In cancer patients, lack of information seems to be related to increased coping problems, depression, anxiety, and uncertainty. In her commentary on the effects of the relationship on the quality of health care, DiMatteo (1994) denoted sJar findings: I Patient satisfaction with their physician's behavior has been positively associated with the continuity of tHe relationship. More adequate histories have been associated with higher patient satisfaction and less delay in reporting symptoms. Enhanced perceived personal call over heahh with collaboration and TBDS. Increased patient responsibility for f.edical decisions have been related to lower likelihood ofblaming the ph};sician if things go wrong. Increased adherence by patients mL physicians provide more information Physician interpersonal characteriJcs and non-verbal communications have been associated with patient health behaviors. 67

PAGE 81

Patient outcomes can be effected by positive expectations about efficacy by both the physician and the patien Improved recovery from surgecy, I decreased use of pain medications, shortened hospital stays, improved clinical improved symptoms, and enhanced management of chronic conditionk have all been associated with effective physician-patient communication. Roberts, Cox, Reintgen, Baile and Gibertini (1994) studied the influence of physician communication on the psychollgical adjustment and decision-making of I 00 post-surgical breast cancer patients. The Leer Diagnostic Interview Scale (CDIS) was used to assess the patients' perceptiot of their surgeons' behavior during the discussion in which they were informed of their cancer diagnosis. The SCL-90-R Scale was used to assess psychological stabs. Results from the CDIS assessment revealed that the physicians' caring attitude was perceived as the most important factor related to psychological adjustment while information giving was a much weaker fador. Those physician behaviors most closely associated with a caring attitude were: "I fllt my doctor cared about me as a person", "my doctor understood my fears and coni\ s" and "my doctor is a warm and caring person". The authors concluded that the u of psychotherapeutic techniques by I physicians can have a significant positive effect on a patient's psychological well-being. Based on this research and that froJ the social sciences literature, they provided guidelines for what the surgeon sJould do in conducting the cancer I diagnostic interview: select an appropriate time and place with the ideal being a 68

PAGE 82

prearranged return office visit, verbalize empathy with direct eye contact, allow the patient sufficient time to absorb the news and react emotionally, provide information about the cancer and the treatment optiJns, engage the patient in decision-making about treatment. In a very comprehensive study, Fox and Stein ( 1991) utilized random digit dialed telephone interviews to investigaJ the effect of physician-patient communication on mammography u4on in a random sample of 1,057 black, Hispanic, and white women (mean age ot53 years). Seven questions from the d =-c. I ab h uali f h mternew were use to secure uu.ormatton out t e q ty o p ystctan-patJ.ent communication regarding mammography! Results revealed that 71% ofHiJanic women had never had a mammogram and that more than half (55.3%) of the 1spanic women over age 50 years had never had a mammogram. The most important of whether women of all racial groups had a mammogram, at any time o1 within the last year, was whether their doctors had discussed mammography witj them. The data also suggested that wit, the physician-patient communication process, physicians were not effectively communicating information about screening and were discussing screening less frequJtly with Hispanic women than with white or I black women. The authors offered several possible explanations for this. First, fewer than half of the Hispanic women had clinical breast exams that reduced the r

PAGE 83

opportunity for mammography referrals. Second, physicians were not using pediatric visits for their children to offer family c,e. Third, Hispanic women tended not to view the doctor as the most believable sdurce of health information. Fourth, Hispanic women cited more barriers to and anxil about the use of mammography than did black or white women. Finally, may have created communication difficulties. With the rapid population 4wth and purchasing power of the Hispanic population in many concentrated areas U.S. (Los Angeles, Houston, Phoenix, Chicago, New York and others), the implications for provider organizations are ,apparent. In summary, the research reviewed here has provided evidence that patients, and to some extent physicians, have been tstrated with sevend aspects of doctor patient communications and the physician-patient relationship. These fiustrations and dissatisfactions have been borne primarily out of traditional doctor-patient relationships characterized by a Parsonia.Q., physician-centered, paternalistic model whereby the patient has played a predo1antly passive, dependent role and the physician has assumed an autonomous, professionally dominant and directive role in the health and medical care decision-mJg process. It is clear from the evidence preseJted here that patients as well as physician and non-physician researchers continue to rvocate a physician-patient relationship and communication style characterized by informed collaboration, mutual 70

PAGE 84

participation, and patient-centeredness. Most importantly, it is also clear from the evidence presented here that such awthes can have profuund effects on societal outcomes of interest: enhanced patient satisfaction, improved clinical and functional health status for patients, cost efficienciJs, and possibly, enhanced job satisfaction for physicians. Given these implications, Ong et al. (1995) noted that an important goal would be to establish a systematic theory and odel of doctor-patient communication. Such a theoretical framework would be useful for: 1) providing a systematic, organized approach for improving physician-patient communications and relationships; and 2) generating hypotheses regarding such relltionships. Thus, the purpose of the I following section is to review some models of doctor-patient communications and relationships which have been champion1 by researchers and practitioners. Collectively, these models may point us tbward the theoretical framework and practical models needed for advancing in this area; and subsequently, for the successful reengineering of doctor-patient communications and relationships. Models for Physician-Patient Communication and Relationships l In creating interpersonal relations ps, there have been different opinions on how to define and develop an optimal inJrpersonal relationship in the physicianpatient dyad. As early as I 956, Szasz and Hollender described three types of

PAGE 85

physician-patient relationships based largely on the degree of control required by each participant for certain medical circumsJces: activity-passivity, guidance-cooperation, I and mutual participation. According to Ong et al. (1995), some believe it is primarily a social relationship; some claim that the is detennined by its therapeutic qualities or "client-centeredness;" some 1escribe the "ideal" as one which integrates both patient-centered and physician-cent Ired approaches; while others advocate a "mutuality" in the relationship. Several u eful models have been advocated for structuring the optimal doctor-patient rei tionship. von Friedericbs-Fitzwater a!. I) noted that the term "patient-centered medicine" was first introduced by Balint, Hunt, Joyce, Marinker, and Woodcock in 1970. This was a contrasting model to thl traditional illness-centered/disease centered/doctor-centered model. Patient-lentered medicine has been characterized as follows: In the patient-centered metrod, the physician attempts to Ascertain the patient's agenda ... the essence of this method is that the physiciay tries to enter the patient's world, to see the illness thrpugh the patient's eyes. This is done by behavior that in'fites and facilitates openness by the patient. The doctor'ls goal is to understand the patient's expectations, feelihgs, and fears .... bringing the patient's problems into the Wagnosis. (von Friederichs Fitzwater et al.,199l;p.I8) von Friederichs-Fitzwater et at. (1991) po ted out that there has been no operational definition of the model. T

PAGE 86

In 1977, Engel advocated a biopsychosocial model of disease and healing which viewed disease as an open syst4 process that incorporated mind, body, and environment. This model was also in cohtrast to the traditional mind-body dualism prevalent in the traditional paternalistic tomedical model. Keller and Carroll ( 1994 ;p.131) asserted that, "While EngJl' s biopsychosocial model of disease and h alin ha ed gnifi I. d h d 1 e g s recetv s1 cant attentton m pnrnary care as an 1 ea, 1t as receive ess attention as a guide for physician behaviL." Emanuel and Dubler (1995) havl characterized the "ideal" physician-patient relationship as one that encompasses: chbice, competence, communication, compassion, continuity, and no conflict lfinterest. Their "ideal" tilted in favor of a d ll b I a1 h f more pat1ent-centere co a orattve, mutu part1crpatton approac : m tenns o choice, they have advocated for an infoL patient who is provided choice among treatment alternatives; in terms of commLcation, they have advocated for listening, understanding preferences for infonnatiol and involvement in decision-making, assisting patients in dealing with the illnJs experience as well as the disease, and a facilitative role for physicians in guiding Ld directing patients through the illness experience by recognizing values and feelgs as well as life, family, and job circumstances. From a legal perspective, lhey also pointed out that this type of communication style can allay adverse evlnts which may eventually lead to malpractice litigation. In tenns of compassion, they eLphasized the importance of empathy; in

PAGE 87

tenns of continuity, they acknowledged he investment of time necessary for d hr d. I f nfl. f prevention an c orne 1sease management; m tenns o no co Icts o mterest, It was expected that the patient's well-being tJe precedence over the physicians' own personal interests. Keller and Carroll (1992) have identified three perspectives which they believe influence the physician-patient relationf: the physicians' philosophy of disease and health, the role which the physician assuies in the relationship, and the communication strategies employed by tlie physician. They have advocated the "E4" model for physician-patient communicatilns. This model goes beyond the traditional biomedical tasks of"find it and fix it" to fne which unites the communication tasks of and with the biomedical tasks of find and fix. Like the model of Emanuel and Dubler (1995),\ this model has favored a non-paternalistic approach to physician-patient communicahons that emphasizes physician sensitivity to I patient's preferences for infonnation and involvement in decision-making. I Similar to Emanuel and Dubler (1995), Delbanco (1992) has advocated for a more patient -centered view of doctor-Jent interperoonal relationships. His approach includes seven dimensions oft which should be considered when dealing with each patient as a unique individual: respect for patient's values, preferences and d ed d d I din d. f expresse ne s; commurucation an e uo1ation; coor ation an mtegrat10n o care; physical discomfort; emotional support and alleviation of fears and anxieties;

PAGE 88

involvement of family and friends; continuity and transition. Delbanco has asserted that this approach should be: woven intt the clinical encounter in bits and pieces until practitioners become more comfortable rth it; incol]lOrated into the early years of training in medical school; adopted by patients to guide encounters with physicians; used to guide quality of care feedback Jom patients. DiMatteo (1994) proposed a based on collaborative TBDS. This model was characterized as one which includes mutual participation; explicit sharing of and expected outcomes by both patient and physician; information exchange about various treatment alternatives and their Jotential effects; and negotiated treatment decisions. To effect this she prol\ eded to identify the required physician and patient behaviors: establishment of a con ucive atmosphere; allow patients to tell their stories; physicians respond to patient's ilLess experience with concern, empathy, edi d. h I f k mun acy, an mterest; p ys1c1an encouragemen t o patients to as queStions; physician explorationofpatient goals, ojtcome expectations, values and quality oflife components; patient and doctor in considering various health management strategies; physicians should assist patienl. in organizing I information given to them; and patients should be allowed sufficient time to make decisions. Ballard-Reisch (1990) advanced aJtheoretical model of participative decision making for physician-patient interactions. This model has three phases that guide the 1/5

PAGE 89

doctor-patient interaction: the phase; the exploration of treatment alternatives phase; and the treatment decision, implementation, and evaluation phase. In tum, each of these phases includes se1ral stages. The diagnostic phase includes Jonnation gathering and infunnation interpretation by both patient and physiciL. During this phase, the nature of the 1 hi b thd d .I 0 0 ed Within" thi 0 re attons p etween e octor an pattent ts negottat s negottatton, there are four relational options: patient ,bdication, patient autonomy, collaboration, or termination. Ballard-Reisch emphasized that collaboration is the ideal model for . d . makin Th 1 I f al h parttetpauve eetston g. e exp oration o treatment ternattves p ase included the following stages: exploratioJ of alternatives and criteria, establishment of al d ghin f al I Th d ternattves, an wet g o ternattves cntena. e treatment ectston, implementation, and evaluation of impleJented treatment (phase ill) included the stages of: alternative selection, decision lplementation, and evaluation of implemented treatment. Each of the thrl phases incorporates outside input from other professionals, fiunily members, and tpport groups. Ballard and Reisch (1990) emphastzed that the physician behaviors most crucial to this model include: information 1haring; gathering information about the patient's goals and expectations, values, blliefs, lifestyles, and sociocultural relationships; active listening and providini feedback about understandings and interpretations; involving family members kd friends where appropriate; and 76

PAGE 90

collaborative decision-making. The aut ors contended that viewing the doctor-patient relationship as a three-phase process wJch progresses through eight distinct stages of decision-making: 1) provides a structuJd framework upon which physician-patient communications can be improved; and tenders decision analysis researchers with a framework for analyzing the quality of Jedical decisions. F redrikson ( 1995) has proposed r input -process-outcomes infonnation exchange model of medical consultation.[ Central to the model and to the core of the process component is infonnation exchange which, like Ballard Reisch's model (1990), emphasizes a very flexible, partijipative flow and exchange of information between doctor and patient in regards to diagnostic and treatment information and involvement in decision-making. Both lodels have embodied input, process, and outcome components. The input component incorporates both patient and physician input. From the patient's perspective, this includes: :frame of reference; motivations, goals, needs, expectations, and personal infonnation. irom the physicians' perspective, this includes: motivations, goals, needs, expectations, and medical information. The outcomes component is linked to the component and encompasses specific patient satisfaction outcomes related to rabors of patient understanding, f d h .. I I. hi appropnateness o treatment, an p ys1etan re at1ons p. 77

PAGE 91

Kreps, O'Hair, and Clowers (1994) provided a conceptual model of communications and how communicatJ s influences outcomes. They have referred to this model as the Transformation Model of Communications and Health Outcomes. They've pointed out that this model is b sed on the systems theory transfonnation model of input-process-output. In systeLs theory, inputs are the antecedents which provide the raw materials for the procesles necessary to accomplish the outputs. In I their model, antecedent conditions included health problem or risk and consumer attitudes, beliefs, and expectations. The component encompassed the communication activities engaged in by patient and the physician: provider/consumer message strategies, tge used, non-verllal cues, channels, and media. The outcomes of the communicatiion and medical care activities comprised the output component of the model: cognitivl, behavioral, and physiological outcomes. Finally, Ong et al. (1995) provide1 a theoretical framework which is very to the inputs-processes-outputii models described previously. Their model mcludes three components: a) backgroun; vanables, b) actual content of communication, and c) patient outcomes. Included in the background variables component are: culture, doctor-patient relationship, types of patients and doctors, di h . c d h lctual f . sease c aractensbcs. ontame m t e l content o commurucat1ons component were the actual communicative behaviors including both instrumental and affective behaviors. Entailed in the patient outcomls component were short-term, 18

PAGE 92

intermediate, and long-term outcomes including: satisfaction, compliance, recall and understanding of information, health sta I s, and psychiatric morbidity. Diffusion of Innovatio : Constructs. Components. and Determinants Two of the primary aims of this ry were to: I) advance our understanding about how physicians view enhanced participation of patients in medical decision making through the integration ofTBDS services, and 2) glean key insights on the implications for working with physician Iganizations in optimizing the integration of such services into medical care delivery s, stems. Since TBDS and the involvement of patients in playing a more informed and Jctive role in the medical decision-making process can be considered a "new" innovbon in the delivery care services, the framework for studying the diffusion lf innovations may assist in the fulfillment of these aims. TherefOre, the purposes ofJ section are to: review the basic com ponents and premises underlying the Jsion of innovations, identify benchmark rates fur the adoption of new innovations and, there possible, delineate factors/determinants influencing these rates. Components ofDiffusion What is the "diffusion of innovati0n" and what are the major premises underlying this concept? Diffusion has beln defined as: 79

PAGE 93

The process by which an innovation is communicated through certain channels over time among the members of a social system. The main elements are the innovation, communication channels, time, and the social system. .. An innovatioh is an idea, practice, or object that is perceived as new ty an individual or an organization. (Rogers, 1995, p.I0-11).1 Innovations in health care include such things such as new drugs, treatment guidelines, preventive services, procedures, diagnolics, therapeutics, and technologies. The rate of adoption of an innovation over time + been consistently demonstrated to fullow an S-shaped curve among various audiences including physicians, public health administrators, teachers, and fanners (Rlgers, 1995a). I This epidemiological S-shaped arption curve characteristically has an early slow phase involving a few "innovators d "early adopters". This is followed by an accelerated middle phase whereby the rJe of adoption usually takes off following adoption of the idea by the early adopterl. The rate of adoption in the final and third phase is slow ending in incomplete In association with this adoption curve, Ryan and Gross (1943) have described the different populations of adopters within a given social system. They are Lcterized in five different categories. I The "innovators" generally approximately 2-3% of the group. They have been characterized as: well endowed with resources, venturesome, risk taking, fascinated with novelty, willing to expenlent and transcend geographical and social boundaries to learn. Rogers (1995a) refels to them as "cosmopolite". Locally, they 80

PAGE 94

are generally somewhat disconnected socially, and are generally not the opinion leaders. They play an important role by Jerving as the gatekeeper for the flow of new ideas into the system. The "early adopters" generally r around 12-15% of a group. They do talk with innovators and selectively new ideas they would like to try out. Locally, they are highly regarded for mlng prudent decisions in the adoption of new I ideas and they are socially well connected. They are regarded as significantly influential opinion leaders with the "early majority''. For these reasons, they are often utilized by change agents as the local evangelist for accelerating change. The "early llllliori!y" comprise rtely one third of the group and are characterized as very local in their perspectives, more focused on current needs and I local practical problems, and do not hold in high esteem lofty conceptual ideas, science, or theory. Their innovation-deci1ion making process has been characterized as very deliberate and therefore slower thL that of the innovators and early adopters. They generally have a high state of readinlss to hear about new innovations that are relevant and practical for solving local pcJblems from peers they know well and trust. They look to the early adopters for indijons of what is prudent to try in terms of new innovations and serve as an importanl interpersonal network link between the early adopters and "late majority". 81

PAGE 95

The next group, referred to as tlie "late majority'', comprise approximately one are more conservative in their views an distrust "cosmopolite" sources of knowledge. They look primmily to the early majo1 for their learning channels, local "proof', and cues in adopting new innovations. The late majority do not adopt new innovations until they become the status quo and p1 pressure eventually forces their adoption. Finally, Ryan and Gross (1943) define the last 16% as the "laggards". They are the ultra-conservative traditionalists L are the last to change and to adopt new innovations. They are isolated in their 4cial networks and are generally suspicious of change, change agents, and innovations. The label given to this group may be an iJ1iustjce. Even though slow to adopt 1 innovations, this group is often useful and valuable in making wise, systematic, calcillated choices based on a rational view of their own limited economic resources. Adoption Rate Determinants Rate of adoption has been defin1 as, "The relative speed with which an innovation is adopted by members of a social system and is measured by the number of individuals who adopt a new idea in a spJified period" Rogers (1995a, p. 250). He has identified at least five variables effectJg the adoption rate of innovations. 82

PAGE 96

First, the rate is influenced by hlw the potential adopter groups perceive the attributes of the new idea. Rogers pp. 250-251) emphasized that those innovations that have the following are more likely to see rapid adoption when compared to those innovations thal do not have these attributes. I) "Relative advantage the degree to Lhich a new idea is perceived as better than the idea that it supersedes. 2) Compatibility the degree to which an innovation is perceived a consistent with the existing values, past experiences, \and needs of potential adopters. 3) understand and to use. 4) Trialabilitythe degree to which an innovation can be experimented with on a limited basis. 5) Observability the degree to which t1e results of an innovation are visible to others." Rogers (1995a) pointed out that 50-900/o of the variance in adoption rates can be explained by these five attributes. As we b see later, these attnbutes have indeed been demonstrated to be influencing facto s on the diffusion of innovations in health and medical care. 83

PAGE 97

A second variable influencing the rate of adoption is the decision-making process I itself. In the process of adopting a new innovation, individuals go through a decision\ making process that includes the following stages (Rogers, 1995a): 1) knowledgeI becoming aware that a new idea exists; 2) persuasion-developing a favorable attitude about the new idea; 3) decision -deteLg whether to adopt the innovation; 4) implementation -putting the new idea 1o use; and 5) eotifirmation -evaluating the innovation for the purpose of continued adoption, modification, or discontinuance. Interestingly, this typology is conceptujy congruent with the transtheoretical model of behavior change that includes contemplation, preparation, action, and maintenance stages and DiClemente, 1983). According to Rogers (I995a), Jr additional variables influencing adoption rates include: 1) the nature of the communication channels within a social network, 2) the nature of the social system throuJ which the innovation is diffusing, and 3) the degree to which change agents are involJed in promotional activities. He speculates I that when considering all five of these Ties, the diffusion process is driven predominantly by how the attributes are pirceived and how this perception is conveyed through the adopter group's interpersonal networks. In addition to Roger's (1995a) clalsic social system theory of diffusion, Green (1980) has provided another typology thal holds significant potential for assessing factors which influence the diffusion of Jovations. It also provides a framework for 84

PAGE 98

targeting interventions for enhancing the diffusion of a given innovation. In his PRECEDE Green has classified lhese factors into three categories: predisposing factors (physician knowljge, attitudes, beliefs,), enabling factors (skills/competence, resources, practice I\ aracteristics), and reinforcing factors (reimbursement, peer support, positive eedback, enhanced self-esteem). The application of this typology will be explJred in a later section examining the diffusion of preventive services. Diffusion of Innovation in Health Care The gap between medical science and medical practice (Greer,l994) as well as the gap between best achievable practice1 and actual practice (Berwick, 1998) is well established. Many medical innovations (nlw drugs, treatment guidelines, preventive services, procedures, diagnostics, and technologies) may take yean or even decades to reach widespread adoptien. Ford, Kalumy, and Sondik (1990) have poignantly noted that the time involved tJ achieve the full transfer of new treatments continues to be problematic in medicine. Even though an enormous body of scholarly research exists on the diffusion of innovations, it is still not a process. A practical standardized process for success has not yet been codified. Thlrefore, it is not surprising to find wide variation in the rates of adoption for new in health and medical care services. The 85

PAGE 99

purpose of this section is twofold. The first is to review, where available, rates of adoption related to new drugs, practice and services. The second is to distill those that have been demonstrated to predominantly influence diffusion and rJes of adoption. We'll begin by first examining one of the earliest and most classic sociometric studies on the diffusion of medical innovations. I Using pharmacy prescription Jrds with information about the month in which each doctor adopted the new dru4 Coleman, Menzel, and Katz, (1966) studied the spread of an antibiotic (tetracycline) Jmong physicians in four lllinois communities. They also collected data from physician r1spondents regarding other physicians in their community with whom they had discussl the new drug. This inlormation allowed the researchers to trace the social networks trgh which the innovation was diffused. The commercial sponsor of the study, ler, provided academic detailers to extensively promote the new drug. This innovation spread rapidly reaching almost complete adoption in 17 months. Physicians reported that the acadLc detailers had little direct influence on their decision to adopt. Instead, the resejchers discovered that information about the new drug spread primarily by word of molth from physician to physician and, generally, a physician did not adopt tetraJcline until a respected peer had shared his/her experiences with the new innovatiol. Greer (1994, p. 434) insightfully noted so

PAGE 100

that what is usually forgotten in this study is that the best predictor of general adoption was not adoption by the physicians who were recognized as opinion leaders, but "adoption by those physicians who with the most other physicians the most often." She emphasized that, "the more people talk and stimulate further talk, the more there is opportunity for consensus, practical problem solving, and social pressure to change." In another prescription drug study, Peay and Peay (1988) examined the adoption process fur temazepam thro, interviews with 124 specialists and general practitioners. They found that within 13 months after its release, 71% of the physicians were familiar with the drug ( time between release and awareness was about 17 weeks), 48% had prescribed it (average time between its introduction and first prescription was 5.5 months), j 27"/o preferred it to the alternatives. I In contrast to the Coleman, and Katz (1966) study, they reported that contact with academic detailers was the Jost consistent predictor of favorable reception by physicians. Sixty-eight p4 reported that their first news came from commercial sources, 40% citing the in particular. Only 28% reported discussing this new drug with other doctot before deciding to prescribe. Eighteen percent reported colleagues as the most important information sources in their first decision to prescn"be. Doctors who had mte frequent contact with detailers (at least once per week) were aware of the drug ear;lier, prescribed it earlier, and were more 87

PAGE 101

likely to prefer it to alternatives than those who reported less contact with detailers. On the other hand, physicians who cited professional sources as their first source of infonnation became aware of it later, prescribed it later, and were more likely to prefer the alternatives than those who did not. l tenns of practice characteristics, there was d ha 1 I diffi ed fr h no evt ence to support t t so o practttilners er om group practitioners m t e reception of temazepam. Peay and Peay (1988) emphaticall noted that there was little evidence to support the Coleman, Menzel, and Katz (1966) depiction of innovators, early adopters, and physician communication n tworks. They contended that their data h th d. I infl h b were more consistent Wit ose stu Ies reportmg an uence on p ysicians y sources and factors external to the classicjsocial system theory. In this case, they contended that there may be a relationshi 1 between therapeutic risk and adoption rates. That is, since temazepam was perceived as a low therapeutic risk, physicians may have been willing to relinquish evaluation in discussions with colleagues and prescribe based on infonnation providld by the detailers. Interestingly, Coleman, Menzel, and Katz ( 1966) noted that detailJs were most often identified as sources of \ infonnation just after the drug became available, but closer to the point of adoption, more professional sources (such as colleaJes) were predominant. Peay and Peay (1988) appropriatel1 concluded by acknowledging that the adoption of new drugs is an extremely complex process and that the consistent 88

PAGE 102

prediction of which doctors will adopt new drugs has thus far eluded those in the field. Adoption rates may indeed be influencld by the interaction of many factors including: h.. 'ct/ ... hill h .. p ystctan conta commurucatton Wit co eagues, age, practice c aractenstlcs, perceived risk of the drug, and academil detailers. Lomas (1991} noted that there le more that 60 organizations in the United States and at least ten countries that technology assessments of new medical ideas, use some form of consensus devellpment in their deliberations, and pUblish practice guidelines. The predominant l of all this activity is to close the gap between the science of medicine and the bractice of medicine, between achievable best practices and actual practice, between wJat is known and what is actually done in delivering health and medical care. UltJtely, to influence physicians in improving the quality of care. Stross and Harlan ( 1979) conducted one of the earliest studies assessing physician awareness and knowledge aboJ the outcomes of a major clinical trial. They surveyed (82% response rate) 228 family and internal medicine physicians to examine their knowledge of the results of l clinical trial demonstrating the efficacy of photocoagulation in diabetic rjinopathy. The preliminary results of this clinical trial were published in April1976. Stross and Harlan conducted their study in 1978, two years following the publication of the clinical trial results in the American Journal of Ophthalmology. 89

PAGE 103

Even though 43% of the family Jhysicians were aware that a large clinical trial had been conducted, only 28% reported\that they were aware of the results. Sixty percent of the internists were aware of tie study and the results. When asked to manage two pa,ent problems mvolvmg diabetic retmopathy, only 33% chose the correct treatment both patients. When physicians who were familiar with the study were asked their source of information regarding the clinical trial results, more than two thirds repol that an ophthalmologist or colleague was the primary source. The authors conclul that this study supported the suspicions of many that valuable information from expLve clinical trials was not being brought to th f . h . dl b. ed. th d e attention 0 practicing p ystCians an ras not emg mcorporat mto e every ay practice of medicine. They suggested that professional interactions with influential colleagues may be an effective comrnunicltion channel and a key mechanism for diffusion. Jordan, Burke, Fineberg, and Hanley (1983) studied selected findings in the diffiJsion of innovations in bum care. Thel study approach was to systematically select a set of innovations in bum care and query physicians about their awareness and use of the innovations. Eight treatments wre selected from a National Institutes of Health Consensus Development Conference (1977-78) and classified according to their complOJ
PAGE 104

were randomly selected to receive a poStal questionnaire (sent out in 1979) that asked questions about familiarity, first knowltge, sources of knowledge, and utilization of the eight selected treatments (response rate of86%). After aggregating results for all treatments, the average awareness level was 75%. More physicians were consensus treatments (82%) than low consensus treatments (70"/o) and low coLple
PAGE 105

delivering CME on new innovations at teaching centers and conferences, there may be better results in spreading innovations ,ouglt local outreach initiatives. In addition, such initiatives should take into consideration three important considerations. 1) What innovations are most appropriate to incljde in such local programs? They suggested that high consensus treatments with low Less levels are prime candidates. For high complexity treatments where awareAess is low, the CME effort should be primarily directed at awareness and infolg physicians of these treatments. 2) Which physicians have the most need for Lrormation? Obviously, CME programs should identify and target those physicianl who have low awareness and use levels. 3) How should infonnation about innovatioJs be made available? It was suggested that the most popular channels of influence rJ awareness and use be identified then subsequently used for communication. Second, the authors noted that several factors tend to influence adoption rates: a) the availability of resources, b) the trJmg and skiUs necessary to use the treatment, c) the regulatory environment, ld d) the physicians' conviction (or lack thereof) that the treatment actually works. The implications are obvious. All the aforementioned observations should be taken into consideration when trying to structure more effective CME programs performance improvement strategies within physician organizations. 92

PAGE 106

Hill, Levine, and Whelton (1988) studied the awareness, use, and impact ofthe 1984 Joint National Committee Consenlus Report on High Blood Pressure (JNC ill). The objective of the study was to utilizl the diffusion of innovation framework to examine diffusion of the JNC ill consls report and to determine to what extent its recommendations had been adopted by !hysicians. They conducted a pre-post-survey (response rate of 44% to both in a stratified random sample of Maryland physicians who were likely to be caring for individuals with high blood pressure. The survey was intended to oL information regarding: awareness of the JNC lll one year after its release, channels most frequently used for I receiving the report, use of recommendations in everyday practice, and practice behavior changes preand post-release report. This research team reported that Jne year after publication, even though 62% of the physicians were aware of the 58% reported having a copy available, 81% reported familiarity with the report, lruy 17% reported use in practice. The comnnmication channels most frequently rted for receiving the Consensus Report were pharmaceutical company representatives, professional journals, and professional association mailings. Second, these researchers found that practice behaviors pre-and post-survey were not significantly different. That is, mhre that two-thirds of responding physicians were already practicing in a manner conJent with nine of the ten treatment 93

PAGE 107

recommendations studied. The authors ncluded that this particular consensus report codified rather than changed practice beLvior. This conclusion was remarkably consistent with the findings ofKosecoff, lanouse, Rogers, McCloskey, Wmslow, and Brook (1987) in their study of the effectJ ofthe National Institutes ofHealth (Nlli) Consensus Development Program on phjician practice. Through 1986, the Nlli had conducted 60 consensus development conferences encompassing drugs, devices, diagnostics] therapeutics, prevention, and public health. The primary purpose of this study was to lxamine how consensus conferences have affected physician, knowledge, attitudes ld practice behaviors. This study collected I data from several sources including a physician survey on the knowledge, attitudes, I and practices of physicians as well as a review of hospital medical records to determine changes in actual practice. This study repjrted on the medical review component and was designed to examine to what extent: 11 conference recommendations addressed areas of needed of change, 2) pre-conference practice behaviors were already in high I compliance with the conference recommendations, 3) the recommendations stimulated change in practice where none was occurriJg, and 4) the recommendations accelerated change that was already under way. The Jedical record review was designed to assess the impact of four conferences: 1) thJ treatment of primary breast cancer, 2) the use of steroid receptors in breast cancer, 3) lesarean childbirth, and 4) coronary artery bypass. 94

PAGE 108

For each conference, the researcllers studied medical records from 10 randomly selected acute, non-specialty +spitals drawn from two periods before and one period after the conference, with eaci period lasting nine to 12 months. Time period one was 13-24 months and time period two was 0-12 months prior to the conference. Time period three began Je to12 months following the conference. Compliance for each of 11 recommendatilns selected from the four conferences was studied. Results showed that the conferences mostly failed to stimulate change in physician practice. For six of the II recoLtions, the compliance level remained at less than 50% following the conferencelln addition, compliance was, in general, increasing during the year or two immedilely preceding the conference and compliance failed to accelerate following le conferences. One of the major findings in this study was that the Nll:I recommenltions addressed areas where there was already high confurmity while at the same le neglecting areas of practice that needed improvement. Consistent with the study 1 Hill, Levine, and Whelton (1988), this research tended to support the fact that the consensus reports codified rather than changed practice behavior. The research bt Kosecoff et al. ( 1987) prompted an astute sociological researcher (Greer,1987, p. 2710) to pose the very appropriate question, "Should we be surprised that the national Jonsensus has caught up to the field?" Greer's (1994) own obseiVations have doclmented that decisive change largely occurs 95

PAGE 109

in local communities of practice and receptiveness to change from sociocultural determinants. In this regard, Kosecoff et al. (1987) acknowledged that changes in practice do not inevitably occur from the disseminat+ of consensus reconunendations. Their reconunendations for changing physician iractices in the dissemination of practice guidelines were consistent with diffusion of innovation theory. First, the consensus conferences should concentrate on areas practice that need the most improvement. This generally means that topics shoold partly on the basis of data regarding existing community practice so that both the state of science as well as the state of practice can be taken into account] Second, it should be recognized that li h d I h d h comp ance Wit some recommen attons ,ay reqwre c anges not un er p ystCian control, such as the acquisition of new tools, technologies, resources, and facilities. Third, successful physician behavior changl will require dissemination and l . th 1 11 d based ilizin" tmp ementatton strategtes at are more o""""" commuruty, an practiceut g respected peers for performance feedback Ld face-to-face endorsement. Specifically, to change the state of practice, a local disjmination .and implementation program I nmst meet some minimum requirements.le information must be timely, clinically grounded, and clinically relevant at the local level. The intended audience should be targeted appropriately and assessed for readiness, willingness, and ability to act on the recommendations. 96

PAGE 110

Lomas, Anderson, Domnick-Pierre, Vayda, Enkin, and Hannah (1989) evaluated the effect of a national consenls statement on cesarean births in Ontario, Canada. Like their U.S. counterpart, the Canadian consensus statement explicitly recommended a lower rate of cesarean births. The guidelines were disseminated I extensively in Canada between March and June of 1986 in a mailing to all obstetricians who were on the mailing list of the natioJal specialty society (Society of Obstetricians and Gynecologists) and all hospitals with Lore than 50 beds. The guidelines were published in both the national medical jot and the bulletin of the society. Monthly discharge data from Ontario hospitals of obstetricians before and after the dissemination of the guidelines were used fo examine awareness, attitude, knowledge, self-reported effect, self-reported practice and actual practice. The analysis revealed that one, after dissemination 94% of the obstetricians were aware of the guidelines and 82-85% meed with them, but specific knowledge of the content of the recommendations was pbor (67% correct response). The obstetricians reported learning of the guidLes from multiple sources with the most predominant being the specialty society bjletin (94%), discussion with colleagues (56%) and the Canadian Medical AssociJion Journal (52%). Even though at two years following release of the guidelines 33l of the obstetricians and hospitals reported they luid changed their practices l a result of the guidelines, data on actual practice showed that c-section rates were 15-49% higher than the self-reported rates. 97

PAGE 111

That is, two years after dissemination, p ysicians' behavior had changed very little. An interesting secondary calculation reiarding rate of change was completed by these researchers. They approximated the n ber of years it would take for the overall c section rate in Ontario to reach 10.5%, le approximate rate in England and Wales. At the rate of change generated by the in Ontario, it was estimated that it would take Ontario more than 30 years reach a rate of I 0.5%! Lomas et al. ( 1989) concluded practice guidelines may predispose physicians to considering change, but behaviors are influenced and sustained by many things besides research evidencl. They identified some possible administrative, educational, patient-centled, and/or economic factors potentially infl h . I fl "d lin ed thr f uencmg p ystctan unp ementatton o practice gw e es: percetv eats o malpractice litigation from patients, inadluate skills, economic and socioeconomic d d"ed I din c.ed p mcenttves, an patient-me tat regar g prererr treatment. omtmg to the work of Greer (1987) and Kosecoff 1 al. (1987), Lomas et al. (1989) acknowledged the local and individualized nature of the forces shaping clinical policies. They concluded that ''unless are embedded in a broader program that addresses the need for translation and imJlementation of the guidelines locally'' (p. 1310), clinical policies will be unlikely to lffect rapid change in practice behaviors and I unlikely to effect the ultimate objective ofiimproved quality and patient well-being. 98

PAGE 112

In another study on cesarean births, Lomas, Enkin, Anderson, Hannah, Vayda, and Singer (1991) conducted a randomizld controlled trial with 76 physicians in 16 Ontario, Canada community hospitals to the effect of audit and feedback compared to local opinion leader education on guideline compliance. The practice I guideline was produced and disseminated in 1986 in collaboration with the Society of Obstetricians and Gynaecologists ofcla. The guideline explicitly recommended clinical actions to increase the trial of lablr and vaginal birth rates. The unit of randomization and intlrvention was the community hospital. Eight hospitals served as a control group and Jceived a copy of the practice guideline. Four hospitals service as an intervention group for audit and feedback (A/F) on physician perfonnance regarding compliance with tl!le practice guideline. Four hospitals served as an intervention group for the use of lojally selected educationally influential opinion leaders (OIL) who utilized a "detailing" in furmal and infurmal interactions with peers. For the experimental groups, lhe duration of the active interventions was 12 months. After 24 months (1988 and 1989), the trial oflabor and vaginal birth rates in the AIF group were significantly differnt from those in the control group. However, the trial of labor and vaginal birth rates for the OIL group were significantly I higher than the control group, 46% (P = .007) and 85% (P = .003), respectively. I Duration of stay was significantly (P<. 001) lower in the OIL group. The mean c-99

PAGE 113

section rate for the OIL group dropped b;y 11.5)0/o from the pre-study year of 1987 to the end of the two-year study period whJreas in the control and the AIF groups, the mean rate increased by 4.7% and 6.7%, rlspectively. Quality of care (measured by maternal deaths, infant deaths, complicaJons, and Apgar scores) was not "ed" "h ... \Th b thah findin compromts m e1t er Intervention group\. e aut ors note t t ese gs can not be explained by case-mix differences between the groups. It was noted that the failure of thj AIF strategy to create any change ran contrary to the literature. They that differences in methodology (change I agent in previous AIF study acted much like an opinion leader) and research environment (teaching hospital vs. commLty hospital) could be partial explanations for these differences. Also, because physlians in the AIF group would only agree to department level feedback and would not to personal individualized feedback (an essential component for AIF e:ffectiveless ), the power of the intervention to significant change may have been comproLsed. Also instructive from a diffusion plrspective was the fact that even in the OIL group, over 50% of eligible women still ulderwent an elective c-section instead of a trial oflabor. The authors noted that this Lay partly be explained by "patient factors", I another element that has been related to cJange in physician behaviors. Specifically, in follow-up interviews with OIL physicians, lhe most common reason offered for lack of 100

PAGE 114

guideline compliance was the refusal of many women to undertake an offered trial of labor. The researchers concluded that educationally influential opinion leaders with educational support can generate change among their peers when I they agree and prepare to be agents of change. The authors speculated that the I success of the opinion leaders may lie in their ability to take local circumstances and cultures into account in their educational/iommunication activities with peers. At the same time they noted that even four years after the release and widespread dissemi nation of a clinically sound practice guideline, the compliance rate among physicians encouraged by opinion leaders was no bjer than 70%. Given this observation and the aforementioned "patient factors" they Lserted that further advances in the I appropriate use of c-sections may have to be derived from patient education and regulatory or economic incentives that 4 already been proven eJrective. Grilli, Apolone, Marsoni, Nicoluccj, Zola and Liberati ( 1991) investigated the impact of a national education program baled on the dissemination of written guidelines for the treatment of breast, colJectal, and ovarian cancer in Italy. Through a survey of 770 physicians and a review o)lmedical records of I, 483 patients, this study examined whether: 1) the guidelines reached the targeted audience of physicians --those working at community hospitals] 2) they were effective in influencing physicians' opinions, and 3) the guidelines Lere being used in patient care. 101

PAGE 115

Three task furces for breast, color, and ovarian cancer were launched in Italy in 1978, 1979, and 1985, respectively. The distribution process for the breast cancer guidelines included: a mailing to t general hospitals, distribution at some national scientific meetings, availability upon request from the task forces' coordinating centers. The guidelines for I olorectal and ovarian cancer were distributed at scientific and specialists' meetings and were made available upon request from the task forces' coordinating centers. Ofthe 1,874 physicians identified or the study, 770 (41%) completed a valid questionnaire. Compared to respondents tended to be female and younger physicians with radiotherapists ,d medical oncologists responding more frequently than other specialists. The medical records review included care provided to all newly diagnosed breast (742), colorectl (641), and ovarian (100) cancer patients I consecutively seen at one of 45 hospitals from March 1986-March 1987. Compliance with guideline recommendations was by examining how closely patterns of care followed 36 recommendations (12 frdm each disease site). The percentage of doctors awareness of the guidelines were 60"/o, 47%, and 44% for breast, colorectal, and Jvarian cancer, respectively. Additionally, those who were more aware of the guideliles reported that the guidelines were brought to their attention informally by o \er colleagues. 102

PAGE 116

Significant variation in compliance with the 36 recommendations, especially in relation to diagnosis and staging, were lealed: 37% to 89%, 48"/o to 82%, and 10% to 97% for breast, colorectal, and ovarij cancer, respectively. Compliance with the 36 recommendations for treatment range1 from 300/o to 86%, 11% to 90%, and from 18% to 91% for breast, colorectal, and otarian cancer, respectively. This research team concluded thaJ guideline development and dissemination had a negligible effect on cancer care. Jey delineated a set of requirements very much like those specified by Kosecoff et l. (1987) for changing the state of practice. The message must be timely, scientificall1 grounded and clinically relevant. Consideration must be given to the context in which local clinical decisions are made. The appropriate audience must be targetel. The target audience must be both willing and able to act on the information. A novel and insightful study conducted by Grilli and Lomas, ( 1994) investigated the relationship between comlliance rates and key aspects of the specific clinical message of the practice guidelines. All studies in the English language medical . literature published between 1980 and 19i! that provided practice guideline clinical content, compliance rates, endorsement official organizations, and which taJgeted providers were eligible for the study. A total of23 studies addressing 143 recommendations across 70 different medilal and surgical procedures were selected for the study. 103

PAGE 117

For each of the 143 recommendations, the authors categorized them according to specialty area, type of procedure ( dilostic, surgical, etc.), and as being high or low on characteristics thought to influenJe diffiJsion. For diffusion characteristics, the authors adapted the five category taxonoLy proposed by Rogers (1995a): relative advantage, compatibility, complexity, tJability, and observability. Lack of information in the studies did not allow tAe authors the capability to assess the relative advantage and compatibility. Thus, each \of the 143 recommendations were classified as being high or low on complexity, trialability, and observability in a typical clinical setting. High complerity, tria/ability, and observability were respectively defined as when a practitioner with usual training and skills, working in an average setting: 1) "perceives it to be difficult to acquire the lkills for or to understand, or has no direct control over any resource changes requir1 for implementation; 2) is able to experiment with it on a limited basis; 3) is able to obtain timely feedback on any impact of the intetvention on patients' outcomeslp. 204). Overall, the mean compliance these studies was 54.5% with significantly (P < 0. 001) higher compliance rates for the specialty areas of cardiology ( 63.6%) and oncology (62.5%), than those fur preventive care ( 46.2%) or obstetrics and gynecology (43.3%). +mmendations regarding procedures with high complexity had lower compliance rates than those low on complexity ( 41.9% vs. 104

PAGE 118

55.90/o; P=0.05). Those recommendations judged to be high on trialability had a signficantly (P =-0.03) higher compliance rte (55.6%) than those judged to be low (36.8%). Interestingly, this study found no influence of observability on compliance rates. As the authors noted, this was surprising in view of the fact that the outcomes is often cited by physicians as major reason for non-compliance. Multiple linear regression was uscl:l to assess the overall amount of variability in compliance rates which could be explled by all the pre-selected characteristics of the message (data source, area ofpracti1 complexity, trialability). Overall, these characteristics explained no more than 4 7% of the observed variability in compliance rates. The authors arrived at several key \conclusions and recommendations that are instructive for informing and guiding the diffusion of innovations in health care. First, the observed compliance rates suggest thJe is still significant room for performance improvement in the current delivery of care. Second, since only about half of the variance in compliance rates could be lxplained on the basis of diffusion and other characteristics, these researchers that a larger percentage of the variance might be explained if other recognized catLries of potentially influential factors were included in the analysis: the source of the lessage, the channel of communication, the clinical setting, and the nature of the receiig audience. 105

PAGE 119

Third, complexity was the single most effective predictor of compliance. Obviously, recommendations with high ctmplexity and low trialability are more likely to result in low levels of compliance. Grilli and Lomas ( 1994) asserted that the nature of diffusion efforts may differ for high coLplexityllow trialability recommendations I compared to low complexity/high trialability recommendations. That is, for recommendations with high complexity Jd low trialability, diffusion practitioners may I effi a I de akin want to p ace greater ort mto pre zsposmg practlttoners to cons1 r m g changes in their delivery of care before laLching efforts that are more focused on enabling and/or reinforcing strategies. In contrast, for recommendations with low complexity and high trialability, less effo1 may be needed to predispose clinicians to change because they may be already recerve to this kind of change in practice. Instead, "efforts could move more to a focus on strategies such as opinion leaders to enable change at the locallevel'1 (p. 212). The authors went on to acknowledge that from a diffusion perspective, even with a highly implementable message, it is still unclear whether to focul efforts on the providers themselves (audience), the incentives and structure oflhe workplace (setting), or a combination of the two. Lawrence (1990) has examined the diffusion of the U.S. Preventive Services Task Force (USPSTF) Guidelines. He has commented that because of the number of topics reviewed and the variation in the qua.Iity of scientific data supporting the 106

PAGE 120

guidelines, the analysis and prediction o1the diJfusion of the USPSTF recommendations are complex. He goes! on to explain that because the guidelines are directed at an entire population with many subgroups with varying needs, and because some of the recommendations are more jechnical (immunizations and screening tests) while others emphasize behavioral (counseling), such differences significantly influence the rate of diffusion. Notwithstanding, he utilized Green's (1980,1988) PRECEDE model (predisposing, enabling, and reinfurcing ,ehavioral determinants) as a frameworlc for analyzing the aids and barriers to the diffiision of preventive services. Since the PRECEDE framework is complementaryrl o Roger's (1995a) and Greer's (1987, 1994) contention that diffusion is largely sociocultural process; it will be instructive to briefly review Green's (1988) and Lalence's (1990) delineation of these three categories ofbehavioral determinants. Predisposing factors for the diffusion of preventive services included physician: knowledge and attitudes toward services, personal health behaviors, confidence in counseling abilities, and beliefs about their patient's interest in health promotion and disease prevention. EnabJg factors included: physician competence to -C'. h.. .lb c. d. peu.onn preventive servtces; p ys1c1an remt ursement 10r ren enng preventive a practice environment organizj to facilitate the delivery of preventive services; time to provide services; efficient reminder systems; physician perception of 107

PAGE 121

guidelines as being clear, coherent, and scientifically grounded. Reinforcing factors included: peer support, positive feedbacJ and receptivity by patients, evidence of intermediate results such as improved he1th behaviors among patients, and an "enhanced sense of self-efficacy in one's role as a healer." This distillation of behavioral detenninants provides a framework for assessing strategies to enhance the diffusion of preventive services as well as a foundation for the review of additional studies on this topic. Three years prior to the release first edition of the U.S. Preventive Services Guidelines, Lewis, Wells, and (1986) constructed a conceptual model and an associated questionnaire to study r determinants of physician counseling practices. The questionnaire measured the agressiveness, indications and techniques used by physicians in counseling patients smoking, exercise, weight control, and alcohol use. The independent variables asLsed motivations, perceived skills and barriers, medical-specialty, and personal hLth habits. A random sample of male physicils (151) with an average age of 51 years al d "'--=1 I d b representing gener an li:U.IWY practitioners, mtermsts, surgeons, an o stetnetangynecologists from a county medical completed the questionnaire. The questionnaire also assessed the health habits of the physicians. Fifteen percent reported they were a current smoker, 58% reported they did not exercise enough, and 24% reported that drank alcohol about every other day. 108

PAGE 122

In tenns of motivations to 87% agreed that smoking is dangerous to your health, 85% agreed that counseling \was important, and 87% agreed that physicians have an obligation to counsel. Even though 74% indicated they knew how to interview patients, only 21% reported rhey knew how to counsel patients and only 12% felt they were effective in counseling. Physician perception of skill in counseling was a statistically significant (P =0. 05) +ctor of physician counseling behavior for smoking, exercise, and weight control. 'Ilhat is, those physicians who perceived themselves to be more skilled in counseJg were more likely to engage in counseling th d kin I. d gh 1 Thi err patients m regar s to smo g, exermse, an wet t contro s ts consistent with the framework provided by Lawrent (1990) who cited physician confidence in their ability to motivate behavioral change in their patients as a predisposing factor for the diffusion of preventive services. j Over half of all physicians report ; counseling all patients about smoking I (52%), wetght control (56%), and alcoho1 (55%). However only 26% reported counseling all patients about exercise. Forty-four to 55% of physicians said they counseled individuals for less than two jutes with regard to all health habits included in the study. Lawrence ( 1990) cited physician health behavior as a predisposing factor to consider in the diffusion of preventive 1= The most significant finding of this early study by Lewis, Wells, and Ware ( 1986) was that physicians who had poor health 109

PAGE 123

habits did not fully counsel their patients about those habits, and physicians attempting to improve poor health habits counseled significantly more than physicians I who were not trying to change their own behavior. The authors concluded that strategies which fucus health promotion 1rts on physicians may have a nrultiplier effect. That is, maintaining physicians' hjalth through risk factor reduction may I benefit both the physician and ultimately the patient. Schwartz, Lewis, Clancy, Kinosl Radany, and Koplan (1991) surveyed 2610 I members and fellows of the American College ofPhysicians from four geographical areas of the U.S. to estimate internists' usl of immunizations (pneumococcal, influenza, tetanus, hepatitis B), screening jests for cancer (breast examination, Papanicolaou smear, stool occult blood tel), and other diseases (electrocardiograms, cholesterol, chest x-ray), and behaviorallunseling (smoking, exercise, alcohol and seat belt use). In addition, information reJard.ing background information, personal health, and record keeping were gathered. Overall, results revealed that this group of internists used effective preventive interventions less frequently and ineffectivl practices. more frequently than recommended by expert guidelines. Y physicians and those whose practices were primarily oriented toward general intimal medicine exhibited greater compliance with expert recommendations for health 1motionldisease prevention. Attitude and lack of knowledge were associated with the use of preventive services. However, one 110

PAGE 124

of the most significant findings of this stlildy supported the observations ofLawrence (1990}, Lewis, Wells, and Ware (1986).1There was a strong and significant (P < 0.01) relationship between physicians' person 1 health behaviors and their propensity to provide the same services to their patienJs. Furthermore, this relationship. remained persistent for all three classes of health prevention activities studied: immunizations, screening, and for behavioral factors. Like Lawrence I (1990) Lewis, Wells, and Ware (1986), these authors suggested that perhaps an I appropriate strategy for improving the diffusion of preventive services is to target physicians' personal health practices. McPhee and Bird ( 1990) studied the utilization of cancer prevention guidelines in clinical practice as recommended by thl National Cancer Institute (NCI}, the American Cancer Society (ACS}, and the U.S. Preventive Services Task Force (USPSTF). Their objective was to answer three basic questions: 1) Is there a I problem? 2) Why is there a problem? 3) What can be done about it? To answer these questions, these researchers compared thl results they obtained in studying 52 physicians from a university general inteJal medicine practice with other published studies. Their results revealed that there is indeed a problem. Even though physicians tend to overestimate their own performance rates, actual performance rates for cancer prevention guidelines are consistently beldw those recommended by NCI, ACS, and 111

PAGE 125

the USPSTF. Consistent with the behavioral determinants framework provided by Lawrence (1990), these researchers cateJorized the major barriers to implementation into physician factors (physician disagreement with recommendations), patient factors (patient apathy, discomfoJ refusal, inconvenience), test factors (cost, risk, accessibility, uncertain cost and cliniL effectiveness), and health care delivery systems factors ( lack of reimbursement, ILk of equipment/staff, scheduling problems, inconvenience). These researchers advocated inte entions that focus primarily on enabling and reinforcing factors for both physicians an patients as the most promising for enhancing the diffusion of cancer prevention guidelines. Specifically, 1) physician audits with feedback, 2) computerized reider systems for physicians, 3) reminder systems and behavioral counseling/educatiln for patients. Battista, Williams, and MacFarlanj (1990) studied the detenninants of preventive practices with respect to cervical, colorectal, and lung cancer I through face-to-face interviews with 552 fee-for service primary care physicians in the Canadian provinces of Quebec and New BLnswick. They gathered information on over 200 items including: a) physician sociidemographics, (b) practice and organization characteristics, (c) continuing medical education activities, (d) knowledge of risk factors associated r each type of cancer, e) physician beliefs about the efficacy and effectiveness of selec:rted preventive practices, (f) perceived 112

PAGE 126

barriers to implementing preventive services into clinical practice, and (g) reported patterns of adult cancer prevention. sJilar to McPhee and Bird (1990) they grouped the perceived barriers to prevention into I physician-related (lack of knowledge, perceived ineffectiveness of procedures, risk of procedures) patient-related (lack of interest, lack of information, loj compliance, time, and cost), organization related (accessibility, time constraints, Jera11 costs). For regression modeling they grouped these predictor variables for eat cancer type into a typology conceptually congruent with Green's {1988) predisposing, enabling, and reinforcing factors: cognitive, sociodemographic and ocsamlonal determinants. Like Mcl'hee and Bird (1990), Jey fuund that when comparing reported practices to actual practices (billing recofs), physicians overestimated their preventive services practice behaviors by 10%. Multiple regression analysis revealed particular predictor variables for each cancer type. !Specifically, knowledge, continuing medical I education, beliefs regarding efficacy, gender, accessibility, and provider-related barriers were significant (P < 0. 01-0.00 I) predictors of detection scores for breast cancer. The most important determinant of breast cancer detection scores was gender, with female physicians having higher det1on scores than male physicians. I For cervical cancer, physicians with higher knowledge scores and younger aged who practices in high volume urban lenters were more likely to perform Pap 13

PAGE 127

tests according to recommendations. Fo, colorectal cancer, the most important detenninant of practice behaviors was the absence of perceived practice barriers (P < 0.001). Additional significant predilors were knowledge, continuing education, in this preventive activity. For lung cancer prevention (anti-smoking counseling), the most significant (P < 0.001) detenninant was the smoking status of the physician with non-smoking physicians being more likely to provide anti-smoking counseling. Three additional predictors were continuing belir and language (English-speaking). For lung cancer detection, the routine use of chest x-rays in asymptomatic individuals, a I test not recommended by guidelines, was r to be associated With the strength of physicians' beliefs in the usefulness of early detection of lung cancer. I Attesting to the complexity of this issue and the limitations of research I instruments, these authors acknowledged that their regression models could only explain between 5-17% of the total varianJe in practice behaviors. Even though reimbursement per se was not examined in this study, the authors cited previous research they had conducted which disclosed a lesser involvement of fee-for-service physicians in adult cancer prevention than lalaried physicians. These researchers concluded that in fee-for-service rei.mburseLent environments without specific financial incentives for preventive practicesl organizational factors were more 114

PAGE 128

important in shaping physician behaviors. Therefore, in fee-for-service environments, organizational factors (accessibility, timJ constraints) may be fertile ground for interventions aimed at increasing physiciL behaviors for adult cancer prevention. Summary. In sum, the rate ofJovation diffusion in any system is influenced by a complex and often confounding +on of many social, cullwlll, individual, organization, and environmental that may possibly include: Perceived attributes of the innovation: relative advantage, compatibility, complexity, trialability, observability. Stages of decision-making: knowledge, persuasion, decision, implementation, and confinnation. Physician predisposing factors: knowledge, attitudes, beliefs, skills/competence, and personal health behaviors. Physician enabling factors: training, resources, practice characteristics. Ph =-c. c. b I d ys1ctan reuuorctng !actors: reun ursement, peer support, positive peer an patient feedback, enhanced self-esteerh and self-efficacy. Patient filctors: patient-mediated p.l...s regarding preferred treallllent; positive patient feedback, patient receptivity/apathy. Organizational/health care delivery sykem factors: local practice norms, available equipment, technologies, facilities, tmie, scheduling, access, convenience. Communication/information attributJ: the nature of the social system and the nature of the communication channelsiwithin a social network, relevancy, timeliness, clinical and scientific grounding, source of the message, appropriately targeted audiences. Environmental factors: regulatory environment. 115

PAGE 129

Rogers (1995a) has noted that ne innovations bring with them an element of uncertainty. "Uncertainty" in the pracril of medicine seems to be a common road in the health care landscape inextricably Jng the research regarding adoption of new innovations. geographical variations (Wr. 1982), and appropriateness (Brook, Lohr, Chassin, Kosecoff, Fink & SolomoJt, 1984; Brook, Kamber, Mayer-Oakes, Beers, Raube, & Steiner, 1990). To help lesolve this uncertainty, physicians often tum to near peers for further information, subjective evaluation of new innovations, and consensus within the context ofthe local\communi{Y of practice standards, experiences, and norms. "This information exchange abou1 a new idea occurs through a convergence process involving interpersonal networks. The diffusion of innovations is essentially a social process in which subjectively perceived information about a new idea is communicated. The meaning of innovatiJn is gradually worked out through a process of social construction" (Rogers, 1995a, p. xvii). "The more people talk and stimulate further talk, the more there is opportunity for consensus, practical problem solving, and social pressure to change .... frequen.t discussion creates a sense of movement .... The problem of uncertainty is Jsolved by a new consensus" (Greer,1994, p. 434). Thus, closing the gap between medical science and medical practice, spreading innovation, and accelerating the rate ofadlption ofbetter practices may be as much a 116

PAGE 130

function of an organization's social capital as its intellectual or financial capital. The successful adoption of any new innovatiln within physician organizations will require an understanding of how to leverage and maximize that social capital. It would I behoove practitioners who are faced with the daunting challenge of successfully institutionalizing best or better practices rthin physician organizations to be cognizant of these factors and to subsequently account for them in any implementation strategy. Summary and Implications for Research First, there is a substantial body Jr evidence regarding patient preference for information and medical decision making. However, there seems to be a coinciding scarcity of information regarding physiciL' opinions about how active patients should be in seeking information or in plicipating in health and medical care decisions. The proposed research will b1d on the work ofStrull et al. (1984) in assessing physicians' opinions about how medical decisions should be made. Because there was relatively little demographic information characterizing the physicians in the I Strull et al. (1984) study, very little else about physician perceptions regarding patient infonnation seeking and involvement in drsion-making could be extrapolated from existing research. The proposed study will provide additional insight by regressing various sociodemographic factors and prabce characteristics on physicians' opinions ab . I d" al d .I. out patient mvo vement m me 1c ectstons. 117

PAGE 131

Second, TBDS services are expenencing rapid growth and integration into managed care delivery systems. Concurr1tly, there is a dearth of information regarding the opinion of physicians about such services. Even though the opinion of the physician stands as a critical variable in optimizing the ultimate impact such decision support interventions will have t physician-patient relationships, clinical efficiencies, patient outcomes, medical care utilization, and costs, reviews of published literature revealed no systematic assessmJnt of physician viewpoints about these types of services. Nor has anyone tried to ascejain the implications of their opinions on the implementation, utilization, and ultimate pact of such services on health care quality, cost, and access factors. The proposed research will be one of the first to: 1) advance our understanding about how physicians view enhanced participation of patients in medical decisionmaking and the integration ofTBDS servibes into care management; and 2) provide key insights on the implications for worJg with physicians in optimizing the diffusion and incorporation of such services across Jhe continuum of care in managed care delivery systems. I 118

PAGE 132

QUANTITATIVE JTIIODS AND RESULTS Introduction s. h iliz.d I h .. ab urvey researc was ut e to ascertam p ystcians oplDlons out: 1. More informed, actively involjed patients as partners in health and medical care decisions. 2. The impact they believe consumers/patients being more informed and taking a more active partnership role in health d medical care decisions will have on: the quality of the doctor -patient relationship, clinical efficiencies, health outcomes, patient satisfaction, health care utilization, health care costs, and malpractice liability. 3. The incorporation ofmos serces into care management and the impact they believe a TBDS service will have om the quality of the doctor-patient relationship, clinical efficiencies, health ojtcomes, patient satisfaction, health care utilization, health care costs, and malpice liability. This chapter will describe the methods, analyses, and results pertinent to the quantitative portion of this study. 119

PAGE 133

Methods I This section will describe the subjects, the instrumentation (Physician Opinion Survey), and the data collection protocol for administering the survey. Subjects The domain of the physician pop ation studied included all primary care physicians (approximately 859) and speciLists (approximately 1306) who were members of the Rochester Independent P1actice Association (RIP A) and the Rochester Community Independent Pract ce Association (RCIPA). All physicians eligible for the study (N = 2, 165) were ed a self-administered Physician Opinion Survey (POS). Physician Opinion Survey (POS) Both research and theoretical constructs supported POS question development. The questions pertinent to preferences for making decisions and patient desire for information was guitd and informed by a substantial body of survey development and associated patienJ research in this area (Cassileth et al., 1980; I Strull et al., 1984; Ende et al., 1989; Sutherland et al., 1989; Brody et al., 1989; Beisecker and Beisecker, 1990; Neufield J al., 1993; Davison et al., 1995). mo

PAGE 134

The development of questions related to the effects of patient decision and self-management support on various quality, lcost, and access parameters was guided and informed by both research (Vickery et al. 1983; Lorig et al., 1985; Greenfield et al., 1989; Mayo et al. 1990; Dentzer, 1991; Long et al. 1993;) and theory (Lynch & Vickery, 1993; Vickery & Lynch, 1995). Final questions and format fori POS were developed in collaboration with the executive and physician leadership from RCIP A and RIP A. Following the guidelines ofPieterzak et al. (1990) for dlvelopment of new measures, a modified multi-stage pretest and revision L used for the POS: I) pre-test with staff, colleagues, and experts 1iuniliar with the troject; 2) use feedback from Step I to guide adjustments and revisions; 3) pre-test with staff and physicians that are not members of guide final revisions. The final POS is illustrated in A. The survey includes a profile of demographic data and physician practice 1haracteristics. The first section of the POS asked physicians their opinions about mJr infonned and actively involved patients in the health and medical care decision-makihg process. The second section of the survey asked physicians their opinions abJut informed consent and telephone-based 121

PAGE 135

decision support (TBDS) services as wei as some open-ended questions regarding the key challenges and benefits of implemenlg TBDS services. POS Administration and Data Collection Protocol The following data collection prolocol was used for the self-administered POS: 1. Administration The survey was administered through the mail by RCIP A using the standard membership mailing list for RCIPl and RIP A. 2. Mailed Survey Packet Contents The contents of the mailing piece included: (a) a cover letter printed on IPA letterhead that included the name, signatures from IP A executive and physician leadership, IP A IUllllber fur questions about the survey; (b) the POS printed on IP 1. letterhead; and (c) a business reply envelope for returning the survey. 3. Mailing Procedures a) b) c) Initial mailing ofPOS to study subjects June 1996. Physicians who did not ..L surveys within three weeks were mailed a follow-up survey. Data collection stopped six weeks following the mailing of the second POS. 122

PAGE 136

4. Confidentiality-All surveys were returned to the University of Colorado Health Sciences Center Departmlnt of Family Medicine. Each survey was I numbered for the sole pwpose response rates. In accordance with the University of Colorado Healtli Sciences Center Research Protocol for Human Subjects, all responses wlre anonymous and confidential. We obtained li . hrl gh h f h . N unp Cit consent to parttctpate t Gu t e return o t e questtonnarre. o physician was identified by name.IAll analytical infonnation was reported in the aggregate. Analysis and Statistical Methods This section describes the selectioh of the independent variables, the dependent variables, and the statistical tests utilized L analyze the POS data. Sample Description Descriptive statistics for the independent variables were used to describe the physician subjects and their practice charaberistics. Description ofVariables Descriptive statistics and frequen distributions were calculated for each of the 30 opinion variables selected for studj. 123

PAGE 137

Selection of Dependent Variabl sand Psychometrics. From the 41 total opinion variables ascertained by the POJ, key opinion variables consistent with the aims of this research were selected for 1y. For those items where there appeared to be a similar underlying "construct" or "fJctor", inter-item correlations using the alpha coefficient for internal consistency were lalculated. The alpha coefficients for the inter -item correlations are summarized 1 Table 3 .I. Table 3.1. Cronbach alpha reliability c9efficients for composite items ofthe Physician Opinion Survey. I ---Question# (B2C,B2H,B3B,B3D) More Involvement 2 I Question# (B2A,B3!A) Minor Issue Decisions 2 I -Question# (BIA,B3C\ ) Physician Time 2 Question # (B2D,B2(J) 885 0.29 856 0.30 893 0.67 None ofthe correlations met the 1pha of 0.70, the criterion for new scales (Nunnally and Bernstein, 1994). TJerefore, 30 survey questions emerged as the physician opinion factors (dependent varilbles) of interest for this study. A summary I of the 30 survey questions utilized in the final analyses is illustrated in Table 3.2. 124

PAGE 138

Table 3.2 Physician Opinion Survey utilized in final analyses. To ic I Question No. More Infonnation Patients should ask more questions Patients want more infonnation Patients need role in asking questions Patients need role in seeking information More Involvement Patients want more involvement Patients need role in decision making Minor Issue Decisions Who should make minor decisions Patients need role in minor conditions Major Issue Decisions I Patients Need Role in Chronic Conditidns Patient Preference Physician Time Physicians don't have time to involve patient Physician don't have time to answer 1atient Physician Awareness ofTBDS Physician Opinion ofTBDS Reasons Health Plans Implemented TBDS Improve quality of care Improve member satisfaction Reduce health care costs Respond to business community Improve patient outcomes Reduce liability claims Improve efficiency Potential Impact of TBDS Unnecessary phone calls Inappropriate visits Patient demand for services Liability claims Length of clinical encounter Health outcomes Physician-patient relationship Patient satisfaction 125 B2C B2H B3B B3D B2A B3A BIA B3C BIB B3E B2F B2D B2G Dl D2 D3A D3B D3C D3D D3E D3F D3G D4Al D4A3 D4A4 D4A6 D4A7 D4Bl D4B2 D4B4

PAGE 139

Selection of Independent Variables. The independent (predictor) variables were chosen for their potential to influence opinions: gender, age, location of d al f I k, d f rest ency, spect ty, type o practice structure, practtce VISits per wee an percent o patients in a capitated reimbursement pl. Two independent variables were eliminated I from further analysis. "Years in practicer was eliminated because it closely paralleled "age" and added no new infonnation. "Salary structure" was omitted due to low priority and minor interest as a study vjable. Selection of Statistical Tests Logistical Regression Analysis. The descriptive statistics and frequency distributions for each of the 30 depend, variables were examined. This examination revealed non-normal distributions of positively and negatively skewed physicians' opinion about patients being encouraged o ask more questions about their medical care. Approximately 94% of the physicians "agreed" that patients should be encouraged to ask more questions leavinJ the remainder in the "no opinion" (3%) and "disagree" (3%) response categories. ThJefore, where there were dependent variables with highly skewed non-normal distributilns, the seven-point Likert scales were 126

PAGE 140

recoded for analyzing the real differences between any "agree" responses vs. "all other''; and "no change" responses vs. "1 other." For the dependent variables that fd not exhibit an extremely positive or negative skewness, but instead present, a more rectangular or bimodal distribution, the seven-point scales were recoded from seven categories to three categories: "agree", "no opinion", and "disagree." the five-poim scales regarding who should make minor and recoded as ''both" vs. "all other." Table 3.3 summarizes the recoded Likert categories for each of the 30 dependent variables and the frequency percentage for each ote categories. Table 3.3 Recoded Likert categories frequency percentages for dependent variables logistical regression analyses. _Dependent_ Variable #lj __ ..fategories __ Minor decisions (BlA) both (56)/ all other (44) Major decisions (BIB) both (68) I all other (32) Patients want more involvement (B2A) agree (88)/ all other (12) Patients should ask questions (B2C) agree (94)/ all other (6) Physicians can't involve patient (B2D) disagree (56) I all other (44) Patients don't want to make decisions agree ( 46)/ no opinion ( 1 0) I disagree (B2F) (44) Physicians can't answer questions disagree (65) I all other (35) (B2G) Patients want more information (B2H) agree (90)/ all other (10) Patients need role in decision making agree (82)/ all other (18) (B3A) 127

PAGE 141

Table 3.3 (Cont.) Dependent Variable (question #) Patients need role in asking questions (B3B) Patients need role in minor condition (B3C) Patients need role in seeking info Patients need role in chronic conditions (B3E) Physicians' awareness ofTBDS (D1) Physicians' opinion ofTBDS (D2) TBDS to improve quality of care (D3Al) TBDS to improve member satisfaction (D3B) TBDS to reduce health care costs (D3C) TBDS to respond to business (D3D) TBDS to improve patient outcomes (D3E) TBDS to reduce liability claims (D3F) TBDS to be more efficient (D3G) TBDS impact on phone calls (D4A1) TBDS impact on inappropriate visits (D4A3) TBDS impact on patient demand (D4A4) TBDS impact on liability claims (D4A6) TBDS impact on clinical encounter (D4A7) TBDS impact on health outcomes (D4B1) TBDS impact on phys-pt relationship (D4B2) TBDS impact on patient satisfaction (D4B4) Categories (frequency %) agree (88)/ all other (12) agree (72)/ all other (28) agree (87) I all other (13) agree (66)/ all other (34) unaware (26) I all other (74) positive (21) I neutral (44) I negative (16) unaware (18) agree (37) I no opinion (30) I disagree (33) agree (59)/ all other (41) agree (83)/ all other (17) agree (72)/ all other (28) agree (38)/ no opiniC?n (30)/ disagree (32) agree (22)/ no opinion ( 40)/ disagree (38) agree (61)/ all other (39) no change (57) I all other (43) no change (57) I all other (43) no change (58) I all other (42) no change (77) I all other (23) no change ( 69) I all other (31) no change (64) I all other (36) no change (51)/ all other (49) improve (36) I no change ( 46) I worsen (18) 128

PAGE 142

The frequency distributions for each of the 30 dependent variables detennined the selection of either binomial or mulJomiallogistical regression analysis to evaluate the simultaneous contribution of the phJsician demographic and physician practice characteristics ( mdependent variables) J predicting physicians' opinions (dependent variables). This technique is effective in analyses where categorical independent variables are used to predict ordinal dependent variables with non-normal distributions (Wolfe and Smith, 1996). The criterioJ variables were physicians' opinions. Gender, I f d al I f . age, ocatton o rest ency, speci ty, type o practice structure, practice VISits per week, and percent of patients in a capitaled reimbursement plan were used as predictor variables. The predictor variables were tegressed on each criterion variable. Odds ratios and 95% confidence intervals were calculated. All calculations were made using SPSS version 9.0 Open-ended questions. The open-ended questions were coded, segmented, and analyzed using methods of qualitativl analysis described in the Qualitative Analysis Section of Chapter 4. Results This section provides a descriptioh of the subjects, a description of the variables, and logistical regression resultJ for the POS. 129

PAGE 143

Sample Description Relevant subject characteristics are presented in Table 3.4. A total of914 clinicians retwned useable questionnairJ, a response rate of 42%. Table 3.4 Phymcian and J.!'"actice cmkeristics .* Variable I No. % Gender Female Male Age 30-39 40-49 50-59 60 and over Years in Practice <11 11-20 21-30 >30 Location of Residency Rochester Other Practice Primary care Specialists 130 197 715 212 325 210 157 214 303 199 171 447 462 379 535 21.6 78.4 23.4 36.0 23.2 17.4 24.1 34.2 22.4 19.3 48.9 50.5 41.5 58.5

PAGE 144

Table 3.4 (Cont.) Variable Practice structure Solo Single specialty group Multi-specialty group Hospital Other/combination Practice visits/week < 100 100-199 200 or more % patients capitated 0 1-49 330 20h I 50 or more 2017 % 28.6 37.4 9.3 7.2 17.5 37.5 26.5 36.0 44.5 27.5 28.0 *Total N = 914. Ns will vary by questithn due to missing data. Approximately 78% of the subjls were male and 21% were female who had been in practice for an average of20 yels (SD = 11). The mean age of the physicians was 48 years (SD = 11 ). ApJroximately 50% completed their residency training in Rochester. For the purposes of this study, primary care jncluded those clinicians in general practice, family practice, general internal medicine, and general pediatrics (Donaldson, MS, Yordy KD, Lohr KN, Vanselow NA-,1996). All others were classified as specialists. Primary care clinicians complsed approximately 40% of the subjects while nearly 60% of the subjects were specialiJs. 131

PAGE 145

A little over one third of the subjects practiced in a single specialty group and nearly one third in solo practice. The mL number of practice visits per week and the mean percentage of patients in capitated leimbursement plans were 283 visits/week (SD = 592) and 24% (SD = 30), resp -lely. Description of Variables Univariate Analyses. For a majori:tY (24) of the opinion variables examined in this study, there was a rather strong and kectionally consistent opinion held by most physicians resulting in either a positive o1 negative skew in the frequency distributions. For a small number of opinion variables (6), there was a greater variance of opinion resulting in more of a rectangular or bimJdal distribution. The frequency percentages for each of the recoded Likert scales of lh dependent variable are summarized in Table 3.3. As described previously, the distributions were utilized to guide the selection of statistical tests. Missing Values. A large number of cases with missing information can create inaccurate or biased results. The variablel with missing values are presented in Appendix D. Tabachnick & Fidell (1983) have outlined the different methodologies on how to treat missing data. For this analysis, calewise deletion of data was selected. This 132

PAGE 146

methodology requires the elimination o observations with missing values for a given variable from the analyses. Logistic Regressions Logistical regression techniques used to detennine which physician/practice characteristics were the most significant predictors of physicians' opinions, the referent criterion variables] Regressions of physician practice characteristics and practice variables on opinions about more infurmed, actively involved patients and ffiDS 1 summarized in Table 3.5. Statistical adjustments for multiple comparisons were not made. It should be recognized that when conducting multiple comparisons, lhe probability of one or more Type I errors is greater than 0. 05. That is, at least one J twenty comparisons may be significant due to chance alone. Table 3.5 Logistic regression analysis o physician characteristics and practice variables on oEinions. Opinion .Y ariables Predictor Variables and P V aloes G PT R s A v c Patients Desire Information Patients should ask more questions 0.33 0.90 0.47 0.18 0.27 0.89 0.68 Patients want more information 0.10 0.27 0.61 0.36 0.99 0.29 0.61 My patients take > role asking 0.19 0.83 0.51 0.26 0.04 0.16 0.20 questions My patients take > role in seeking inf1 0.29 0.19 0.53 0.72 0.15 0.39 0.10 Patients Desire Involvement Patients want more involvement 0.44 0.73 0.84 0.74 0.33 0.20 0.80 Eatients take > role in decisions 0.10 0.11 0.35 0.95 0.53 0.25 0.57 133

PAGE 147

02inion Variables I Predictor Variables and P Values G PT R s A v c Who Should Make Minor Decisions l Who should make minor decisions 0.30 0.60 0.78 0.27 0.32 0.47 0.99 My patients take > role tx minor illnes 0.35 0.04 0.06 0.00 0.54 0.27 0.57 Who Should Make Major Decisions 0.75 0.81 0.76 0.49 0.00 0.96 0.91 My Patients Take > Role in Chronic 0.58 0.37 0.34 0.00 0.38 0.73 0.08 lllness Physician Time MD doesn't have time to involve 0.69 0.11 0.17 0.43 0.96 0.74 0.30 Patients MD doesn't have time to answer 0.70 0.51 0.15 0.12 0.01 0.20 0.01 patients Patient Preference for Decision Making 0.73 0.39 0.61 0.08 0.08 0.85 0.70 Physician Awareness ofTBDS Services 0.10 0.26 0.00 0.02 0.02 0.17 0.06 Physician Opinion of TBDS Services Physician opinion ofTBDS (all) 0.84 0.67 0.14 0. 70 0.63 0.80 0.10 Physician opinion of TBDS (aware) 0.37 0.13 0.50 0.03 0.69 0.11 0.84 Physician opinion ofTBDS (unaware) 0.99 0.00 0.40 0.09 0.00 0.16 0.05 Reason Health Plans Implemented TBD Improve quality of care 0.45 0.40 0.68 0.02 0.74 0.42 0.84 Improve member satisfaction 0.78 0.34 0.62 0.00 0.72 0.86 0.49 Reduce health care costs 0.44 0.96 0.02 0.31 0.55 0.12 0.87 Respond to business community 0.52 0.80 0.02 0.00 0.20 0.19 0.21 Improve patient health outcomes 0.73 0.02 0.04 0.06 0.83 0.88 0.64 Reduce medical liability claims 0.03 0.09 0.04 0.69 0.48 0.76 0.58 More efficient care of capitated patients 0.41 0.74 0.60 0.06 0.05 0.86 0.25 I r

PAGE 148

Table 3.51Conq 0 inion Variables Predictor Variables and P Values G PT R s A v c Potential Impact of TBDS Volume of calls to practice 0.58 0.40 0.52 0.01 0.71 0.54 0.09 Number of inappropriate visits 0.08 0.78 0.52 0.01 0.90 0.43 0.04 Patient demand for services 0.72 0.46 0.26 0.47 0.26 0.00 0.02 Number of medical liability claims 0.27 0.37 0.10 0.69 0.05 0.87 0.13 Average length of clinical 0.79 0.30 0.46 0.99 0.65 0.31 0.21 encounter Patient health outcomes 0.10 0.35 0.83 0.97 0.68 0.76 0.25 Quality of doctor-patient 0.66 0.42 0.29 0.14 0.85 0.97 0.28 relationship Patient satisfaction with medical 0.56 0.36 0.96 0.21 0.54 0.52 0.14 care (G) gender, (PT) practice type, {R) lofation of residency, (S) specialty, (A) age, (V) number of visits/week, (C)% capitation. Patients Desire More Information. Physicians were of th,e opinion that patients do desire more information, should ask Lore questions, and should be encouraged to seek more infonnation about health and r.edical care issues in the medical decisionmaking process. The percentage of physicians agreeing with the statements: a) patients should be encouraged to ask 1re questions about their medical care, b) patients want more information regarding health and medical issues, c) I would like my patients to take a greater role in asking Juestions about their care, and d) I would like my patients to take a greater role in seeJng available information a bout a medical problem or condition was 94%, 900/o, 88r, and 87%, respectively. The relationship between age and physicians' opinions about their patients taking a greater role in asking questions about their care was siJmcant (P = 0.04) but nonlinear. Physicians 135

PAGE 149

in the 40-49 and the 50-59 year old age groups were more likely than physicians in the 30-39 and the 60 and over age groupk to agree that their patients should take a greater role in asking questions about thrl. care (89.7% and 91.8% versus 84.7% and 85.4%). Patients Desire More Involvement. Overall (N = 891), 80-90% of the physicians were of the opinion that desire more involvement in the health and medical decision-making process. The pJcentage of physicians agreeing with the statements: a) patients want more involvjment in their health and medical care decision-making process and b) I would tle my patients to take a greater role in the medical decision-making process was 8,. and 82%, respectively. There were no significant differences of opinion related to physician/practice characteristics. Decisions About Minor Health Issles. Slightly more than half (56%) of the physicians indicated both physician and pJtient together should make decisions about I minor health issues, while nearly 34% in,cated mostly the patient. There were no significant differences of opinion related te physician/practice characteristics. Almost three fourths (72%) of the physicians agreed with the statement that they would "like their patients to take a role in treating their own minor conditions." Practice type (P = 0.04) ami specialty (P = 0.002) were significantly related to decisions about minor health isles. Physicians in multispecialty group I practice were more likely to agree (85.5%1 that their patients should take a greater B6

PAGE 150

role in treating their own minor conditions whereas, physicians in solo practice were less likely to agree ( 63.7% ). Primary le physicians were more likely than specialists to feel positive about their patients a greater role in treating their own minor conditions (81% versus 65.8%). Decisions About Major Health Issues. Slightly over two thirds ( 68%) of the physicians indicated that decisions regarbg treatment for major health issues should be made by both physician and patient tlgether while 28% indicated mostly the physician. Age was significantly (P = 0. bo7) related to physicians' opinions about who should make decisions regarding treatmL fur Illl!ior health issues. Younger I physicians were more likely than older physicians to believe that both physician and I patient together should make decisions about major health issues. Older physicians (50 and over) were more likely than youlger physicians to believe that decisions for major health issues should be made by the physician. Chronic Conditions. Two thirds lr the physicians agreed that they would like I their patients to take a greater role in self-managing chronic conditions. Specialty was significantly (P = 0.000) related to this olinion. Primary care physicians were more likely than specialists to agree that they I ould like their patients to take a greater role in self-managing chronic conditions. Patient Preference for Decision aking. Physicians were asked to what extent they agreed with the statement, "patients want their physicians to make decisions for 137

PAGE 151

them." Opinions were varied on this factor with approximately 46% agreeing, 44% I disagreeing, and 10% with no opinion. There were no significant differences of opinion related to physician/practice cJacteristics. Physician Time. Slightly over hJ (56%) of the physicians disagreed with the statement that physicians do not have time to fully involve patients in the decision-making process. There were no significant differences of opinion related to physician/practice characteristics. Nearly two thirds (65%) disagreed with the statement that physicians do not have adequate time to answer patient qukstions about medical care. Age (P = 0.008) and % of patients capitaled (P 0. 007) tere significantly related to this opinion. There was a linear relationship between and physicians' opinions about having adequate time to answer patient questionl. That is, with increasing age there was an increasingly higher percentage of physiciLs disagreeing that physicians do not have adequate time to answer patient questionl. In particular, physicians 60 years of age and older were more likely than younger to agree that physicians do have d . I a equate tune to answer patient questions. There was also a linear relationsJj between the percent of patients capitated and physicians' opinions about having time to answer patient questions. W h h .I gl high f h It mcreasmg capitation, t ere was an mcreasm y er percentage o p ysiCians hah"" d h dl. . agreemg t t p ysiCians o not ave a ate tune to answer patient questtons.

PAGE 152

Additionally, there were no statistically ignificant relationships between visits/week and physicians' opinions about having time to answer patient questions. Approximately two thirds of all physi1s, regardless of the number of visits/week, disagreed with the statement that physi]. ans do not have adequate time to answer patient's questions about medical care. Physician Awareness of (TBDS) Services. Physicians were asked to report I whether or not they had heard of the TBDS service that had just been implemented within the community by the health planJ; and, to what extent they were familiar with (N = 888), 73.7% had heard of the servi e. Of those reporting that they had heard of the service, 31.3% indicated that they were familiar with its purpose and available services while 42.4% reported that even kough they had heard about the service they knew very little about its purpose and av1able services. Location of residency (P = 0.002{ specialty (P = 0.016), and age (P = 0.02) were all significantly related to physician familiarity about the TBDS purpose and available services. Physicians who had clmpleted thdr residency in the community I reported being more familiar with the TBDS purpose and available services than physicians who completed their residency elsewhere (38% versus 24.9%). Primary care physicians reported being more familiar with the TBDS purpose and services than specialists (38.2% versus 26.5%). Physijans in the 40-49 and the 60 and over age 139

PAGE 153

groups were more familiar with the service than were physicians in the 30-39 and the 50-59 year age groups ( 35.1% and 341 versus 28.4% and 26.8%). Physicians' Opinions About the (TBDS) Service. Physicians' opinions about the TBDS service were analyzed for thrl different groupings: I) all physicians, 2) only those physicians who reported familiar with the purpose and available services, and 3) only those physicians 1o reported indicating that they were unaware of the service or knew very little about its purpose. Physicians' opinions for these I three groups are summarized in Table 3.l Table 3.6 Physicians' opinions ofTBDS service. Physician Group I Opinion Neutral Negative All physicians* 2I!I% (I83) 44.5% (387) I6.0% (139) Knowledgeable ofTBDS service 32]ZO/o (89) 42.0% (II6) 25.7% (7I) Unaware ofTBDS service 36]8% (25) 48.5% (33) I4.7% (10) The sum does not total to I 00% as I8% of all physicians reported being unaware of the TBDS services. As illustrated in Table 3.5, for the overall group (N = 869), there were no significant differences of opinion related to physician/practice characteristics. For those familiar with and knowledgeable about thl mos service (N = 276), specialty was significantly (P = 0.03) related to physicils' opinions about the service. Primary care I physicians were nearly three times more likely than specialists to have a generally positive opinion about the TBDS service. I40

PAGE 154

For those physicians who reported being unaware of the TBDS services (N = 68), practice type (P = 0.001), age (P = 0.002), and% ofcapitated patients (P = were significantly related to physicians' opinions about the service. Even though these relationships were statistic1y significant, small cell sizes precluded any practical significance. Reasons for Implementation of TBDS Service. There were several primary reasons why physicians believed that the I ealth plans were implementing TBDS services. The reasons are summarized in descending order in Table 3.7. Table 3. 7. Physicians' opinions about hy health plans implemented TBDS. Reason I "Agree" Percentages* To reduce health care costs [ 83.3% (708) To respond to business community interests 72.4% (617) To more efficiently care for capitated patients 61.1% (521) To improve member satisfaction 59.1% (501) To improve patient health outcomes 38.0% (323) To improve quality of care 37.5% (317) To reduce the number of medical liability claims 21.6% (184) The sum will not equal to 100%. Each percentage is the percent of physicians who reported "agree" out of the total nJmber of physicians reporting for that particular survey question. Implementation Reason: Reduce
PAGE 155

Implementation Reason: Respond to Business. Location of residency (P = 0.02) and specialty (P = 0.001) wJe significantly related to physicians' opinions I that the health plans implemented TBDS services in response to interests of the I business community. Primary care physicians were more likely than specialists to I agree that the health plans implemented rns services to respond to the interests of the business community (78.7"/o versus r-9'/o). Physicians who had completed their residency in the local community were more likely than those who had completed their residency outside the local community tJ believe that the health plans implemented TBDS to respond to the interests of the Lsiness community (75. 7"/o versus 69.2%) I Implementation Reason: More Efficient Care. Even though there were no statistically significant differences of oplon related to physician/practice characteristics for this variable, there weJe some interesting trends. A greater percentage of physicians with the highest percentage of capitated patients (50% or more) than physicians with a lower percentage of capitated patients (zero or 1-49%) believed that health plans implemented thl TBDS to more efficiently care for capitated patients ( 68.2% versus 61. 7"/o and 57. Io/1There was an inverse relationship rh age. With increasing age, a lower percentage of physicians believed that health plans implemented TBDS to efficiently 142

PAGE 156

care for capitated patients. When compared to specialists, primary care physicians were more likely to believe that the TBD\S was implemented to more efficiently care for capitated patients ( 67% versus 57%)1 Implementation Reason: ImproJ Satisfaction. Specialty was significantly (P = 0.005) related to physicians' opinioJs that the health plans implemented TBDS to b p I h likl h unprove mem er satis1actton. nrnary care p ys1C1ans were more e y t an specialists to believe that the health plans implemented TBDS to improve member satisfaction (65% versus 54.8%). Implementation Reason: Improve Health Outcomes. Opinions were varied regarding physicians' beliefs about improLg patient health outcomes as a reason for why health plans implemented TBDS. Ejr though 38% agreed this was one of the reasons, 32% disagreed and 30% had no epinion. Location of residency (P = 0.04) and practice structure (P = 0.016) were sikmcantly related to physicians' opinions that the health plans implemented TBDS t improve patient health outcomes. Physicians who completed their residency 'in the community were more likely than physicians who completed their residency butside of the local community to believe that health plans implemented TBDS to patient health outcomes ( 40.4% I versus 35.1%). Physicians who were academic hospital based were more likely than

PAGE 157

physicians in other types of practice stlctures to believe that health plans implemented TBDS to improve patient iealth outcomes. Implementation Reason: Improve Quality of Care. There was also heterogeneity in physicians' opinions .fut improving quality of care as a reason for why health plans implemented TBDS. Even though 37.5% agreed this was one of the reasons, 33% disagreed and 30% had nl opinion. Specialty was significantly (P = 0.02) related to physicians' opinioJs that the health plans implemented TBDS to improve quality of care. Primary care ptysicians were more likely than specialists to believe that the health plans implementel TBDS to improve quality of care (45.5% versus 31.9%). Implementation Reason: Reduce Medical Liability Claims. There was a wide variance in physicians' opinions about re6ucing medical liability claims as a reason for why health plans implemented TBDS. ty 22% agreed this was one of the reasons. However, 38% disagreed and 40% had no opinion. Gender (P = 0.03) and location of residency (P =0.04) were both significanlly related to physicians' opinions that the health plans implemented TBDS to redule the number of medical liability claims. I Males were more likely than females to that this was a reason (39.9% versus 29.8%). Physicians who had completed their residency within the local community I were more likely than physicians who had completed their residency external to the local community to disagree that this wj a reason (40.8% versus 35%). 144

PAGE 158

Potential Impact ofTBDS Service. A consistently predominant percentage of physicians were of the opinion that th1 would be no change (versus positive or negative impact) on certain health care variables as a result of implementing TBDS services. The selected health care variablls and the percent of physicians reporting no change for each are summarized in de.Jming order in Table 3 .8. I Table 3.8. Physicians' opinions about the potential impact ofTBDS services on selected health care variables. I Variable I %(No.) Believing No Change* Number of medical liability claims 77.4% (639) Average length of clinical encounter 69.1% ( 570) Patient health outcomes 64.0% (529) Patient demand for services 57.5% (476) Number of inappropriate visits 56.9% (468) Volume of phone calls to practice 56.6% (467) Quality of doctor-patient relationship 51.5% (425) Patient satisfaction with medical care 45.9% (380) The sum will not equal to I 00%. Each percentage is the percent of physicians who reported "no change" out of the number of physicians reporting for that particular survey question. Potential Impact: Number of Medical Liability Claims. Age was significantly (P = 0.05) related to physicians' opinions rout the potential impact ofTBDS on the number of medical liability claims. Physicians in the 30-39 year old age group were more likely than older physicians to belieje that the TBDS would have no impact on the number of medical liability claims. 145

PAGE 159

Potential Impact: Length of Clinical Encounter. There were no significant differences of opinion related to physicibpractice characteristics. Potential Impact: Patient HealtJ Outcomes. There were no significant diffi f . I ed h .1_, h . erences o opinion re at to p ysici&:uu practice c aractenstlcs. Potential Impact: Patient Dejd for Services. Number of visits per week (P = 0.004) and percent of patients capJated (P = 0.02) were significantly related to I physicians' opinions about the potential pact ofTBDS on patient demand for seiVlces. Physicians with less than I 00 vis ts/week were more likely than physicians with I OO-I99 visits/week or greater than 200 visits/week to believe that the TBDS would have no impact on patient demand for services (63.I% vs. 53.90/o vs. 54.3%). Physicians with I-49% of capitated patijnts were more likely than physicians with zero capitation or greater than 50% capitatioj to believe that the TBDS would have no impact on patient demand for service (61.6% vs. 53.9% vs. 54.3%). Potential Impact: Number oflnappropriate Visits. Specialty (P = 0.014) and percentage of patients capitated (P 0.,) were significantly related to physicians' opinions about the potential impact of TBDS on the number of inappropriate visits. Specialist were more likely than primary lare physicians to believe that TBDS would have no impact on the number ofinappro\ riate visits ( 62.7% versus 48.7%). I46

PAGE 160

Physicians with 1-49"/o of the their patiTs capitated were more likely than physicians with zero capitation or greater than 50% capitation to believe that TBDS would have no impact on the number ofinappropria visits (62.2% versus 53.5% and 52.3%). Potential Im act: Volume ofPho e Calls. Specialty was significantly (P = 0.008) related to physicians' opinions about the potential impact ofTBDS services on phone call volwne. Speciali,s were more likely than primtuy care physicians to believe that TBDS would have no impact on phone call volume (62.5% versus 48.6%). Potential 1m act: Doctor-Patient Relationshi There were no significant differences of opinion related to physici practice characteristics. Potential Impact: Patient Satisfaction with Medical Care. Even though there were no statistically significant differencel of opinion related to physician/practice characteristics, physicians with 50% or Jbre capitation were more likely than physicians with with zero or 1-490/o capi tion to agree that the TBDS service would increase patient satisfaction with medical bare ( 43% versus 33% and 33%). Logistic Regressions SIIIIllll3!Y. lpproximately 90% of physicians were of the opinion that patients desire more infonnahon about health and medical issues and generally desire more active involvement in the medical care decision-making process. A nearly equal percentage (82-88%) weri of the opinion that they would like to see their own patients take a greater role in asking questions, seeking information, and 147

PAGE 161

participating in the medical decision-making process. Both of these opinions were l allh"" \d. h .. CODSlStent y strong across p YSlClanr practlce C aractenstlCS. When considering patient invol Iemen! in decision making for minor, major, and chronic conditions, approximately thirds of the physicians were of the opinion that the physician and the patient should collaborate in making decisions about minor as well as major health issues. Similarly, two thirds were of the opinion that they would like their own patients to take a more active role in the management of chronic conditions. However, for minor health ilsues, there was a tendency for physicians to believe that patients should take a role and for major health issues there was a I tendency for physicians, especially the older physicians, to believe that physicians should take a more predominant role.,early two thirds of the physicians, especially older physicians and physicians with no capitated patients, believed that physicians have adequate time to answer patient qulstions about care and to fully involve them in the decision-making process. l Even though a consistent majori of physicians were of the opinion that, in general, patients desire more in the decision making process, that they want their own patients to take a more aLive role, and that physicians have adequate 1 th I h .. b h t1n1e to mvo ve patients, ere was a greater vanance m p ys1e1ans opm10ns a out t e patient's preference for decision making] Slightly less than one half were of the 148

PAGE 162

opinion that patients wanted their physicians to make decisions for them while an equal percentage disagreed and 10% hJ no opinion. Approximately three fourths (73 j7%) had heard of the 24-hour TBDS service that had been implemented within the coLunity in the two months prior to this research. Of those reporting that they hld heard of the service, 31.3% indicated that they were familiar with its purpose and services while 42.4% reported that even though they had heard about the seh.ice they knew very little about its purpose I and available services. Depending _upon r.iarity and knowledge about the service, 20-40% of physicians held a generally ,itive opinion, 40-50'/o a neuttal opinion, and 15-30% a generally negative opinion. Ph sicians reporting a higher number of patient visits/week were generally more positive about the TBDS service than were those seeing fewer patient visits. As summarized in Table 3.7, physicians believed that the primary reason the health plans implemented TBDS services were for reasons of cost control in response to the interests of the business community, as opposed to reasons related to clinical quality as measured by patient health outJomes, quality of care and malpractice. As summarized in Table 3.8, physicians predlminantly believed that the TBDS services I implemented within the community would have virtually no impact on selected health care delivery parameters associated with lfficiency and effectiveness. 149

PAGE 163

Of all the predictor variables, logistic regression analysis detennined specialty to be significantly related to the greateJ number (9) of physician opinion variables. Specifically, when compared to specialiks, primary care physicians (general practice, family practice, general internal medi4 and general pediatrics) were more likely to: 1) agree that their patients should take J greater role in treating their own minor conditions; 2) agree that their patients slould take a. greater role in self-managing chronic conditions; 3) be familiar with ld knowledgeable about the TBDS services; I 4) have a more positive opinion about lDS services; 5-7) believe that the health plans implemented mos to respond tor of the business community, to improve quality of care and to improve member satisfaction; 8-9) believe that TBDS services would decrease the volume of cbs and the number of inappropriate visits. L a1 di I ed th 1 f d regresston an ysts m cat at ocatton o rest ency was significantly related to five physician variables as well. When compared to I physicians who had completed their residency external to the local community, those I physicians who had completed their residency within the local community were more likely to: 1) be familiar with and about the TBDS services; 2) believe that the health plans implemented TBDS to reduce health care costs, 3) believe that the health plans implemented TBDS to respond to interests of the business community, I 4) believe that health plans implemented rs improve patient health outcomes, and 5) disagree that the health plans implemented TBDS to reduce medical liability claims. 150

PAGE 164

Logistic regression analysis indicated that age was significantly related to four physician opinion variables: (1) decisionl for major health issues, (2) time to answer patient's questions, (3) their patients tlg a greater role in asking questions, and I (4) awareness ofTBDS services. Specifically, older physicians were more likely than younger physicians to believe that: a) dlisions for major health issues should be made mostly by the physician and b) physiaJ. bave adequate time to answer patient questions about medical care. The relatibnship between age and physicians' opinions about their patients taking a greater role in asking questions about their care was nonlinear. Physicians in the 40-49 and t e 50-59 year old age groups were more likely than physicians in the 30-39 and the 60 dover age groups to agree that their patients should take a greater role in asJng questions about their care. The relationship between age and awareness lfthe TBDS services was also nonlinear. Physicians in the 40-49 and the 60 and ot age group reported being more familiar with the service than physicians in the 30-39 and the 50-59 year age groups. Even though there was a statistically significj relationship between age and opinion about TBDS services (60 and over physicians brg more positive) in the group of physicians who were unaware of the TBDS service just implemented in the community, the small number of physicians reporting that they Lere ''unaware" of the service precluded any practical significance to this finding. 151

PAGE 165

Finally, logistic regression analysis determined that the percent of capitated patients was significantly related to two physician opinion variables regarding the potential impact of TBDS on utilization. Physicians with zero or greater than 50% capitation were more likely than physicians with some capitation (1-49%) to believe that TBDS services would have an impabt on patient demand for services and the number of inappropriate visits. Similarly J physicians with a higher number of visits/week(> 100) were more likely tJ physicians with fewer visits/week(< 100) to believe that the TBDS would have an imract on patient demand for services. There was also a linear relationship between %1 of patients capitated and physicians' opinions about having adequate time to answer patient questions. With increasing capitation, there was an increasingly higher percent1e of physicians agreeing that physicians do h d . I not ave a equate tmle to answer patient questions.

PAGE 166

c I TER4 QUALITATIVE .METHODS AND RESULTS In1uction Qualitative inquiry (key inteJViews), was used to explore more in depth physicians' opinions about: I) Co,sumers/patients being more informed and taking a more active partnership role in Health and medical care decisions. 2) The impact they believe a more informed, actLe consumer/patient will have on the quality of the doctor-patient relationship, clinicj efficiencies, health outcomes, patient satisfaction, health care utilization, healtJ care costs and malpractice liability. 3) The incorporation of telephone-based decisiot support (TBDS) services for patients/consumers into the care manage, ent process. 4) The impact they believe the incorporation of TBDS in the health care delivery process will have on the quality of the doctor-patient relationship, clinical e ciencies, health outcomes, patient satislilctioo, health care utilizatioo, heal, care costs and malpractice liability. This chapter will describe the methods, analys s, and results pertinent to the qualitative portion of this study. 153

PAGE 167

Met ods This section describes the instrumentation, and process used in conducting the key informant interviews\ Sample Selection Patton (1990) has pointed out that, "Qualitative inquiry typically focuses in depth on relatively small samples, even sLgle cases (n=l), selected purposefully .... The logic and power of purposeful samplg lies in selecting information-rich cases for study in depth" (p. 169). In contrast probability sampling used in quantitative research, sample size in qualitative res,ch is not dependent upon probability statistics and power calculations to compute sample size. Instead, "The validity, meaningfulness, and insights generated from qualitJive inquiry have more to do with the I infonnation-richness of the cases selected and the observational/analytical capabilities I of the researcher than with sample size" (Jatton, 1990; p. 185). Because the purpose of this aspect of the study was to obtain lonnation-rich "thick description" of physicians' opinions about the key researlh questions, a fonn of purposeful sampling known as maximum variation sampling wL used to select the study subjects.

PAGE 168

"Maximum variation sampling occurs when one seeks to obtain the broadest range of information and perspectives ol the subject of study. Guba and Lincoln (1989) claim this is the preferred stratel for qualitative inquiry" (Kuzel, 1992; p. 3 7). This approach is one of deliberate!{ seeking out diverse variations and coounining important common patterns in their experiences. Initial data are then elaborated upon and given more breadth and strength by teliherately searching for confinning and disconfinning cases that will ... suppoi or challenge the investigator's understanding ofthe topic of study" (KuzeL 1992; p. 41 ). When conducting maximum variation/confinning-discoofinning sampl subject selection continues to the point of saturation or redundancy with 12-20 Jata sources recommended (Kuzel, 1992). Therefore, 21 clinicians were seljcted for key informant interviews from the domain of the physician population to be studied ---all primary care physicians (approximately 859 family practice, generalists, internists,) and specialists (approximately 1,306) that were memberl of either the Rochester Independent Practice Association (RIP A) and the Roclester Community Independent Practice Association (RCIP A). Seven key informLt physicians each were selected from those f h d I ali h d b a1 groups o pnmary care p ys1Cians an speet sts w o reporte emg pos1t1ve, neutr or negative about the TBDS service. 155

PAGE 169

"Key informants are individuals io possess special knowledge, status, or communications skills, who are willing to share their knowledge and skills with the researcher and who have access to or observations denied the researcher" (Goetz & LeCompte as cited in Gilchri11992). They are "trustworthy, observant, reflective, articulate, and a good story teller" (Johnson as cited in Gilchrist, 1992). Snowball sampling 1992; 1990) was used to select the physician key informants who met the above criteria. r is, using the key informant criteria and the requirement to identifY physicians who held preconceived opinions (positive, negative, neutral) about the TBDS setv:ice, +live director and the physician president of each IP A were asked, "Who should we t81k to?" The IP A executive directors and I physician presidents began the process of identifYing initial key i.Dformants who in turn directed them to additional physicians wJo were thought to be "information-rich" informants. This iterative process continjed until the full complement (21) of key informants was obtained. Physician Opinion Interview (POI) Guidel Following the guidelines for quali ative interviewing, the standardized openended interview approach (Patton, 1990) r used to conduct the key informant interviews. The questions comprising th,Physician Opinion Interview (POI) were extracted from the Physician Opinion Survey (POS). The questions were formatted to

PAGE 170

accommodate the interview process and to acquire more in-depth information about the topic of interest. Like the POS, final questions for the standardized open-ended interviews were developed by the research team in collaboration with the executive and physician leadership from RCIP A Jd RIP A by following the guidelines of Pieterzak et al. ( 1990) for the developmlnt of new measures. The final POI guide is illustrated Appendix B. The survey includes a profile of demographic data and physician practice characteristics. The first section of the POI asked physicians their opinions about informed and actively involved patients in the health and medical care decision-ma.lfutg process. The second section of the survey asked physicians their opinions atut informed consent and telephone-based decision support (TBDS) services as well as some open-ended questions regarding the key challenges and benefits of implementLg TBDS services. Interview Structure and Administration Upon arrival at the scheduled tim and place for administering the interviews, the following agenda was used with each physician: a) complete introductions; b) review of the purpose of the interview, the interview process including tape recording, what will be asked in the intelews, how the information will be used, and the voluntary, anonymous, and confidentt nature of the interviews; c) answer any questions or concerns and obtain consent; d) conduct the interview and 157

PAGE 171

e) conclude appropriately. Physicians were given the opportunity to decline the interview, stop the interview at any poinl or refuse to answer any questions with which they were not comfortable. The face-to-face structured inte ews were conducted by the investigator during the week of August 22-26,1996. The interviewer took notes during the interviews and all interviews were audio 1 aped. Audio tapes and notes were anonymous and confidentialidentified inly with an identification number. The avernge length of each interview was rtely 60 minutes. AD interviews were transcribed, coded, segmented, and anal}lzed as detailed in the following analysis and statistical methods section. AnJ yses All audio-taped, standardized op ded interviews were transcribed, coded, segmented, and analyzed using methods lr qualitative analysis (Patton, 1990; Crabtree & .Miller, 1992). Methods of qualitative rs generally encompass three primaiy functions: I) preparing, organizing and ceding the data in order to identify key data I areas and to facilitate future data retrieval and analysis; 2) segmenting the text data into analytically meaningful components; land 3) interpreting the data through self-158

PAGE 172

immersion in retrieving, sorting, making ronnections, examining patterns, and testing interpretive models of the coded and segptented components (Crabtree & Miller, 1992; Patton, 1990; Reid, 1992). Transcription, Coding and Segmentation All audio-taped standardized ope -ended were transcribed into data files using standard word processing. All text files Lere converted to ASCIT format in preparation for use with NUD-IST 4 (QLntative Solutions and Research Pty Ltd. I 1997) text-analysis software. Using NUID-IST 4, all text files were imported, numbered, and printed for initial coding. The first step in coding was the development of a template in the form of a codebook used fur organizing and interprg text. A codebook fur tbis study was created by using a template analytic approach, ". . beginning with a basic set of codes based on a priori theoretical understandinks and expanding on these by readings of the text" (Crabtree & Miller, 1992; p. 95). The beginning set of a priori codes was based on the investigator's I understanding of the issues and initial review of text files and field notes. Using an I iterative process that encompassed multiple readings, interpretation, and recoding of the text files, the primary investigator indlpendently expanded and refined the initial set of codes. Collaboration and consensJ among the two primary investigators I 159

PAGE 173

determined the "final" codebook to be used in text analysis and interpretation. All text files were computer coded for future retl.eval, sorting, and analysis using NUD-IST 4.0. Data Analysis and Codebook The goal of data analysis in qualitative inquiry is to provide infonnation-rich descriptive data ("thick description") by rbing salient patterns, themes, and categories that emerge from the data. Iligenous typologies are classification systems comprised of verbal categories used by the key infonnants. In contrast, analyst conslructed typologies are classification comprised of patterns, categories, and themes constructed by the investigators (Patton, 1992). "Negative case analysis involves addressing and considering altitive interpretations of the data, particularly noting pieces of data that would tend to Lme the researcher's reconstruction of reality" (Erlandson, Hartis, Skipper, All+ 1993; p. 121). In this study, particular attention was given to negative case analysis and to the development of both indigenous and analyst-constructed typolbgies in developing the codebook and in analyzing the data. Using an iterative process ofmultip1e readings and interpretation of the text files, the initial analysis was expanded 1 refined by the primary investigator. The outcomes of the iterative analytical process included: a) refinement of coding 160

PAGE 174

constructs, b) construction of a profile summary for each case document (interview), c) construction and import of a case ,table for linking quantitative and qualitative interview data, and d) question-by-que,on analysis delineating direction or how the key infonnants felt about a given topic negative, neutral) as well as why they felt the way they did. The final codebook used for text analyses and interpretation is summarized in Appendix C. Inter:pretation The goal of interpretation in inquiiy ... involves going beyond the descriptive data ... attaching significancj to what was found, offering explanations, drawing conclusions, extrapolating lessoJs, making inferences, building linkages, attaching meanings, imposing order, and 1ealing with rival explanations, disconfuming d d 1 f I. th abil. f cases, an ata lffegu annes as part o testmg e VI tty o an mterpretat1on .... The emphasis is on illUmination, understlding, and extrapolationrather than causal determination, prediction, and generalizatlon" (Patton, 1992; p. 423-424). For this study, data analysis and lerpretation incorporated a modified furm of I member checking to enhance methodological soundness. "Member checking provides I for credibility by allowing members of stake holding groups to test categories, interpretations, and conclusions" (Erlandtn et a!., 1993; p. 142). A modified furm of member checking was conducted by pro 'ding a copy of the preliminary analysis and

PAGE 175

interpretation to three independent phys cian reviewers who were not a part of the study population for critique and feedba k. One of the physician reviewers was David B. Nash, M.D., M.B.A, Associate D1 and Director of Health Policy and Clinical Outcomes at Thomas Jefferson Universir Hospital and Jefferson Medical System in Philadelphia, Pennsylvania. The second !physician reviewer was Mark Earnest, M.D., Internist at the University of Colorado Health Sciences Center General Internal Medicine Clinic and doctoral student in lhe Health and Behavioral Sciences Program, University of Colorado at Denver. physician reviewer was Zach Gemarg, M.D., Internist, ChiefMedical Officer of a 140,000 member mixed model HMO and President/ CEO, Gerbarg and Associ1 In order to guide and to provide a more uniform, standardized approach to the checking process, each of the physician r were asked to test the data analysis, interpretations and conclusions by answ,ring the following questions: I) From a physicians' viewpoint, did you detect a different pattern of key themes and issues emerging from the data? If so, they? 2) From a physicians' viewpoint, were any key themes or issues missed? tr so, what were they? 3) From a physicians' viewpoint, did you arrive at different int rpretations and/or conclusions about the data? If so, what were they? 162

PAGE 176

Minor adjustments and revisions from the independent physician reviewers were incorporated into the final analysis. Collaboration, and consensus among the two primary investigators guided the "final" analysis and interpretation. Results In the information that follows, dtcriptive characteristics of the interview subjects are provided. Each section beJs with a specific interview question followed by a balance of analysis and thick d4on utilizing representative participant quotes to illustrate the pertinent themes that emerge through in-depth analysis. Each section closes with a summary of the .key themes lthat emerged for that question. A llll!ior purpose of this analysis las to present and organize key infurmant responses in a way that provided more in depth insight to each question as well as a more clear overall pattern of emerging themes and issues. A concluding synopsis of common overarching themes that emergl across all questions is provided. Subjects As previously described, criterion-based, maximum variation snowball sampling was used to select 21 (20 males land I female) key clinician informants from the approximately 2, 000 IP A physicians bls studied. Relevant subject characteristics are presented in Table 4.1. 163

PAGE 177

164

PAGE 178

The mean age was 46.6 years (SI = 8.3. Fourteen completed their residency Rochester. There were 13 primary care hysicians (internal medicine= 9, family practice = 2, pediatrics = 2), seven specia.Ity physicians (ophthalmologists = 2, gastroenterologist= 1, orthopedic surg,n =I, nephrologist= 1, allergist/immunologist = I, clinical psychiatrist = 1 ), and one clinical psychologist. Opinions About Patient Involvement in Medical Decisions The first series of questions. ( Q 11) were intended to explore clinicians' opinions about patient involvement in hfh and medical care 1. What do you think about patients w ling more information regarding health and medical issues? Twenty out of 21 clinicians had a ery positive opinion about patients wanting more information regarding health and medical issues. Physicians placed major emphases on: a) better decisions leading Jo improved health outcomes and quality, b) the doctor -patient relationship, and c) Lnsumerism and patient preference. Better decisions, outcomes and Physicians said that a more informed consumer/patient would be more likely to make better decisions generally resulting in superior quality, more efficient care and iproved health outcomes. Generally I think that's a idea. I think informed patients do take better care of themselves. If they understand what's going on, they make bettd decisions. (Pediatrician) 165

PAGE 179

In my experience thus far, it makes the care a lot easier. When it's one-sided it mhlces it very difficult to convey the importance of taking medications, following a specific treatment program, espetially if the patient doesn't ask I questions or doesn't get involved. Because if they're having any difficulty with the prbgram they may not be that likely to voice that and will just nbt follow the program. I tell patients outright from the front tHat I'll always offer them the best treatment plan, but if thete' s something about that treatment I plan that doesn't jive with what they expected, or doesn't fit with their life style, we do. find a middle ground. I find that a lot easier, it puts the patient at ease, and it allows them to speak to me. And thai has a lot of pluses, not only in just management of the but, I think it allows better rapport between the patient and tre physician so they're likely to tell me something if something's not working or something goes wrong. (Internist) I think it's great. From my perspective, the more patients know in terms of their medicalibowledge base the easier it is for me to take care of them because they understand the rationale, the treatment. And it makes thy job easier and it makes good medical care easier. (Internist) In general I think it's a gopd thing. Patients that are better informed are more likely to have a good result and do well. (Gastroenterologist) Although most physicians supported and endorsed patients being more informed, one physician expressed a concern that more rormation may lead consumers to wrong conclusions. He said that he didn't believe one could generalize that a patient having more information was a good idea. The concern is that they use a little bit of information and think they understand it aAd come up with totally wrong conclusions . .. like with a lot of things, a little knowledge can create big problems. Then 166

PAGE 180

you have to spend a lot of time undoing what they think they know. (Internist) However, the same physician, fully supported patients desiring more active participation: j I tend to find that as the c1atient gets more involved in making decisions, things usually run more smoothly. (Internist) Doctor-patient relaJionship. Not iy did physicians think that patients wanting more infonnation about health and medical issues was positive, most indicated they actively encouraged information seebg by patients. They saw themselves as I 11teacher11, 11facilitator11 of decisions, and the most reliable source of information for patients in the context of the doctor-pati1t Patients need to be educatL as much as possible. The word doctor means (Orthopedic Surgeon) I think one of the ways rl useful to my patients is being able to provide informatiob in an appropriate context. (Internist) I like to stay away from a paternalistic practice and I like to shoot for a collaborative modeled medical practice as much as can be done. (Fanilly Practice) I don't see my role as pateLalistic. I see my role purely as an advisor and to give niy best possible professional opinion. (Internist) J I find myself frequently g to a patient. I had a conversation yesterday wit a man who had been recommended a surgical p+cedure by a specialist surgeon. And I asked him what the surgeon explained to him about 167

PAGE 181

the potential risks. He e lained the benefits, the patient could explain the to me but he knew nothing about what the potential risks 9r what the failure rate was. And I said, 'well, I think that's more information you need from I that surgeon, and certainly it's information I would want to know if I were in your pdsition before the procedure itself' I find that comes up quitJ a bit. Getting patients to go back to their specialist and sayl 'I need more information to make a proper decision here.' It's interesting how frequently they come to this office after donsulting with somebody and wanting more information saying that they didn't get enough. (Internist) Consumerism and patient preferlnce. Informants acknowledged that health and medical care have entered into an erl of consumerism characterized by patients desiring more information about health ld medical care issues and often desiring more active participation in health and ,dical care decisions. They pointed out that physicians will need to be prepared to work with the strong and growing trend of health care consumers who might have a strong preference for more health and medical information. It's clearly the trend where things are going, more consumerism, and I think its what we all see in the office. (Internist) It's a natural part of having a more educated population. It's natural and it's an empbwering thing that's happening. I'm not surprised it's I think if it's done-it will happen. The question \is whose going to provide the information. (Family Practice) I just think that the days the doctor being dogmatic I and saying, 'Here's the pill, you take that and you'll get better are over.' People arb understanding, listening to their own bodies, their owh environment, trying to make 168

PAGE 182

their own decisions with a little bit of guidance from the physician. (Internist) I have no problem with tHat. In this day and age with I modem technology I think it's perfectly appropriate for them to want to have mol information to base their decisions. (Pediatrician) Summary. Overall, clinicians held a positive opinion about patients and consumers wanting more information abJut health and medical issues. They acknowledged that there is a growing coLmer trend characterized by patients who have a preference for more information rlther than less. Informants viewed themselves as the best source of health and medical lormation for their patients. Many conceded that with an increasing number of informed patients, their communication style would need to be more predominantly ''teacher" and "facilitator" rather than paternalistic decision maker. Physicians Jndamentally believed that a more informed patient would result in better health and ledical care decisions leading to improved quality of care, health outcomes, and dodor-patient relationships. 2. What do you think about patients wanhng more active participation in the medical care decision -makingprocess? Twenty out of 21 clinicians positively endorsed a more active patient in the medical care decision-making process. PJysicians most frequently emphasized issues related to the dynamic of the doctor-patieht relationship, recognition of the variation in 169

PAGE 183

patient preference for active participation and the subsequent need for physicians to be I sensitive to varying patient preferences or involvement in the medical care decision-making process. Doctor-patient relationship. Ph sicians once again noted that more active involvement of patients in the medical dJcision-making process is a significant trend in today' s health care, a trend they must b1 prepared to deal with in the doctor -patient relationship. They believed that generally more active involvement by patients in the doctor -patient relationship would lead J increased patient compliance, and therefore satisfaction with care. It's a real phenomenon. If-egarding the importance of patient participation in the diagnostic and therapeutic process, I think it's probably the most significant change in the dynamic of the dodor-patient relationship. Whether physicians like it or not, they think it's important enough, isn't really the issue. It's that it's a reality that needs to be addressed. (N I think if they take a personal interest in their care rather than relying on the to do everything for them that they're much better, first 0f all, to bring things to your attention that you might otherwise have missed; to be compliant with their therapy because they understand. (Gastroenterologist) I think that when patients get to own more of the decision making process they're mbre comfortable with whatever I the results might be of their clinical choices. I think that a I lot of the chagrin and the feelings of disaffection that patients have of their medical care is when they feel like they were led down some garden pa$, some complications occurred and if they had been more !actively solicited or set up that 170

PAGE 184

path, I think they're more willing to deal with the consequences. I give them active I involve my patients actively. They're given alternatives to make a decision. I put the ball back in their court. I to find that problem, explain it to them, give them alternatites. (Orthopedic Surgeon) One alternative to enhancing doctor-patient relationships was explained by a clinical psychologist. He pointed out that, increlsingly, he assumes the role as one who facilitates a better understanding and communications between patients and other physicians. Patients that I work with, part of my role that I serve and play is to help them become more active with their physician. It works out better, overJ.n, in terms of clients otherwise, they don't understand why probedures are being done, they don't understand the rationale. The more they understand the better. It's not written down that that's what rm supposed to do, but that's ultimately what I try to do. The doctors here are very willing to go along with that as well. I can give you an example of one patient who felt lack of control. Her personality was one of needing a lot Jf control. She's in a high power job and used to a lot of contrbl. Now she's sick and the control is out of her hands. She getting really angry at the doctors for prescribing all these mediations. 'What are these for?' she didn't understand it. And \after doing some work with her she was able to articulate this to the doctor and the doctor was able to give her contrbl or some say about when they'll increase the dosage, whed they won't. And they're working more cooperatively. Ultuhately, she's much happier. That's what a lot of it is. I think :doctors, often times, medical doctors don't realize that Jl>atients don't understand what they are saying. It really takes burturing a person through the process. I went through this yesterday with another patient. She was much more willing to tak9 the kind of medication and under stood why they wouldn't give her the Tylenol with codeine in it by the end of our discussion. She was much more ready to be a partner. To, kind of, 1eally take part in the treatment. 171

PAGE 185

Patient preference. At the same time, however, it was noted that patient involvement in medical decisions is somewhat dependent upon patient preference. That is, some patients clearly do not wJt to be involved in the decision-making process. Instead, they would rather theJ physician make the decision for them. Interestingly, several physicians noted tJt they were not comfortable with patients who preferred that the physician take a lore paternalistic, directive approach. I It bothers me a bit _when get that line from the patient, I propose choices to them and they say, 'well, you're the doctor, tell me what to drl.' They just want you to tell them. I don't like that. I feel with that. Patients like to be told what to do rather than participate in decision making. Being told what to do and feeling comfortable that the doctor is able to tell when not feeling fully equipped to participate in those ch ices. But I think it is my job to help them be comfortable with participating rather than just telling them what to do. Some perhaps, aren't even capable of doing that. Some patients, for whatever reason, whether it be their illness or reasons, some patients just aren't able. You can try and a proxy who can participate for them or you can just plan role and make the decision that you think, is the best without their input. You have to work with what you have to wdi with. (Internist) Physicians also acknowledged that for those patients who do prefer to be more 1 1 d. h d akin I :fi active y mvo ve m t e eetsxon m g process, patient pre erence or consumer choice is paramount, even if the physicii tends to disagree with the patient's decision. Something that patients want is to know my opinion as the medical professional. I really think is best even though there may be a range and they can have some choice and they 172

PAGE 186

can ask questions about ach choice. They want to know what I think is the best option bd what I would recommend doing. So I think this all comes into being part of the proper communi cation about treatment enabling the patient to make an active choice. And when a disagrees with me, even about accepting any treatment at all, I explain to them that I understand that and that I'm not to take away their autonomy, that that is paramount. They iwml always have the right to choose not to accept treatment. And after I've explained to them potential benefits risks of treatment, nontreatment, then I know that they can truly make an informM decision and that my conscience rests I easy because they've made a decision with the best available infonnation. (Internist) I'm all for it. They're the one's who are taking the medication. They're the one's who getting the tests. They have the right to make decisions, irrational I may think they are and however frustrated it makes me when they come in with their herbal preparations or desire for this test or that test. All I can do is tell them what Iithink is the right way to go and why I think that, and if my reasons are valid to them, which they usually are. And sometimes reasons are valid to me and they con vince me that we should choose another course. It depends on the severity of the 4n. (Internist) Summary. Overall, physicians heltl a positive opinion about patients and con-. .. I. h ed'cal d makin sumers wantmg more active partictpatJ.on m t e m 1 care eCJ.Stong process. Physicians contended that a patient who was more actively involved in medical decisions would be given a greater sense oi control and self-confidence culminating in greater compliance with agreed upon diagnostics and therapeutics. This would result in higher overall patient satisfaction with lcare. At the same time, it was recognized that patient involvement in medical deciilns is somewhat dependent upon patient preference. That is, some patients clearl}l have a preference for more involvement in 173

PAGE 187

medical decisions whereas others do not want to be involved. Additionally, for those patients who clearly prefer an actle involvement, physicians generally respect consumer choice, even if the physician tends to disagree with the patient's decision. 3. What do you think will be the impacl on health status/outcomes o(those patients who seek out infonnation and more hcnvelyparticipate in medical decision making (better. same. worse)? Sixteen physicians representing t)[oth primary care physicians ( internal medicine, family practice, pediatrics) and specialists (gastroenterology, ophthalmology, nephrology, allergyfunmunology, and pr) believed the impact on patient health status and outcomes would be better. Two (orthopedic surgeon and pediatrician) believed the outcomes would be the saml, two (internists) believed that patient satisfaction with treatment would increaJe regardless of the outcome, and one (internist) believed that the impact on he1th status/outcomes would be dependent upon the disease. I think those patients shoti.ld be healthier in the long run, because they will and, as I said earlier, comply with their recommendatiops better. If they understand better that a certain medicine has been prescribed, and the importance of it, then they would be rhore likely to take it than if they didn't know. 'The said do this, but I don't think it helps me very much. I don't why he wants me to do it.' I think in that situation theYt're much less likely to comply with the medication. If perhaps, are some side effects, and they understand that the side are something they have to put up with for a greater goo4, they would be more likely to take it. Whereas if they don't really understand, 'that damn stuff gives me 174

PAGE 188

diarrhea' or 'it makes feel tired, what good is it? I'm not going to bother with it.' So, I it would help them to comply better and then they should be ih better health. (Pediatrician) Th "lh 1 .. "hd"" akin e more active y t ey m t e ectston m g process with respect to their health care the more invested they are in fully participating. When patiehts feel more invested in the process they are more likely to with whatever decisions are ultimately made .... I think anybody is more willing to do something that they participate in. If the patient feels as though he or she is being forced to do something that have had no input into, then they're probably less likely to comply with whatever the recommended regiment is. (AIIergiTologist) The best example I o: off the top of my head, would be to look at diabetic control, which I think there's lots that shows when a person is actively fngaged, and takes responsibility for their management. That you can achieve a lot more than by blinded directing of a patient towf,rd achieving something that they're not even sure what it is. (Intjst) I think they are more to be compliant, not only with appointments but with meldications and with sudden changes in direction ofhealth care. If something goes wrong, if they have an understanding this is a possibility, or have an I understanding why this is happening, they won't get frustrated and feel like they're being excluded from the process. I think, overall, it's going to have a very positive impact, much better than the old paternalistic approach to health care. (Internist) Th I ed th f =-.t: b ere was some very mmor concern exprss at nnsmterpretatton o uuormatton y patients may lead to poorer decisions and thus worse outcomes. Probably it will be much bltter. I think occasionally you may I find a person who will misinterpret some of the information or will direct some of the ormation they get in the wrong way. (Internist) 175

PAGE 189

In other illnesses it may be a hindrance to appropriate care. Things like chronic syndrome, or fibromyalgia's where the request for more infopnation leads them to places that physicians' traditionally liave not felt comfortable or have not found any clinical utility . . (Family Practice) Patient satisfaction. Even in thTe few physicians who believed that clinical and/or functional health outcomes would be equivocal, they acknowledged that a more active, informed patient would be a morl satisfied patient even if the outcomes were not as positive as hoped for. I think it's going to be positive. I've seen some of the research I that's come out so far it looks like patient's that are actively involved in their health calre, not only feel more satisfied, are more happy with their heJith care overall. (Internist) Ihink bin h 'dl. .. p. t one t g t at It oes 1s It mcreases satistactiOn. at1ents feel more satisfied with outcomes. And even if something doesn't go to wishes. And people who spend the time and effort thi.nkiDg and teaching themselves and gaining knowledge about illness apd health to be an active participant in decision making begin to 1eam, 'hey, this is not black and white exact science.' There's a lot of judgment, interpretation and there's a lot of dealing with probabilities, and dealing with uncertainty so that decisions are often that, well in 60% of the time this treatment works but 40% [doesn't, or 70% doesn't. And they come to appreciate that so that f.hey're not as likely to be disappointed, upset, or have negative feelings when things don't work out. You just come back and you'll go through the process again. So I think that people who are more 1active in their medical decision making, once again, are going to happier. More satisfied with outcomes and less disappointed in nbgative outcomes. (Internist) Medica/liability. Physicians belilved that increased patient responsibility and I accountability in the medical decision-making process would lead to improved patient

PAGE 190

satisfaction and health outcomes resulting in less potential for litigation-even if the outcomes were not as positive as hoped for. I think, in general, it will robably decrease if they are more involved in the decision making. But I don't think it goes I away because, unfortunately, they can always say that they didn't understand the expianations, and you're still the doctor, you're still the expert. Sq I don't think it will go away. But I think, I've always found tpat if you work with the patient to gether that there's much lfss possibility of them saying, after the fact, 'how come, this,! this and this', if they see that they're involved in helping you re that decision. (Internist) Patient preference was recognized as a factor influencing outcomes. As previously described, some patients d, more infonnation and want to be involved in decision making whereas others do no Also, it was acknowledged that there may be a selection bias in those who seek out more information and who actively participate in medical decisions. That is, such individuals may already have this type of coping style and will naturally do better the more information and the more I involvement they have in health and medical decisions. My bias would be to say J.uer, but I also think that people that seek out information have a coping style that information will be beneficial. But people that won't seek it out given the information, it will onl make it worse. I think some people are information seekers add some people are deniers. Those I deniers have a hard time if given too much information. They get overwhelmed. If they lare seeking out information I think it's probably better to givel it to them. I don't think it's so hard and fast and the people wlio are deniers, if you do some work with them, they come aroJnd eventually to use information, but they need a little workf They take in information as they're ready to take it in. You can give it to people, if they're ready. 177

PAGE 191

They'll ignore it or it 'Will increase their anxiety, or just not pay attention to it if they're 1ot. (Clinical Psychologist) Outcomes would be better because that's the personality of the I person. If you had the same person who wanted to seek the information, I don't knoJ whether you can do that, but generally the person who wants to lseek out the information is one who wants to be more It would be more likely, that they're involvement in learning about different quality of life issues, that things may change. If thJy tend to be more active in their care. But is it because they the information or is it the personality of the person? I don't know what would be affecting that more. (Ophthalmologist) Summary. Providers believed that those patients who actively seek out -S:. d h 1 .I. edi al d . "11 h b huormat1on an w o more act1ve y partlapate m m c ec1s1ons W1 ave etter clinical and :functional status health outjmes. They thought this was particularly true for patients with chronic illnesses. The lechanism for better health outcomes and reduced potential for litigation was relat1 primarily to improved understanding, . d b I. I din d li d part1c1pat1on, an uy-m to treatment regunens ea g to 1mprove comp ance an overall satisfiu:tion regardless of wheth, the outcomes were negative or positive. Patient preference was again acknowledied as a fitctor influencing outcomes. Some patients desire more infonnation and want to be involved in decision making whereas others do not. Health outcomes will be iLproved if physicians are sensitive to patient preference for active involvement and jtch their response to that preference. If physicians fail to match their response to preference, then health outcomes and overall patient satisfaction may be diminilhed. It was pointed out that selection bias 178

PAGE 192

might be contributing to better outcomes of those who more actively participate in medical decisions. J 4. For a minor health issue (sprains. 1ains. cold/flu $Jl!11ptoms). who would you want to make decisions about Please provide the reasoning supporting your opinion. Explain how this would i#tpact: clinical efficiency. the doctor patient relationship. patient self-eQicacy. healtH outcomes. health care costs. and medical liability. Table 4.2 illustrates provider's opinions about who they would want to make decisions about treatment for a minor ( sJrains, strains, cold/flu symptoms) health issue. One internist reported somewhere between "only your patient" and "mostly your patient with input from you", depen
PAGE 193

Clinical efficiencies and tpproprateness. Questions about clinical efficimcies and appropriateness generated the most Physicians emphasized that there would be a vecy positive impact on clinic efficiencies for patients who participated in decisions about treatment for minor heal issues. That is, the number of medically inappropriate visits, the volume of unne ssacy phone calls, and clinical encounter length would all be reduced over time. Anything that's minor is to be self limited and therefore would not require muca put from a physician. In these times of physicians, especially s ecialists, seeing more patients with more severe diseases, t they have in the past, ultimately we will have less time to deal patients with minor complaints. complaints the more effici nt will be our utilization of our time. (Allergist/Immunologist) As far as a demand man ement issue that, certainly, we've been talking about for years. e need to change patient demand or it's going to be impossible to ontrol costs. A well-informed patient who demands care approp ately, that's going to turn out to be a good thing. (Internist) Probably it would have a ositive effect on clinical efficiency. Hopefully, there would be less time spent in the office, lower bills and so on and so because mostly patients can do that kind of stuff at home. lot ofthat from the practical point of view becomes easier for tile patient, saves them an office visit, saves the time in me for thf office for other more serious problems and so forth. So it's probbly good. (Ophthalmologist) I think it's tremendous. I t in terms of giving patients the right tools to understand things understand what's serious and what isn't, you can avoid a lot o the inefficiencies that occur as a result 80

PAGE 194

of mundane complaints co g in to an office and clogging up an offices precious time and aking away from patients who have much more serious complaints t at necessitate direct patient contact. (Internist) I think it probably would rfduce, to some extent, the number of office visits, and, the number of hospital emergency visits as well, that are inappropriate. I think in that, if patients learn about illnesses and can re'*ember that, and use that the next time that illness arises then theYj don't have to go through the same scenario all over again, su e, it should make it more effective -things that really are self "ted, and could be handled at home. One of my pet peeves, in articular, is sprains and strains going to emergency rooms, and sitt g there being untreated, no first aid, a patient with a sprained e sitting there with no ice, no ace, the ankle is down on the floor getting worse for two or three hours because they're not going o be seen early in the emergency room, they're put down on the b ttom of the triage list. It's a total waste oftime. (Pediatrician) To be a little nihilistic, for r lot of these things it doesn't matter much what particular you choose. So, I guess, from a notion of resources I'd spend more of my time advising people about matters of gr ater importance that would have a greater effect on their heal h outcome than to advise them about relatively minor matters. amily Practice) I think that if you have pe pie who can basically do first aid with minor illnesses it does help with office efficiencies in that it enables the internal office to keep the schedule open. The maintenance visits lthe chronic illnesses, a large part of what we do, the physical e s that people need to help maintenance in that regard And then the urgent day to day illnesses that come up than to have them clogged by people who are coming down cold symptoms or early flu symptoms or minor strains that they clm treat themselves. (Internist) 181

PAGE 195

Doctor-patient relationship. Ph sicians believed that patient involvement in minor health decisions would contribute o an enhanced, more positive, collaborative doctor -patient relationship. I would hope it would m e it better. I think this whole idea of the doctor being on a ped stal and mandating what the patient has to do, I think that wa a bad relationship. I just think it's the give and take and a far m re pleasant relationship to do it this way. (Internist) Probably an improvement because for minor little things that they could have taken care of emselves. For me to ask them to come in here and wait an hour i my waiting room would probably ruin my doctor/patient relationship. Not ruin, but they'd be pissed. I would be too. (Family Pr ctice) Even though physicians had a positive in ,Jination toward more significant patient involvement in making decisions about treatment for minor health issues, they also viewed patient encounters for minor issu s as an opportunity to build the doctorpatient relationship. The relationship between patient and a physician is, I trust my physicians' style. I like the way he approaches these problems. You don just talk about their cold or their sprained ankle when they come in. You talk about other issues, not just their lillness, but what's going on in their lives. The more contact you have with a patient, the more you develop a and a trust, and a knowledge base. So I look at encounters over minor illness as beneficial in those ways as seri us illnesses in building on our relationship in some fashi n. (Internist) 182

PAGE 196

But I think they're getting more than just the treatment for their minor health issue w en they come in. If they're anxious about their illness, what e er is bothering them, then you can put their anxiety to rest. d that encourages the doctor/patient relationship, it enforces t t. (Internist) Although physicians were predo antly positive about patient involvement in decisions regarding minor health issues, t ey expressed two concerns that patient involvement may have a less than optimal impact on the doctor-patient relationship. First, some were concerned that, while gi g the patient a more prominent active role in the decision making process the physi an might appear to be a "wet noodle" or "indecisive", ''uncaring" or "unconcerned in the eyes of the patient. I think it's definitely a plus It all depends -lots of times you'll get patients when you say, 'well, what do you want to do?' and immediately they're g ing to say, 'well, you're the doctor.' I think that puts them off. I think that's kind of a poor way out of it. If you couch it prope ly, where you don't get too -a lot of people are busy, they wan to come in and get out and get back to their life. I think if you ouch it in a sense that, okay, this is what the problem is. I thi the best way to deal with this is to do this. What kind of ork do you do. Do you have any responsibilities that could ave impact on? 'Yes, I have to do this, that, or the other and that could really be a problem if I do that.' Well, here arr some of the other options, it may effect its ability to it may effect the outcome, but it may allow you to do those! things. 'Well, I'm not interested in doing that if it's going to affect the outcomes,' or, 'okay let's do that and maybe I can back and see if it's progressing too slowly, we can try somethirg else.' I think it has definite, positive impact on the relationship, as long as you don't look like a we noodle, like, 'well, you make the decision.' I think it's got t be the type of thing where you really 83

PAGE 197

have to approach it from consumer point of view. This is the best option, this is the w rst option, what's your lifestyle? (Internist) Patient preference. Another fear expressed by some physicians was that many patients may not prefer to take on an a ve decision-making role, even with minor health issues. If forced to do so, there ay be a negative impact on the doctor-patient relationship and an accompanying redu on in patient satisfaction. That may be a little bit p ient dependent. I think one of the things that we're seeing i that many patients do want to take more of an active role in g for themselves, or their family members, and they woul like this. However, I think there are some patients who still p efer kind of the paternal physician. They really just want to I ve everything up to the doctor, and if the doctor's trying to them to become more active in decision making, and they're not terested in that, that's an imbalance. They may resist that and e less satisfied. They may feel the doctor's asking them to dttoo much. (Pediatrician) Those people who are us to having physicians tell them what to do in a situation with health problem would probably be less satisfied with a sy tern or an approach that causes them to be a little bit more self. reliant, but I think most patients would not have a problem (Allergist/Immunologist) There will be a lot of pati ts, we know this to be true, patients who want to be told wha to do who are very frustrated by having the options presented to them. (Clinical Psychiatrist) Patient satisfaction and Physicians believed that a more collaborative approach would provide p lents with more confidence and control in managing a minor health issue and woul generally result in higher patient satisfaction. 184

PAGE 198

I don't know what the real significance is, but certainly it empowers the patient a bit and I think most people feel positive about that and gi fS them more control over their own health. (Ophthalmollgist) I think that improves as I said earlier, because informed participating patients feels that theyre empowered and that they're in which is what many people want. (Internist) Health outcomes. Many physici s noted that patient involvement in decisions about treatment for minor health issues w uld have neither a positive nor an adverse effect on outcomes. Most physicians hel the opinion that most minor health issues were self-limiting, would run their natural course, and that patient participation in treatment decisions would have no signi:fi ant effect one way or the other on clinical or functional status outcomes. I don't know that health o comes are at all effected in the minor category. I think that my ew of these categories is that people, in are not going t 1 have adverse outcomes, or better outcomes whichever is taken. (Internist) Since most of the minor illilesses are going to get better on their own anyway, I don't know as it's going to make a big difference on that. (Pediatrician) I guess the way I feel abou this for minor health issues is most of these minor issues are s If-limiting no matter what therapy is selected. (Family Practice) At the same time, physicians were conce ed that some patients might mistakenly per-ceive a major condition as a minor one, d would not appropriately benefit from physician input. They frequently comment d that patients should always have the 85

PAGE 199

option and feel comfortable consuhing wtth their physician at any time about "minor'' health issues. So, I think they should hare some involvement with their physician. I think, after a "fhile, they should have the -knowledge to make a fair number of -fuose decisions without input from their physician. But, they shouid always have that option if they have a question, for whatever rbson, that they can call a physician. I think as patients become ;mewhat more educated about illnesses, minor illnesses, and under tanding them, that they could make decisions on their own. B t, hopefully, would still recognize that there are certain times whfn it would be important for them to check with their doctor. fthey're not sure about what seems to be a minor illness, and the want to be sure about it, they should still have that availability o check with their physician. (Pediatrician) But I think that if patients can take care of things themselves and they are minor, and they ay need help in deciding that they're minor, then they should. I think that would only improve our relationship. (Internist) Minor, which is there, is open to interpretation. I've seen sprains and strains that are more serious than fractures, and we've all seen what sJem to be cold and flu symptoms, which tum out to be seriohs illness. I think that an informed patient can make bad decirions. They need some input. I think that it should be a partnership. (Internist) To me it's always the decision ultimately. But I'd like them to make that decision with me. If they're decision is different from mine I'll respett their decision. (Clinical Psychologist) Health care costs. Many physiT postulated that a more active involvement of patients in minor health issues would reduce health care costs. 186

PAGE 200

Generally, if people take e of things themselves it costs less. If they come in it costs a Httle more. (Internist) I believe they would go Less office calls fur minor problems. I think also in the ofthings that ifyou're thinking about cost of health care involving time away from work and avoidance of accidents and so forth that that should also decrease. Generally I would think the overall cost to society of health care should decrease. Health care costs, I think, ould be less when the patients would be able to make decisions n minor health issues with relatively little input from physicianJ. (Allergist/Immunologist) I think it will decrease hejth care costs. With managed care, it's funny because if there's a co-payer, or a $15 co-payer come in here, you would think twiee if you had some sniftles and you were going to pay $50 to see mf. as opposed to $5 to see me. So, the copay has made it such that is no financial liability for coming in here, or no significant finahcialliability. But I think if I had to pay I $50 to come in for a cold I'd think, "gee, what can I possibly do. Should I look up in my Cdlumbia physicians' and surgeon's guide what I should do for this?'! It's where managed care has actually, I think, tremendously physicians financially. I tell people, this is something you _can handle at home. You don't need to see me for this. (Internist) I think it definitely should feduce health care costs because it should reduce some of visits to either physicians, or emergency rooms for things that really are self limited, and could be handled at home. (Pediatrician) However, one physician was adamant that patient involvement in minor decisions and the implementation of a patient support slrvice like TBDS would cost physicians money. Well, of course we give away billions of dollars of telephone time, we collectively. We discussed it at a meeting last night 187

PAGE 201

as a matter of fact. The loss. We're the only profession in the world that doesn't charge r1 or telephone time. We will continue to lose money, hand over oot, as we have in the past. It will never change. If they don' have to come into the office it will save them money. It will lose us money. In terms of the big picture, having them take bare of it themselves or with some telephone advice is least to the system. But somebody pays for that and that's me. So whenever we have any telephone advice it will always be frck and that's just the way it's going to be. So healthcare costs, it will have no impact other than we will lose money. It won't cost the any money, it just means we lose money for every minute spend on the telephone. But we have been willing over the years to give that up and I guess will con tinue to be willing to do that. (Pediatrician) Another physician expressed the view tjt patient involvement in minor decisions was being driven solely by a cost managemenl objective without due consideration for patient well being. Economic reasons, this is product that's being driven by, 'how can we lower health care costs'' rather than necessarily always what's best for the patient! I got an invitation to come to a meeting and this is what Jere going to do, saw a lot of business leaders in Rochester, and I have to say I think this is being viewed as business issue rather thkn as a patient welfare issue. I think the decisions are pretty driven by the bottom line, that's what business is about. It's beeh seen for a couple of years that dollars are more important than dutcomes sometimes. (Internist) Negative!disconfirming cases. J contrast to the majority opinion, two physicians, held the opinion that decisionl about treatment for minor health issues should be made mostly by the physician 1th input from the patient. Part of the rationale behind their opinions centered ound the relationship of what they 188

PAGE 202

considered "minor'' and the implied resp nsibility of the physician to take a more predominant decision-making role in rellvely more complex clinical conditions. I don't take care of that J ..... y minor things. Minor things to me is like a that doesn't need to be taken care of tlris week, but can wait a couple of weeks from not(Gastroenterologist) I don't treat a lot of cold d flu symptoms, but all sprains and strains are not! minor, at least the ones I get .... I'm the next of treatment of the minor things. So I don't see as fuch of the minor stuff Usually when it gets to mr it's more severe. (Orthopedic Surgeon) Patient preference was another part of tHe rationale for the gastroenterologist's opinion. Even when the condition was Led to be minor, the gastroenterologist believed that patient preference is for thj physician to tell them what to do. When I go to my doctor flr minor things I tend to want to be told what's to be done and to have it straight forward and simple, and I don't want to involve discussion and all pf the options for minor things. Often times the patient population I see anyway, they want to kno-lv, "Is it a minor thing or is it a major problem?" I They want you to answer that question. And then if it's a minor thing then they mostly just want me to tell them, I "what should I do about this?" They want to know about alternatives some'tirites, but it's mostly just ... it's much more along the fmes, ''tell me what I should do about this to rltake this better. I want it to be fixed and go away and be done with it." 189

PAGE 203

Even though not as straight forward, th orthopedic surgeon implied that for health issues that were clearly determined to b minor, the treatment decisions should still remain mostly that of the physician with put from the patient. The really minor I would tend to go with number 4 (mostly physician with from the patient) .... But if I do see the minor, I'd like a more infonned patient. If they have a contusion o they put ice or heat on it? They should put ice on There are still a lot of people out there putting heat on and their ankles swell even more. So yeah, the need for ed cation is there. But at my ... I Guess when it comes to rp.Y level, my own problems, I don't invest a lot of time, defining the problems. Summary. Physicians differed in terms of their preference for involving patients in minor health decisions. How ver, the primary tendency was towards the patient having a more prominent role in oecisions regarding minor health issues. I Respondents emphasized that there would be a very positive impact on clinical efficiencies for patients who participate in decisions about treatment for minor health issues. That is, the number of medically inappropriate visits to their office and the volume of unnecessary phone calls woul be reduced. This would, in turn, allow h . :fi fth . I . h illn M p ys1etans to ocus more o err t1me on pat:J.ents Wlt more senous ess. ost physicians indicated that having the patilnt more involved in minor health issues would reduce health care costs. Although physicians were predo antly positive about patient involvement in decisions regarding minor health issues, here were two distinct concerns expressed 190

PAGE 204

about patient involvement having a less t an optimal impact on the doctor-patient relationship. I) The appearance or perion that by giving the patient a more prominent role, the physician would appT indecisive or lack understanding and concern. 2) Many patients may prefer not to take on an active decision-making role, even with minor health issues. Physicians felt that a more collaborative approach would provide patients with more confidence and control in managing a minor health issue and would generally result in higher patient satisfaction. They also believed that patient involvement in decisions about treatment for minor health issues would have neither a positive nor an adverse effect on outcomes. An exceptioJ to this may occur when a patient mistakenly perceives an illness or an injury as minor rl hen in reality it may not be minor at all and would appropriately benefit from physicia input. 5. For a major health issue thospitalizatlon. chemotherapy. surgery). who would vou want to make decisions about treatment? \Explain how this would impact: the doctor patient relationship. clinical efficiency. hkalth outcomes. patient satisfaction. health care costs. and medica/liability. I Table 4.3 illustrates physicians' oJinions about who they would want to make decisions about treatment for a major chemotherapy, surgeiY) health issue. One clinical psychiatrist indicated Js opinion would fall somewhere between I "mostly your patient with input from you' and ''you and your patient together." Two internists, one allergist/immunologist, on ophthalmologist, and one pediatrician 91

PAGE 205

indicated their opinions would fall some here between ''you and your patient together'' and "mostly you with input r. your patient. Table 4.3. Physicians' opinions aboutJ:ho should make decisions about treatment for a major (hospitalization, chemoth tpy, surgery) health issue. Only your Mostly your y(] u and your Mostly you Only patient patient with pa tient with input you input from you to1 ether from your patient (61 Internists (2) Internists (2 FamPrac (1) Opthalmol ( 1 Gastroent (1) Nephrolog (1 Clin Psychi (1) Peds Similar to decisions about treatm ed for minor health issues, physicians differ terms of their preferences for involvin! patients in decisions about major medical making for major health decisions trend primarily towards the physician having a more prominent and paternalistic role. Doctor-patient relationship. Ph sicians generally agreed that patient participation in decisions about major he th issues would enhance the doctor-patient relationship. I think it definitely impro es it. If you sit down with your patient and you discuss oz;tions, which is what we're supposed to do --providing info ed consent before someone undergoes a surgical procedure or c emotherapy. Discuss options, then it goes a long way in terms 1 f developing trust. (Internist) 192

PAGE 206

The doctor/patient relatio ship, again, I personally feel is enhanced. I guess you have t ask the patient that, but from my point of view this enhance the doctor/patient relationship. I think it's better understan g of each other's needs, and a deeper level of respect when one i willing to take the time to talk to the patient and recognize their needs and their desires to be involved in a relationship. I think t at enhances the relationship. (Ophthalmologist) At the same time, there was a strong sen ent that the physician should play a more prominent and directive role in these type of decisions. Many physicians believed that severity of illness influences the degree o patient and physician involvement in major medical decisions. It was a widely held op on that the severity of major medical issues required greater physician involve ent. The obvious difference is tpat involving the physician is going to be more irnportanf as the illness becomes more complicated and more critifal. The patient is going to have less ability to think clearly, and to think rationally and to remove fear and thr at from the decision-making process as the illness beco es more severe. (Internist) Some patients do not perc ive the severity of their asthma. Those patients need more input from the physician. (Allergist/Immunologist) I think that in major illness most physicians are going to find themselves, not necess1arily equal participation but more significant contribution from the physician. The key word here is acuity. You don't bkve time to coax the patient into a participant's role. If thinks are changing rapidly, there has to be a fairly prompt decisibn made as to what the next step is. (Nephrologist) The level of information n ed is higher. The level of sophisti-cation. the amount of inpu from physicians that you need is 93

PAGE 207

greater. But at the same time there are clearly different points where the factors in makfug decisions are not medical, but are more personal. (Family Ptactice) Patient preference and sati.ifactibn. Physicians were of the opinion that patients generally like to be involved in ljor medical decisions; and that, when in volved appropriately, patients should be Lore satisfied. However, physicians made a subtle but significant distinction regardink the relationship between patient satisfaction and patient preferences for involvement L the decision making process. Many phys icians emphasized that there may be a rJge of major medical decision-making preferences by patients. Some patients Jay prefer a more passive role ("I prefer that my doctor tell me what to do") while sot may prefer a more collaboralive approach ("I prefer that my doctor and I make the (iecision together''). It really depends on the pLent. Some patients, which are really older patients, may kant you to make the decision for them. And I still want inp t from them but I have a mostly younger patient populatio so they want to make a lot of the decisions. They are relati ely educated, so they want to make decisions. (Internist) It's a Welbyish way ofbe' g to say that patients want-there are a minority of patients -kho would really want me to take an iron hand and just makf decisions for them. But in the long run even those patients be dissatisfied if they felt like all the decision was taken away from them. (Internist) I I think those issues are going to fall under either category 3 or 4 depending, a little bit, ort the medical sophistication, and willingness of the patient tb become involved in the decision making. It's the --how the patient feels about becoming actively involved in decisi 1 n making versus just saying, "you're 194

PAGE 208

the doctor, go ahead" and "you tell me what to do." I've been in practice for 25 years, and when I first started practice I think a far greater numbJr of my patient's were in the, "you're the doctor, tell me t do" mode, than I see now. But I still see a significant percentag who feel they would like me to do most of the decision g. But more and more who would like to be partners in the decision making. (Pediatrician) Several physicians conceded that if clinicir fililed to be sensitive to the patient's own preferences in decision-making preferences for major medical decisions, physician patient discordance may occur, resulting i lower patient satisfaction and a less than optimal doctor-patient relationship. I think in most instances people are satisfied when they're involved. If they feel they're getting f fair evaluation of where they're at. lfthey feel that the advice.rou've given, ifyou laid forth the reasons why you've given it, are going to be satisfied as opposed to not knowing. (Internist) Again, for most patients, satisfaction should be improved for those patients who want a collab rative relationship. For patients who won't want a collaborative elationship they wouldn't like it. (Clinical Psychiatrist) There's nothing worse a patient who really wants to partici pate, trying his best to with an authoritarian physician of the old school that says, 'don't ask me any more questions. I've told you exactly what you should do.' And that was the way our grandfathers, to some practiced a lot of medicine. (Nephrologist) Some subtleties here. I think patients revert a little bit to the, 'you're the doctor, I'm the patient' mode, as they get sicker. It's kind of a regression, it's almost a parent/child kind of a role, 'I don't know what I'm do g, you do, I'm sick. This is scary, threatening to me, tell me hat to do.' They may be more satisfied with a physician w o doesn't keep throwing it back in 95

PAGE 209

their lap, but rather takes a more active role in, trust me. I'm going to help you throu this. We're going to get you better. I'm going to make the decisidns, and you're going to do well. I think that you kind of start to in to that as you go along as that patient's physician. How :rpuch, how able they are at that point in time to participate, but satisfaction, obviously, is going to vary from patient to patient. patients are going to want to be told, and some patient's are going to want to be involved. It depends on desires and the patient's (Internist) Many patients want in the decision making process. And I don't think it's the 1ght thing to do to just hand that off to people who are, in many eases, less prepared to deal with even the concepts. It isn't that I to take their right to decide away from them, but in major illness,! considering the short time factors that are often involved, it doesn't seem to be fair to just look at the patient and say, 'okay, it's your chll.' That's the dark side of patient empowerment. (Nephroldgist) If they're involved in that lhoice and they feel comfortable with the choice that they'+ making and they know that they can always back up and go another route, then they're I satisfaction will be high. is a minority of patients who would really want to take an iron hand and just make decisions for them, but in the long run even those patients would be dissatisfied if thY felt like all the decision was taken away from them. I can prbsent my cases forcefully to a patient about what I think needs tb be done and I can render my opinion, but they still nee to be involved in that process. I think they'll feel better, ev the patients who want me to make I decisions for them. And they'll feel more satisfied. (Internist) l Additionally, physicians frequently comm nted that they would respect the decision of the patient even when they didn't agree lth the patient's decision. It's really 50/50, me and tJe patient. Where, really, the reality is my opinion is very valuJd by the patient and that there's 196

PAGE 210

probably more of a directire component there than I feel com fortable with. (Family Prafice) When we go to the major pealth issues, I think we're dealing with issues where we may[ have therapy that is truly more effec tive that we need to push for in the medical point of view. For example, this week I had a patient who had a heart attack at home, didn't want to go td the hospital, and I had to tell her I think it's the wrong decisibn. But if that's what they want, this is what we need to do what your choice is. I think in this situation I'm much more likely to say, these are the things that have been shown to improre your longevity, or your outcome. I think you need to be these. What do you think about it? And then let them modify my treatment plan by their expectations, rather than put the out and say, 'it doesn't make any difference, you decide.' So, rm more likely to direct situations there. I still want them iniolved. I think people still need to understand that they need o be part of it, and they need to buy on with it. But, I think, where a major health decision where I can give people solid evidence of change and outcome, I have an obligation to tell them that this is the treatment plan that has the best outcome. When I do that, it is more directed than throwing out the options t saying, 'choose one.' (Internist) I think I have a responsibility to my patient's to tell them what I feel is clearly a best treatpent. Or to tell them that I don't know, and let them know in making sure they have a chance to modify that plan based 9n their values system. I think that that's beneficial, rather than being absolutely directive saying this is what you have to dot or being totally open ended and saying, what do you want to do about things. I think they need to see that you have dlear feelings on it, make it known that that's what you want, you're willing to modify. People respect that, I think. (IntenUst) I think there it becomes a 5[10/50 decision ---actually the decision is going to be the patient's. The patient has the final eventual decision making -going to make the decision, let's face it, I'm not going to make the decision but I think it's a partnership and we're both going to do it. They can make the 197

PAGE 211

wrong decision, in my eyes, and sometimes they do. But they sometimes have valid rea.kons. A woman was here today, she keeps coming back rotator cuff tear. She's 84 years old and believes she is too frJil for surgery. I have to respect her opinion on that. I point dut to her though that just last week a I 78-year-old man who's come back 5 months after surgery has full range of motion, no dam what so ever. Perfect result. I point that out to her. if she makes the choice not to have surgery, that's her choice.\ I'm not going to force her into it. There's some people I don't respect their decision. They're making a foolish decision! But it's their decision. I continue to support them. It's their to make that decision. It's their body. I tell them I wouldh't advise them making that decision, but I respect their right t1 make it. (Orthopedic Surgeon) One pediatrician aptly summarized the whole perspective about doctor-patient re lationships and patient satisfaction in to patient preference for involvement in major medical decisions: I think, and this is part of rhat I think is what the art of medicine is. And that is, is the physician appropriately reading the patient's desire for\involvement, and matching his response to that desire? And if the physician is able to do that, then that should enhJnce the relationship. If the physician and patients aren't cdmmunicating that need appropriately, then there will be problemk. Let's say, if the patient really I wants to be more actively decision making, and the physician doesn't pick that up and adopts a more paternalistic approach, then the patient's not going to be satisfied. If, on the other hand, the physician feels tliat it's important for the patient to be completely involved in the decision making and starts I providing the patient with F,ormation and expecting patient's to make decisions, and they're not ready to do that, and they don't want to do that, thenj that's a miscommunication in the other direction and that coWd cause problems also. So, I think, the key here is for physiciah and patient to sort ofbe able to 198

PAGE 212

read each other, and to each other where they are in that desire to be in1olved in decision making, and then there would be a good fit and a good match. Health outcomes and complianc Most physicians contended that outcomes They cited improved patient compliance, control, and empowerment as the mechanism largely responsible for improved outcomes. And, I think that the style of care that I like to deliver to my patient's is such that they feel a little more empowered. They feel that whether I can affect the circumstances of their illness directly or not, I will be thb-e with them through that illness. And I think that when spend the time with your patient that way, I think, that improves outcomes. (Family Practice) An informed patient, an i4olved patient, is a more satisfied patient, and emotional outcomes, and clinical outcomes are improved if the patient is iltvolved. (Internist) I Generally, I would expect that outcomes should be improved. If patient's their major illness better and the implications about the treatment and how important it would be to continue, often times, these treatments would long term medications, and the use of those, and tlle patient uses them because they understand their illnesk better, then their health I outcome would be better. rediatrician) I would hope that health outcomes improve in this situation. If you give peopie enough information, and I that they buy on to a plan that has medically proven efficiency, I think you're better off. I think if people are engaged in their decision ,th proper direction that you'll get better compliance and buy on to a treatment program. (Internist) 199

PAGE 213

It very much improves it. ren you're in the hospital, you're wearing that you don't have your clothes on, you feel you can't do !htything for yourself In order to do this that doctor has sit down with you and go through a set of options that a patient has to make a decision. There are choicds, so that's empowering for them. It really helps peoplb. They have some control over their illness rather it just happening to them. (Family Practice) Clinical efficiencies. Physicians ixpressed conflicting opinions regarding whether or not involving patients in decisions about major medical issues would be more efficient. The gamut of opinions jged from those who believed it would definitely be inefficient and would take mire time to those who believed it would definitely be more efficient Others beliejled that involving patients in these types of decisions may take more time initially, bu in the long run it would be more efficient and would result in higher patient satisfjion. This may be less efficient in terms of dealing with patients rather than absolutely diredtive approach. It's going to take more time in the procbss. I don't see how this gets you away quicker than a only' approach to things. (Internist) It's more efficient to get somebody involved early that has the knowledge base than it \would be in something that may well be self limited. I think it's always more efficient to get the patient involved early That's true. across the spectrum. (Internist) But often times, I think, pafients have a lot of good common sense. If your job is to presf.nt it to them in an understandable way, and if you succeed in hat at all, then I think sometimes 200

PAGE 214

it does improve clinical e, ciency. You can hone right in on productive. (Gastroenter logist) That should be improved, because the patient should be able to -if they're more invol ed they should be able to make some of the decisions, or hnderstand the decisions more I quickly. Now, there may be sort of an investment oftime up front on a major issue this, where it may take more time to educate the patient. But that, I would think, would pay dividends later on becaus5 then the patient would be better able to manage the care and you wouldn't have to be trying to constantly re-explain or get the patient involved later on. (Pediatrician) 1 Health care costs. Physicians eXIjressed uncertainty and mixed opinions about what effect involving patients in decisionk about major medical issues would have on health care costs. The spectrum ranged Jom those who were of the definite opinion that patient involvement would reduce heLth care costs to those held an equally determined opinion that patient involvemt would increase health care costs. Well, unfortunately, it's of unclear how that's going to play out. Because, I thiilk, patients want everything done, or a lot of patients ,ant everything done for things that aren't necessarily Needing an MRI scan for their headache. (Internist) I I am not so confident that it does anything for health care costs. I think it may health care costs. I think the more informed people are more likely they are to seek access to good care and the less likely they are to be satisfied with less care. The more ?}formed people are the better informed they want their d0ctors to be, the more specialists r

PAGE 215

they're going to want to see. The more they're going to want things like cholesterol scrJening and PSA determinations to be done. (Gastroenteroloii.st) For a major health you're talking about something like hospitalization, or surfery the cost is going to be the same whether the patient has made the decision, whether I've made the decision, or whether made the decision together. I find that many patients, when they understand the situation, in fact, will comfortably defet certain problems, that would be in the elective situation. In non-elective situation, I don't know whether there's a difference or not. There may be no difference whatsoever in hbalth care costs in that area. With cataract surgery, for example, as an elective procedure, many patients who are :functiorurtg satisfactorily, as long as they understand that the surgerY is available but there isn't a reason to do it as long as they legally drive, and their visual needs are being met every day. :rytany patients, I find, are happy to defer surgery. (Ophthalmologist) I think health care costs be reduced because it will provide for better treatment, is, hopefully, more optimal treatment which usually would be cost effective, because patient's presumably under the should recover more quickly and should have less complications. (Pediatrician) Health care costs, again, I lhink there's no doubt that it's going to improve it. Again, are less likely to fly into the ED if they know what to expect from a treatment course from their I illness. People are less Iikefy to be ringing the nurse's bell and being sent to the ICU because no one really knows what's going on with the patient over weekend. When they're nervous about something that they llave in regards to their heart failure, their heart attack, their pnehmonia, their cancer. So I think that would help drive down heaith costs. Anything where the patient is playing a role in the and having an understanding in their disease, I think is going to decrease cost. Conveying some information to a patient doesn't cost a cent, it really doesn't. Everything else does. Havihg them come into the ED because I they're nauseous from the chemotherapy medicine That's going to 202

PAGE 216

cost money. Or having sent to the CCU because they're having some twinges of chest pain and they're not sure that it's their reflux or if it's their eart and the doctor never explained which is which so, 'we'll, send them to the CCU.' That's going to cost money. Somebod who has pneumonia, that's on a home treatment and was never what to watch out for the signs of, suddenly gets very dizzy and passes out and is found cyanotic at home, where they werejfeeling sick three days earlier but really didn't want to bother That's going to cost more money. Any situations where you intorm the patients about what to expect with, not only about a but what the complications from the disease can be short and term, and what the complications of the therapy can be short and long term. It's always going to save money because there will less phone calls, less panic reflex with regards to their illness. (Internist) Medica/liability. Physicians affi1ed that involving patients in decisions about major medical issues would definitely redtce medical liability. However, "involvement" was defined as more than j st the rote process of informed consent. Involvement meant being able to establis a genuine relationship with the patient as a whole person. In included a relationship t at would allow the physician to understand patients and their entire and, to empathize with them as well as educate and infurm them about treatment rptions, benefits, and risks. When physicians practice with this style, medical liability is reduced, even when the outcomes are not as positive as expected. Well, my own sense about that is, medical liability decreases, almost exponentially as an inverse to the degree of investment in the relationship between he doctor and the patient. (Family Practice) 203

PAGE 217

I think it lowers it. I thinl(: that it's trust building. It's getting the patient and you on thJ same team and participating in decision making as a I think it really lessens your likelihood of any malpractfce because you're working as team rather than butting II think a lot of malpractice issues probably could be headed off with more of a team approach. (Internist) J The informed consent do rine is that patients need to be explained the risks and bertefits of any treatment and the risks and benefits of any alternative treatment that might be available. If you fail to provide that your patient you're failing in your obligation as a physician and you're subject to medical liability I under the law." Providing informed consent can be a very emotionless event that really help you connect with your patient. I think it's not so much involving the patient in decision making as emoting with your patient. Because there are lousy doctors who have bad outcomes who never get sued because they're good to patients. They're just nice people who listen to their patientsJ And there are great doctors who I have the best gifted hands in the world who will get sued because they're asses. So, I think that more has to do with -not so much sitting down and risks and benefits, but sitting down and empathizing and emoting with your patient. (Internist) As I tell our medical studeJts, 'you should make a friend and I partner' of your patients.' .t}nd that, generally speaking, that patient will be a satisfif. patient who will not sue you even if a goof-up is made. Gener:ally they won't do that. (Orthopedic Surgeon) The more you inform your patients, the more they have an understanding of what options are and why they chose an option, there's going to less concern for liability. No one can really hold you liable if they were involved in the treatment plan, if they werel involved in the decision process. How can somebody sue yoJ if .. 'Yeah, I'm upset with the doctor because I had my entire removed instead of just the lump removed' 'According to thJse notes the doctor discussed with you the options including 1Jmpectomy, radiation, and mastectomy. 204

PAGE 218

Your words at the time ,ere, 'take the whole thing off: I don't care.' 'Apparently you made the decision with him.' 'Well, yes that's true.' 'Do you feel tou were infonned? What our options were?' 'Yes, but I was vef.y upset at the time.' 'It says here also that you didn't have to the decision right now, you could call him back in two or three days.' I think as long as you're infonned I and you understand, you literally put yourself in the shoes of the patient and say, 'how woilld I feel in this situation. What would I want to know?' That's situation where you reduce your liability dramatically. You're going to reduce it to zero. There's always going to be somebody whb wants to sue, that's going to happen if somebody slips on your :frbnt step. If you're terrified of it, you can be paralyzed in providing care. The most you can do is docu ment what you do and be as informative as possible to your patients. That's the most you can db. (Internist) I think there's a very big blnefit here for medical liability reduced problems, because, I when patient's start thinking about suing most of the time it's because of communication problems between I them and their physician. I think if physicians involve and engage patient's in decision then the communication process, of necessity, has to be there's much less likelihood for misunderstanding, and the patient then later on saying, 'well, the doctor never told met s or that' and 'if he had, such and such, wouldn't have happehed.' Then you're more likely to get a suit. (Pediatrician) Summary. This group of physicians differed in their preferences for involving I patients in decisions about major medical issues. In contrast to minor health issues, preference for decision making of major hLth decisions primarily trended towards the physician having a more prominent and palemalistic role. Physicians contended that . 1 . d" al d .I. ld h b d pattent mvo vement m maJor me 1c eciStons wou ennc t e octor-pattent relationship and enhance patient satisfactiJn if clinicians were sensitive to individual patient preferences for involvement in declsion making. They believed that health 205

PAGE 219

outcomes would be improved and the po entia! for medical liability would be reduced if k 1 d .\. b .t:'. di al patients too a more active ro e m eas1ons a out treatment J.Or maJOr me c Ph'' d ld nfli' .. b hth tssues. ys1c1ans expresse uncertamty an co cttng opiruons a out w e er or not I patient involvement in major medical decisions would improve clinical efficiencies or reduce health care costs. j 6. What do you think about your patien -s taking a more active role in the self management of chronic disease conditiofzs (e.g. asthma. arthritis. diabetes. hypertension. low back pain)? Explain hdw this would impact:.the doctor-patient relationship. clinical efficiency. health Diltcomes. patient satisfaction. health care costs. and medica/liability. This group of physicians were unanimous in their belief that patient involvement in the self-management of cJonic disease was absolutely essential for optimizing health outcomes, patient satisf1ction, doctor-patient relationships, clinical efficiencies, cost management, and in redjcing the potential for medical liability. Health outcomes. Physicians _j emphatic that a more active patient role in the selfmanagement of chronic disease jould result in better clinical and functional status health outcomes. I think this is the area where it can substantially increase the health outcomes results. Diabetes in particular. Asthma secondarily. But, yeah, the patient does affects the I outcomes more than what the physician does. Sometimes I'll have a patient that's got k poorly controlled chronic condition who I'll send to a \specialist and I find that they don't control them any better than I do because the problem is not the knowledge base ofthe medical provider, the problem is the patient doesn't have t e knowledge or the motivation to 206

PAGE 220

do what they need to do. So chronic diseases most of the management is self management. (Family Practice) The only way that patientl do well with chronic disease and that you get good health Jutcomes is if the patient's become dramatically involved with their weight loss, with their diabetes, with their hypertension. Bfcause, usually, with these diseases they are major life changes they have to do to take affect and the patients have to be in bharge of it. Otherwise, it doesn't matter how many times ydu tell them to do it, if they're not actually doing it and gettirlg involved realizing themselves what needs to be done, just not going to work. There's no way you can get good outp>mes unless patients are dramatically involved with their own clironic illnesses. (Internist) A chronic illness is for the rest of their lives. It's not something that's going to go and it's largely dependent on their compliance, and insight, and their understanding, and their buy-in to a treatment plan for them to do well and I minimize complications. li'he more they know and the more they participate, the better they're going to do in general. (Internist) Now, if they've got a proolem such as diabetic retinopathy, that may be a different but to the extent that they're educated about the and about the types of things that they can do in terms of control, proper medication use, physical exercise, follow-up with the doctor who's managing their diabetes, thbse kinds of things, I think, those I patients clearly do better. fhe health outcome, I think on these patients it's better particularly with the chronic problem. Obviously, diabetes for exJmple, someone who doesn't under stand the nature of their is likely to be a poorly I controlled diabetic. With the types of chronic eye problems that we see, these patients ho far better with an active role in self management. (Ophthalinologist) I think of chronic disease, as arthritis, where management techniques are largely techniques, methods to control symptoms, to control their functional capacity. Medically 207

PAGE 221

their joints may be shot in the same amount of time no matter which approach is taken, but if a person has learned to cope with their disabilities, witH proper access to arthritis association I programs and techniques for managing their disease, they may ultimately need a hip at the same time, but between here and there they may have functioned a lot better. (Internist) They're going to be less Jely to be in the ED, less likely to be in the hospital, less likely to get their medications confused and more likely to have a course of the disease. I think, in the outcome, have better morbidity and mortality in the long run, especially with astruJtics, diabetics, COPD'ers, and even Alzheimer's patients. satisfaction, same. (Internist) Patient satisfaction. These physidians had consensus that patients taking a more active role in the Jf chronic disease would result in higher patient satisfaction. It's enhanced, once again. Diabetes, for example, 'when Patients have an understanffing of their illness, know how to use their glucometer, adualiy know what to eat and what not to eat and are not tethered by the telephone with I every blood sugar that's abnonnal. They actually know what to do. They love it. l.\nd they'll be grateful for it. They feel much more in cohtrol of their illness. (Internist) I would think that would Jcrease because, again, there's a great satisfaction of doink something yourself and realizing that it has been ydur own input that has helped and made something betterl Whether it's fooling around in the stock market or cuttfug your lawn or taking care of yourself. (Internist) It's higher. lllnesses are disempowering. Decision power is empowering. So I think beople feel better about their illnesses no matter where they are in the extent of their illness. (Family Practice) r

PAGE 222

Doctor-patient relationship. M y physicians indicated that, when patients take a more active role in the self-managlment of chronic disease, it improves the doctor-patient relationship through enh ced trust and the removal of barriers to communication. No doubt on the doctor-patient relationship because when someone has a chronic those are the patients you're going to be seeing the most of And they're the ones that are going to get very aisgusted with the medical system unless they feel like a true advocate out there for them. I think that's where the physician really has to play a role As their advocate, an und9rstanding what it is that they want. So I think it has a very positive impact on the patient physician relationship beca!use they can really feel like, 'this is my doctor, he knows that's going on.' They don't feel like they have to ljuggle things between six or seven different doctors or eight Hundred different medicines saying, 'well, I have to trust that s guy or this gal knows everything that's going on.' I think it as nothing but a positive impact on the patient-physician re ationship. (Internist) I think it's very positive. I I hink the more the patient is educated the more they talk to the doctor with comfort. The more the physician treats the patient like a peer the more the communication fall between them. It substantially enhances the doctor-patient relationship. (Family Practice) I Patient preference. Individual lees in patient preference for in-volvement in chronic illness care were mentioned by only two physicians who made b bib" lc. .. I hr "illn nunor comments a out t s emg a concern LOT patient mvo vement m c omc ess care. 209

PAGE 223

In most cases it will enhance it. It's going to worsen it in those situations where people iant to be treated as opposed to involved. Some people d1 n't respond to that approach. Some people are very much, the doctor, tell me what to do' approach. I think the majority of patients are going to respond to it positively once they buy into that system. (Internist) For a majority of the physicians, it was Lost as if it were a given, a necessary com ponent in the continuum of care that patiLs be involved in the self-management of chronic illness. Clinical efficiencies and cost management. Many physicians remarked that this is definitely one area where a more Jtively involved patient can have a significant impact on health care delivery system effilencies and cost management without sacrificing quality of care. That's one place where I tlililk that you clearly can show a benefit. Studies show a rltore efficient delivery if you empower the patient to mJnage their care, their chronic disease. That's one that I don't think there's a lot of argument with. (Internist) Clinical efficiency, this is Pfobably where I think the biggest savings are. If you have somebody who has an understanding of their discltse process, they're not going to call you every time their blbod glucose is 205. They're not going to show up in your dffice every time they cough up some yellow sputum. not going to fly into the ED every time they get a little i.vheeze, if they're an asthmatic, or the family's not going tol rush them into the hospital because grandma's getting confused again and left the stove on and almost burned the house ddwn. These are the patients, that if they have a better understanding of what the disease process means, that it's chronic, it's probably going to get worse with 210

PAGE 224

time but we can control you can avoid diabetic acidosis, you can avoid hypoglycemia _1 hich are both things that might require more clinical cost of time and effort and so forth. So I would think it would be better. (Internist) I think that would lower care costs. Fewer office visits, fewer complications. (Internist) What I've seen in studies Jays improvement. Less dollars because of recurrent acceJs to a higher cost things by allowing people to it own their own. (Internist) This is one lruge area for health care costs. I was just looking at an article on diabetes management, and a very, very large percentage of care costs go to deal with the complications of diabetes. \Its been shown that if you can improve the patient's of their diabetes you can reduce these health care costs by huge, huge amounts, billions of dollars for the (Pediatrician) Self efficacy. Physicians attributJ improved health outcomes, patient satisfuction, doctor-patient relationships, tlinical efficiencies and cost management primarily to the mechanism of enhanced patient self-efficacy. They explained that when patient takes a more active role in re self-management of chronic disease, they feel more in control of their lives and the disease and therefore more empowered to deal with the disease in a positive fashion. Confident, knowledgeable, and in control, which is, with chronic disease, largely what determines a patient's perception of their disease. (Internist) If they have an active role ip their health, 'look, you have this illness, it's a chronic illriess, it's not going to go away, but you can take an active in controlling it and this will allow you to control your life.' So, the diabetic who 211

PAGE 225

knows that when the sugar moves up to 250, they have a sliding scale, they can adfi.Ist them, they don't have to pick up the phone for the doctor and they can still go out with their friends or they can still gb to work. They don't have to end up in the ED. I think can handle it, that's very empowering. (Internist) So they're more productire and, once again, they feel healthier, they feel happier that they're in charge, that they're managing an illness. (Internist) I They're much more satisfied if they can control their own outcomes. Because they \feel as if they're in control of their own body. People, especially with chronic disease, like to feel that way. (Ophthalmologist) Medica/liability. Numerous physicians asserted that the potential for medical liability would be reduced significantly +en patients are more actively involved in the self-management of chronic disease. Patients who have a better understanding of their when things do go wrong. I think the more people have bought into a system on their own, and are given choiccls, the less likely they are to accuse you of being the cause a bad outcome. People aren't going to sue themselves because they didn't manage their disease properly. (Internist) The medical system, you mow, just treated her as a disease. Once she felt like she was 1\somebody who had a disease and she had a role in her care, she was hardly ever in the ED. And I knew in my heart of hearts no matter what happened to this lady she would never kue because she was involved in her health care. (Internist) 212

PAGE 226

If they're taking a more aC1:ive role in taking care of themselves then liability would go doWn. They wouldn't be hanging on every decision that I gospel and wouldn't, if there was a bad outcome with a decision, they wouldn't always I be looking to me. They'd be looking towards themselves. I think it would, in all these casesi I think a more active patient involvement would decrease liability. (Internist) I think that the relationshiJ is enhanced and the potential for liability is less. Patients tatkmg an active role is a positive all the way around. It's the Studies show that patient's who have a good rapport with their doctor are much less likely to sue. They may well be better care. I find that patients, for example, lot's of diabetics who are interested in their disease may know much more abobt their condition than a number of the health professionals they 'fme into contact with, and will evaluate any recommendation you r-ake on their fund of knowledge, and may suggest somethinp to you which you hadn't thought of, or makes more sense iq their particular situation. They just end up getting better dare, I think. The likelihood of a problem is somewhat lessdned by that. I don't know what studies would show in this\ area, if it's ever--ifliability costs versus patients taking an active role have ever been looked at, but I would strongly that you would see a lower liability cost. (Ophthalmologist) Summary. Physician informants 4ressed unanimous support for the utility of patient involvement in the self-managei of chronic disease for optimizing health outcomes, patient satisfaction, doctor-patient relationships, clinical efficiencies, cost management, and reducing the potential fJ medical liability. Out ofthe three d ab I. 1 <. illn scenanos use to measure opwons out anent mvo vement mmor ess, maJor illness, and chronic illness), physicians asized that patient involvement in the

PAGE 227

management of chronic illness held the most value and would have the most significant overall impact on health care system qu1ty, cost, and efficiency factors. This, personally, I think il where something lnce patient involvement is going to have thej biggest impact on all of these things. With the minor stuff it's going to be a minor impact, with the major stuff it's going to bb a pretty good impact, but this, I think it's going to have the biggest impact. (Internist) Opinions about Telephone-Based DecisiL Surux>rt The next series of questions (Q 11 0) included more specific questions regarding physicians' opinions and perceptions of a 24-hour telephone-based decision support (TBDS) service that had been iJplemented for 80,000 families two months prior to the completion of these intervie s. Eighteen physicians reported that they had heard of the TBDS service and were familiar with its purpose and availal)le services. The remaining three physicians I reported that they had heard of the TBDS service but knew very little about its purpose and available services. Physicijs expressed a range of opinions about the TBDS service: generally positive (7), netral (11), and generally negative (3). I 7. Please describe how you think the telephone-based decision support service will impact vour patients and your practice. Physicians differed in their opinions about the potential impact of implementing l h b dd .. .I h. d o a te ep onease ec1S1on support servtce on t err pattents an practice. ne opmton group (n=9) indicated they were unceJ or "not sure." 214

PAGE 228

. My patients know about it, theyre talking about it, they're accessing it, they're in and telling me that they've used it. Not sure of the nrtpact on my patients. Whether it's been a good thing, or a bad thing, because it's just getting started. (Internist) I don't know that much ab0ut it so it's a little hard to know how it will impact. It's ... I'rtt not sure. It would be nice if it took away some of the phone calls. (Internist) I b .I . am not sure ow tt s gomg to unpact my patients m my I practice. For eye issues, I don't have a sense at this point, for example, what percent bf my patients may call with ophthalmology issues to a like this. My guess would be that it wouldn't be a number and I would guess there wouldn't be a large impact (Ophthalmologist) Another opinion group (n=9) said that the service would have "no impact" on their patients and their practice. Many in this group were of the opinion that they personally fulfilled the role of the service Jy providing their patients with the information and counseling they needed tjl be more active, involved consumers. They viewed themselves, as one physician put it, as a "TBDS supplier" for their own patients. I don't think it will. I get a lot of phone calls. They kind of pile up during the day and I renlm them after hours. I think of myself as a "TBDS" supplier becadse the patients call me up and they ask these questions that I'm they ask the people at the TBDS service. Perhaps the only thing is if there was a 1-800 number, patients might feel freer to call. TbJy may feel somewhat inhibited to call me because I'm busy. I haven't\found that to be the case though. They still seem to call. I talk with them five, ten minutes on the phone for 2 5

PAGE 229

what ever problems theyt'e got and we make a decision whether they should come in or if we can take care of it by phone with just some simple advice. (Internist) I think it will have a minimal impact for our patient's in. our practice because, I suspebt if we're providing the kind of information that we should be providing for our parents, they would have I little need to call the TBDS service for further information. I And, we have for each ofJ our well-child visits a one page handout talking about hea.Ith and safety issues and feeding I issues. We also have probably at least a dozen disease related handouts how to manage \children with colds, how to manage children with fevers, diarrrea. head injuries, what ever. We have handouts that explain aboft those things that we give to those patients when they make the visit with us. We also make fairly hberal use of the Academy of Pediatrics pamphlets on a whole range of subjects, and we\have some that we handout specifically, and a whole bunch that we have in both the waiting rooms and the examining rooms for to pick up and peruse or bring home. (Pediatrician) There are four of us that share on-call and we've always encouraged patients to ca.h us. I've received several calls this weekend. In other wbrds, one of us covers for the whole practice. Much of }vhat we do is give out advice on the telephone. I'm not so rure it will make any tremendous impact on our practice. O*ternist) Patients with real big we see to it that they're pretty well informed, if they're to that, through getting those patients connected with Alnerican Liver Foundation Local Chapter. We think the TBDS service is a commercial venture, as I see it, fulfilling a need\ that has arisen that is fulfilled largely by volunteer organizations. The TBDS service does some of the same things volunteer orgkuzations have been doing for patients. Patients have sought this of thing out for quite a while. Patients get information frbm these organizations about their disease, past what they siJply get in the doctor's office. (Gastroenterologist) 216

PAGE 230

The third opinion group was a minority of physicians ( n=3) who felt the service would have a "positive impact" on their Jatients, mostly related to clinical efficiencies. Their positive reactions to such a service \generally clustered around the issues of clinical efficiencies and the doctor-patienJ relationship. I Clinical efficiencies. Two physicians in the "positive impact" opinion group I said the service would improve clinical efficiencies by reducing the number of unnecessary phone calls and visits as well as the average length of the clinical encounter. I think overall it's going to be a plus just, hopefully, to cut down on the number of phbne calls, and as an educational tool for the patients. I don't thirik: the TBDS service's goal is to make I radical medical decisions. [ think they're just there to basically to inform you of options are available for particular illnesses or to provide you with with regard to particular illnesses or questions, and I think that will enhance it. Average length of the clinibat encounter it may decrease that mainly because if they \did contact the TBDS service and they say, 'you really should see the doctor because there's a concern for this', they'll be I much more definitive for what they're coming in for. Thet called up and they said 'this is what I should do.' I actujy had an experience with somebody who did use the TBDS seryice, or used the book that came along with it, and looked up something about their eye and it wasn't getting better and called me and were very clear on what they wanted. It actually reduce the length of the encounter. The volume of Calls to the office -it probably will reduce that. That's probablt the biggest impact there. It's those little annoying calls that yolii get about, 'I was bit by a mosquito and it still itches', like that. (Internist) I I think you may be able to decrease the amount of triage, probably half of the triage questions are problems that really 217

PAGE 231

could very easily and ju 1 as effectively be handled by just by it. half that would call be able to really focus it down to the important information. Nkses that triage would be able to get at the problem in a more fashion. (Family Practice) There was an interesting dichotomy in tJe views expressed about clinical efficiencies mong physicians of the same opinion + On the one hand, there were several physicians who expressed a "not sure" or a "no impact" opinion who also expressed a hope that the service would have a impact on clinical efficiencies by reducing inappropriate or unnecessary phone calls I I dont know that much aoout it so it's a little hard to know how it will impact. It's .. .Ijm not sure. It would be nice if it took away some of the nuisance phone calls. (Internist) On the other hand, the same physicians J the "not sure" and "no impact" groups commented that there was no such thing L inappropriate or unnecessary phone calls or visits. Furthermore, they viewed thesJ so called "nuisance calls" as opportunities to build on and enhance the doctor-patient rllationship. If the doctor and the pati.L are meeting over a concern that a I patient has, it's an appropryate visit. No. I don't think there's such a thing as an inapprofriate medical visit. As a country and a society, you have to [decide how they want to spend their dollars, but is it an appropriate visit to a doctor if a patient has a concern, yes. It's like saYmg there's a stupid question. If you have a question that needs\ an answer, you ask it. A lot of visits that maybe weren't there are a lot of interactions that didn't necessarily fix somdhing that needed to be fixed, but they address that need for the at that point in time, and those limited dollars and limited hme, and limited access, you have to 218

PAGE 232

decide what's more appro1priate and less appropriate, or more important, and less impottant, but not inappropriate, and appropriate. It's a continuum. I'm sure the calls that would have t.o me will to as I've before. Over mmor illnesses I think mjh of the time.tt will turn out to be okay. (Internist) I'm a finn believer that 95f, 98% of patients are very reasonable, they're responsible and it's the 2% or 5%, probably 2% that call for everything. It's hard to ctlange their behaviors if their anxiety based behaviors. It's hard to kn?w what type of impact this service would have on the nuisance phone calls too. Example of that would be, often times when a patient with a nuisance phone call during the day, it's simple dvice, a 'yes, n0' or 'do this', typically I'd give it back to one of our senior staff to ban the person back. Inevitably, if it's one I of the people who call a lot I get the chart back on my desk, 'answered question, they kill want to talk to you.' And you tell them the same thing and it's likb, 'okay, I just wanted to hear it from you.' (Internist) So, in my population, I don't want to discourage patients from calling me. I don't calls my front desk and they say they want to talk fo the doctor about a little pain in their leg. I don't think I want my secretaries to say, 'have you tried the TBDS service?' to try to decrease my phone load at the end of the day. I consider those to be key to insuring our relationship and providing patient \satisfaction. There's a possibility it could be detrimental to out relationship because it kind of throws another provider in there, tmd I don't want to inhibit my patients access to me. (Internist) I think that anytime you interact with a patient you're building a doctor-patient and then if the interactions went down with certain patients, beca'lilse of using the TBDS service, that could I perhaps negatively effect the doctor-patient relationship. Perhaps it would be positive, at for the doctor's side of it. I've always thought that's one of the rriain reasons for doing a physical exam on patients, is the health mhlntenance issues you work through. But a lot of the value in physical exam, in my estimation, is the 219

PAGE 233

time you spend with the patient interacting with them, building a relationship. (Internist) Doctor-patient relationship. Other physicians reiterated that any positive or negative impact on the doctor-patient rjationship would fundamentally resuh from: a) how the physician positioned the se]-le with his/her patients and b) whether or not the information provided by the service 1 ould support the physicians' viewpoim. It depends on how the phrsician handles it. If the physician says, 'oh, what the hell did you1call them for?' 'They don't know what to tell you. It's going to 1 ave the patient more confused than anything. (Internist) I think it all depends on h wit's used on the other side of the phone. I think, if it's used as an e ucational tool it could be helpful and supportive of the doctor afd it could improve outcome. If it's going to create confusion in a Pjtient to create second guessing, it could obviously drive a wedge llletween the doctor and the patient. (Internist) Summary. Physicians differed in feir opinions about the impact of implementing a telephone-based decision surport service on their patients and practice. There were three distinct opinion groups rh a majority (n=l8) of the physicians falling into the first two: I) not sure, 2) no impact, and 3) positive impact. Many physicians viewed themselves as a decisiol support "supplier" for their own patients. Reoccurring comments clustered around lhe issues of clinical efficiencies and the I doctor-patient relationship. There were divergent opinions about the impact on clinical efficiencies. Some physicians in + "not sure" and "no impact" opinion groups stated that clinical efficiencies would be improved through a reduction in unnecessary 220

PAGE 234

calls and office visits. Other physicians the same opinion groups believed that there was no such thing as an unnecessary call or visit. Furthermore, these same physicians viewed calls and visits for simple self-c i issues as opportunities to enhance the doctor-patient relationship. Many physicians reiterated that a positive or negative impact on the doctor-patient relationship would primarily result from: a) how the physician positioned the service with his/her patients and b) whether or not the information provided by the service wo+ support the physicians' viewpoint. 8. What will be the most significant challenges vou anticipate following the implementation ofatelephone-based decision SUJ!POrf service for patients? Physicians asserted that the two lost significant and related challenges facing h I f I I d . 1) h .. t e unp ementat1on o a consumer: pat1ent ec1s1on support servtce were: p ys1c1an endorsement and utilization in the everyd y practice of medicine and 2) patientconsumer utilization. Physician endorsement. Resolute concern was expressed that physician buy-in and support for the community telephone-based decision support service had not been sufficiently obtained. I think adoption by the physician community. If we embrace it and refer to it and encmftage our patients to use it, then I think it would only be subcessful. If your physician didn't refer you to it, patients be distrustful of the information otherwise. If an insurance company said to call this number in an effort to avoid emergency room visits and avoid medications and avoid doctors visits, I think that people would be very distrustful with that infontiation. So, a challenge would be, the physician community e1 bracing and using the service. 221

PAGE 235

Referring their patients I told you before, I don't feel comfortable with that yet because I'm not sure I trust the information yet. I think t at's the biggest challenge. I think that there's no advertising that would work other than physician referral. That's the one kr.1 element for the success of the service. (Internist) I think at the community evel, for physicians in general, I think there is a challenge for them to understand the role of the service. Some physicians' are comfortable with patient involvement in decision making than others, and they -some physicians' are I not going to be happy with being 'second guessed', and particularly if they make some for a course of treatment and the patient's may then service and learn about some other options and then call physician back and say, 'hey, you said do this, I heard threugh tlie service that I could do that.' It may not sit well with some ptcians'. (Pediatrician) Community wide the mos' significant challenge has been sus picion, professional about inappropriate or un professional interference in the physician/patient relationship. Along with a concern for people in that does the information meet community standards? Is it in line with community standards? That it's not a one size fi, all. (Gastroenterologist) If I see my patients getting advice that I have a conflict with, then I would be unsupportive. (Internist) It all really depends on I suppose the reason that this is being done. In other is it to just help patients and educate patients, on the one hand, rhich may be the most benign use of it, which I think there still may be potential for interfering with the doctor-patient (Internist) Just the one other thing tht physicians' in the community are wondering about is, how this being paid for? Everybody's talking about saving healtH care dollars. How is this project being funded? I hear a lot !of rumors, I don't think I really have any facts. I don't that being discussed in any explicit way at that informational meeting I went to. In an era where 222

PAGE 236

we're trying to contain cof!s, what is the cost of this and what is the benefit of this, needs to be evaluated, not just by the medical leadership that you're talking about, but I think the physicians' who are buying into it neJd to know what they're paying. If in fact they are. (Internist) I don't want to alienate pJtients in the sense that if you're sitting there and talking to them bd at the end of the conversation you I say, 'and there's the decision-support service now and you can call them.' They're goingjto feel like, 'he's trying to give me the brush off here. Is he sayiqg that I shouldn't have called him for this problem?' You don't \want to have that disjointedness. I'm trying to figure out how te promote to my patients without making them feel alienated, I thinlt: that's a concern. If you say, 'now you can call this 1-800 numbe{ instead of speaking to me.' That would make me feel like 'ph my god, a 1-800 number.' You can talk to me, but this is an option. It's got to be couched in a certain way. It's got to bel something where it's informational, general information. Shouldn't be something that's touted by the I physician as, 'Are the HMO's making you do this?' That type of thing. I think it's got to_bJ couched in a way that if you have any health concerns, things that. (Internist) I will make one very and affirmative statement right now. And that is, the to which this succeeds or fails is going to be determined 'o a large extent by the enthusiasm and participation of the phsicians to take care of their businesses. If the care givers say, 'oh, this is a bunch of crap.' It will be very difficult to the patient's the message of support that you're really trying to Again, I'd make the point that the key dynamic is how reeeptive the physician is to empowering the patient by sharing kno.J.rledge. The limiting factor of this is still going to be to what eJtent the physicians feel that they are able to participate and thefre going to be able to facilitate the I patients in learning and helping. In other words, the potential is that both patients can and physicians' but they both have to come to some understandibg. If I had any reservation about this project it would be that it rbight be just an incredible benefit to have, even at this point, a +ore focused physician education component. I could tell yo that the success or failure of the 223

PAGE 237

whole thing is going to hilige on that. And if it fails it isn't going to be because the concept[s wrong, it's because you haven't got the kind of quality participation that you need to make something I like this show its merits. rephrologist) Patient/consumer utilization. Physicians emphasized that without physician advocacy for the service and without intJgration into the daily practice of medicine, patient utilization of the service would njt be optimized. With sub-optimal consumer/patient utilization, the overall -vllalue of the service to all key stakeholders (physicians, patients, insurers, employers may be compromised. Possibly Patients are used to being able to contact their physicians offices and some people will have a hard time accepting the idea of advice coming from somebody with whoT they have not had any contact. I think that would probably be a potential problem that would have to be overcome before it was widely accepted. (Allergist/Immunologist) The most significant challenge that I see is getting people To understand what the is and to keep it up front in their mind when they begin to have questions. So, I guess it's consumer education about the service is the I biggest challenge. Maybe I should be saying physician education about it, but I tHink the bigger challenge is to have people plugged in and physicians would have to fullow under those (Internist) Probably the biggest one is going to be getting patients to actually pick up the phone land use it. If you think about yourself, if you fell and twisted your ankle, you had a cold and it's been 4 days and not quite getting better, who are you going to call? If you Have a very good relationship with your physician, you're gouig to want to call your physician. (Internist) 224

PAGE 238

First I think it's getting patients to use it. Getting patients to use a valuable source infonnation. It's new, it's different. It's not something that's Pfrt of their health seeking behavior. Frankly I think that the biggest boundary is just that, getting people into the habit of thlnking about this as one of the things they do when they feel a medical issue instead of flexibly calling their doctor, whicH is what they're used to doing. (Family Practice) Summary. The two most significant and related challenges facing the 1 f I d I. . I) h .. unp em.entatton o a consumer pattent easton support semce were: p ystetan endorsement and integration of the servif into the day-to-day practice of medicine and 2) patient/consumer utilization. The primary reasons cited for a lack of widespread physician buy-in and support included: 1. Lack of extensive participation in the development, and decision to implement the consumer decision support service. 2. Insufficient information or misunderslding about the service. 3. Inadequate experience with the servil. 4. Unfilmiliarity with the protocols 1 by the telephone-based nurse counselors. 5. Potential conflicts between the information provided by the telephone-based service and the physicians' opinion abbut appropriate utilization of medical services. 6. Suspicions regarding the primary aim of the service. 7. Uncertainty regarding its value in tenJs of impact on patient health and overall cost-benefit. 8. Uncertainty regarding how to "positior" the service to avoid alienating the patient within the context of the doctor-patient relationship.

PAGE 239

The primary reasons cited for lack ofpa ient/consumer utilization included: 1. Lack of personal physician support, ltilization, and referral to the service. 2. Lack of full understanding by patien\ s about what the service is. 3. Introducing a new mind set and a ne behavior to patients in regards to accessing health and medical infonnation from someone other than their own physician. 9. How do you propose these challenges be addressed? Recommendations for addressing the challenges anticipated following the implementation of the telephone-based decision support service naturally fell into two primary categories: a) how to augment Jhysician endorsement and integration of the service into the day-to-day practice of mldicine and b) how to enhance patient utilization of the service. Physician endorsement/support. Physicians frequently conceded that creating change and spreading a new il among physicians is often "complex and problematic." They also acknowledged tJat compounding the problem in this community initiative was the monumenJ challenge of communicating with and obtaining a critical mass of support from Jhe approximately 2, 000 physicians who comprised the two independent practice Jssociations. They pointed out that mailings I and meetings don't generally work for informing physicians and for obtaining participatory buy-in and support for a nej idea. Additionally, they acknowledged that there will always be a certain subset of who will be reluctant to change irrespective of how well you infonn or en1age them in the change process. However, 226

PAGE 240

informants made several practical recommendations for amplifying physician endorsement and support of the decision support service. I. The most consistent recommendatim1. was to establish strategies for tapping into the already existing formal and infoAnat network relationships within the existing physician organizations. Physicians Jlked about nurturing these fonnal and infonnal networks in more meaninJl ways to engender more physician d &: h I. p () partiCipation an support .10r t e serwce. nmary attention was gtven to: a working with and through the exiJ IP A subcommittee structure in the design, development, implementation and ev1uation of the service and (b) identifying and working with a few key influential p1sicians who are recognized as opinion leaders among their peers. The value If informal physician network communications such as the hospital doctor's lounge should be recognized as a b1 h.&: din I. ab th VIa e mec antsm .10r sprea g commurucations out positive expenences WI the decision support service. j I think we've learned man times the hard way that with any kind of a new program, ydu need to have physician involvement in the design and of the program for it to become accepted. We fun into this all the time as we're trying to improve of providing care, and managing care. You may have a idea but ifyou sort of try and say to the docs., 'do this', doesn't work. But if you say to the physicians, 'we think that fuaybe we can improve the way patients with asthma, or dikbetes in this community are treated, and that a lot of the are not individual physician related, but more systems related. \How do you think the system can help you provide better c 'e.' Then, they get involved and make '2.27

PAGE 241

suggestions. Then you can say, 'we had a group of primary care physicians, and asthma sdecialist's who sat down and they worked on this and these where their suggestions. And then the physician community says, 'okay, I\know, so and so, the internist', and 'I know this asthma specialist, and if they think this is the way we should do it, and that's gding to make things better, okay.' But if it comes from the insurbce company, or some other nonphysician source, it's proBably not going to be accepted. (Pediatrician) It's real hard to make an argument to put the manpower into trying to get out to eight+n hundred physicians on some kind of an individual basis. I think you have to think about what are your efficient kind of like a telephone situation. Say there's a bomb down the street and Joe Blow I calls somebody else and clills three or four other people. I think that if you think ab9ht the existing structures that are there in the medical community and whether you work through medical staffs or whether work through chiefs of departments, departmental meetings; ways that you can give a presentation to 30 physicibs, 15 physicians or 50 physicians. That's the way that I wouid do it if I had to make some choices I around resources. The other piece of that is that I would decentralize that as much as !possible. And in my own work with networks, even just the fact that I was willing to go ten miles and five miles and fifteen fnes to meet with docs at their parent hospital and not expect to come to me went a long way at showing good faitli. (Family Practice) Where we find that we gel the best payoff is when we get down I to the sub-committee level. The 30 sub-committee's representing different physician specia!ks. That, generally, breaks it down into a small enough group\ so that they can understand it and they communicate it to their colleagues, that works pretty well. The only one group that we hare a little difficulty with is the internist's, because there are 400 of those, the sub-committee may be a dozen or so, it's just not big enoJgh to communicate to the rest of them. Most of the other sub-conlmittee's --they represent specialist's that vary in size from 30 to 100-150, and then a sub-committee of 10 or 12 can communicate to many, but to communicate to 400 is a 228

PAGE 242

little trickier, and we're working on ways to try and do that. But I think the sub-committee tbvel, for them to truly understand what the new strategy is, and buy Jto it, that seems to be our biggest payoff. (Pediatrician) I don't know, we tried, sp 1 eaking as a medical director for the IPA, we tried hard to communicate that sort of infonnation, and we had reasonable turn outs.\But, communicating any kind of infor mation to 2,000 physicians is not easy. As a people, they're hard to get together for You know darn well that mailings only accomplish so much, and ffyou can't reach them through the mail and you can't get them to \come out to meetings, how do you communicate to them? AS usual, the best way is through word of mouth, and probably the cbmmunication that takes place in the doctor's lounge in the is probably one of your more effec tive methods, but it's also 1extremely time consuming because you can only reach one or two I docs at a time that way. I think you know what will happen is that solme physicians will become hopefully more comfortable with this type of service and actually see that it helps them take better care patients and say to a.colleague, 'boy, you know I had this patient who had diabetes and they just wouldn't do anything I told them to\do. The decision support service kind of laid it out to them how imE. ortant it was that they manage their sugar closely, and, by golly, this atient really now is working with me and wants to do that.' I think ose kinds of experiences repeated often enough will spread the wotd and that's when the service will become more valued by physicians (Pediatrician). 2. Establish a physician oversight committee with key physician opinion leaders to help guide the initiative. l The continuing communic tion to the physician community I in terms of having a physician oversight committee. That's such an important thing. I \got at least a half a dozen nasty letters about this demand program. Everyone of them after a personal phone call was, 'oh, I understand. Okay, let's see how it goesJ' So having meaningful ties with authentic physician or physician leadership is very, very important. Docs can overtly or covertly consciously 229

PAGE 243

or unconsciously do a lot to influence this. The emphasis is, more can be done to encourage the docs. (Pediatrician) 3. Educate and inform as many doctors in the field as possible about the nature, purpose, benefits, and value of the service to them and their patients. I I think that you'd have to some kind of an educational campaign directed at the that would encourage their trust in the service, and provide concrete examples of how physicians in othet communities have better, happier lives and get home and can actually see their kids because the service has influenced the number of calls they receive. (Internist) 4. Arrange opportunities for key physicians to "experience" the service first hand. Provide examples of how certain cls from patients would be handled. My concern is probably 4 same concern of other people. Even though we've been told a little bit of what's going to be said (by the nurse ), we just haven't heard it ourselves, if you lmow I mean? It is as if we'd like to call up and pretend we're patient, have a conversation, hear what they tell us and, j I think, unless I've kind of heard it myself how this is worldng, I'm not sure I'll feel totally comfortable with it. (Interhlst) 5. Afford key physicians and subcommibees the opportunity to review, approve, and even participate in deVeloping the tot 10-20 protocols that would be most frequently used by the nurse counsel0rs. I have not seen any protolls for the most common things that I would run up againsi. I'm curious as to what they are. You're going to have to lobk at the data and be sure they have data on the commondst calls that have been encountered. Then they have to take the! top 10 commonest calls and provide the protocols to t e physicians. You may only have 130

PAGE 244

to send the top 10, maybe back pain or something to the orthopedist, or a primary care physician and let them know what information is being given out. It will I be a lot of work but it will be worth it in the end. Credibility is the challenge. (Orthopebic Surgeon) 6. Provide frequent physician-specific rback about the perl'onnance of the decision support service in terms of rumber of patients, type of call, and patient satisfaction with the service. I Try and give people feedbrck on how it's being accessed in the community. I think that's going to make me feel better to know the volume is going if you're able to measure any impact on patient satisfaction, or on medical costs in the community that you give drompt, periodic reports back to the practitioner's in the comm'bty about how this product is effecting those areas. (Inttst) 7. Provide physicians and subcommittees not only the opportunity to participate in approving and developing the counJling protocols, but also an opportunity to influence the training and education lf the nurse counselors providing the service. One way of addressing thJ would probably be if physicians in various specialties, as as primary care physicians, took part in the education of nurses that gave advice. If the advice could be made, to seme degree, specific for each patient's physician, that may be difficult, but if somebody called in and identified herself as Dr. I patient, that some how there could be information some of the information that the nurses had at their could be specific for Dr. __ In other words, so the patient could be reassured that Dr. has approved\ofthe information that they're going to be handing to patients the phone. For example, if there was a block of information having to do with asthma, if that could be shared with those of us who take care of a lot of asthmatics and we could s 'well, that's good', or 'perhaps, 231

PAGE 245

I think that could be done a little bit differently, or a little bit I better', and I think it coutd be done on the individual physician basis, or in the tase of the IP A, it could be done by the allergy subcommittee. Each major practice, almost every allergy pradice in the area has a representative I on the subcommittee of IP A, on the allergy subcommittee. I think that would be a of making sure that those physicians who see most bf a given problem could have some input as to what kirld of advice is going to be given. Then I think they would more comfortable with it, and then patient's, in turn, would know that the information that theyve been given been approved by the IP A allergy subcommittee, or if they iCientify themselves as a patient of a pulmonologist, they can y the same thing about the IP A pulmonary subcommittee (Allergist/Immunologist) 8. Assist physicians in how to "positio '"the decision support service to avoid alienating patients in the doctor-pari nt relationship. To actively promote it yo 've got to be careful. It's kind of like advertising yourself in the DNC, it turns people off. Patients expect a certain of professionalism and advocacy from their physifian and if they feel like at the end of it you're saying, 'and now you can call this 1-800 number and get your taken care of without ever talking to me and you won't have to bother me any more', is how it will come across\ It's going to come across like an MCI commercial or something. I think it's got to be couched in a certain way l The glitzy. stuff can be sent to the house but for the physfcians' office, it has to have a professional appearance to it, something that says, 'TBDS I for our patients with concerns.' (Internist) 9. Involve and educate nursing and office staffs who play a key role in patient education. rl32

PAGE 246

I think a way to operationalize this might be to connect with Office managers as much more as with physicians because they are usually the people who kind of do that leg work and get those kinds of things Jut where they belong. Nursing staff would be another key of people to connect with because nurses often get involved in doing patient education activities. So if they sort of know this is a way that they can connect with something,_which wi make job easier, that might be another good thing. (Family Practice) I 0. ClarifY what exactly it is physicians +t and expect from such a service. I think it does need to be elear what the physician organizations want from thls program. If we want it to be strictly, 'okay, you're of twisting your ankle, here are the options that ybu have available to you,' they can give advice of, stay of it, elevate it, put ice on it, that's an option of what you can do yourself It would be listed in the book that was distributed to patients. Versus, another option is, you can call your physician. Call the emergency room and rela)"'l your symptoms to them. Is it strictly options based, and that's what the physician organiza tions want? Fine. If they want to be more directive saying you want people giving that's going to resolve challenges by giving clear guidance as to what physicians want. I see my perceptioimay be different than what the physician organizations a ally asked for in this. Because I do know that physicians, physicians in the community have been clamoring for a ,ong time about how to get a re duced number of phone calls, reducing the amount of telephone based management. Once there may be mixed messages about that. It's kind of, don't want the volume of phone calls but we also don't want faceless strangers dispensing information to our And yet a service like this is just that. The physicians have a conflict here of how to efficiently manage patient dare and not be tied down to the telephone day and and yet giving patients access to all the information (Internist) 233

PAGE 247

11. Complete _an opinion survey of all field physicians regarding their perspectives about patient decision support serviLs and suggestions for successful I implementation prior to program dtopment. I suspect before this was implemented it probably would have been a I idea to have had a significant questionnaire go around all the physicians, clearly the pediatricians, the family the internists who represent basically all thejdoctors that they would be calling about. I think tha they should have surveyed the Rochester physician group to find out how they felt about this sort of thing. Would they support it and if they support it would they have any suggestions on how to do it. (Pediatrician) 12. Assist physicians and employers in forming alliances and partnerships that will c. h r I opturuze success 10r t ese types o commuruty truttatlves. When you start with the in health care, it just changes all the rules and doctors are very worried about this because they're\ not business men. It may seem to everyone on the outside that they're business men, because they make a lot of money they're not business men. They're not used to how functions. There are different ways of forming 1alliances in business, and friendship means different things in ousiness than it does in the personal relationship or than it in the medical society. What is friendship, what is an alliahce? How do partners function? Same way that doctor's dJnt understand patients, I think, there's that same lack ofubderstanding between doctors and I business people. (Clinical Psychologist) 13. Develop a local consumer/patient dLsion support service franchise that is physician owned, directed and operatld.

PAGE 248

Probably, the best solutio? to this challenge would be that the local health care 9ommunity comes up with its own locally based program, protocol's developed and designed by local and staffed by locally based people. I know therr's a joint committee that looked at this program and then gave it the go ahead, both physician and based. individuals, and decided to go ahead It as the pilot trial. (Internist) Patient utilization. Recommendations for increasing patient utilization clustered around two basic strategies: a) physician support and integration ofthe service into the day-to-day doctor-ratient relationship and b) direct marketing and education campaigns to consumers/patients. Patient utilization is the and if you don't do it through the physician, if RCIP A, which I know will go through the physicians. But if their insurance company wants to bypass fue physician, it will back fire, because the physician will be distrustful. Patients will always call the next day to make sure that the advice they got from TBDS was dorrect. If they sense any distrust on the part of the physician, they just won't call again. They'll call the dhysician. (Internist) I'm sure utilization is a ch].enge. The question is what do patients know. What of patients know about it, and those who know about it, where's the number posted? Lot's of times when you nef
PAGE 249

do that kind of stuff. All the types of things that one would think of to utilize Ia new service that patients are not familiar with. ( Ophthalnlologist) Need to have a structure in the workplace that's reinforcing to peo:Rle. I think there ought to be a communicative in the physicians' office, in the physicians' waiting rootd. Otherwise, it's kind ot: well I work is one part of my life and my health care is another part of my life. You ;eally need to integrate that. (Family Practice) Summary. Suggestions for addressing tie challenges anticipated following the implementation of the telephone-based decision support service naturally fell into two . ) h lh . d d. f h pnmary categones: a ow to augment p ystCian en orsernent an mtegratlon o t e service into the day-to-day practice of edicine and b) how to enhance patient utilization of the service. Recommendations for amplifying physician endorsement and support included: I. Working through the existinJ formal and informal network relationships within the existing physician \organizations, especially the IP A subcommittee structure and Key physician opinion leaders. 2. Establishing a physician overlight committee of key physician leaders. 3. Educating as many in the field as possible about the riature, purpose, benefits, and value of the service to them and their patients. 4. Arranging opportunities for Jey physicians to "experience" the service first hand and providing examples of how certain calls from patients would be handled. 5. Affording key physicians and 1subcomrnittees the opportunity to review, approve, and even in developing the top I 0-20 protocols that would be most frequently usetl by the nurse counselors. 236

PAGE 250

6. Assisting physicians in how tq "position" the decision support service to avoid alienating patients in the doctor-patient relationship. 7. Providing frequent physician_\ pecific feedback about the pelfonnance of the decision support service l terms of number of patients, types of calls, and patient satisfaction with the service. 8. Providing physicians and the opportunity to participate in approving and developing counseling protocols as well as an opportunity to influence the training and education of the nurse counselors providing the service. 9. Involving and educating nurs1 g and office staffs who play a key role in patient education. 10. Clarifying what exactly physijans want from such a service. 11. Completing an opinion all field physicians regarding their perspectives about patient support services and suggestions for successful implementation prir to program development. 12. Assisting physicians and empleyers in forming alliances and partnerships that will optimize success for these types of community initiatives. 13. Developing a local decision support service franchise that is physician owned, directed and operated. Suggestions for increasing patient utilization included: (l! enhancing physician support and integration of the service into he day-to-day doctor-patient relationship and (b) direct marketing and ucation campaigns to consumers/patients. 10. What will be the most significant bejfits you anticipate (pi/owing the implementation of a telephone-based deJision support service for cnsumers!patients? What else would you like me to know abdut your thoughts regarding the value of the telephone-based consumer decision support service to the community? I Physicians indicated that the most significant benefit to the community would 237

PAGE 251

be increased consumer/patient access to good quality, uniform information about health and medical concerns. Access to good informati1 n, is the biggest benefit. If this can improve the quality o the information that's readily accessible, it's going to b a benefit. (Internist) And I would just say that this TBDS has that one strength. And that is t least you know the informa tion that they are giving is uniform. (Pediatrician) Physicians suggested that increased infoii ation access would ..;pport patients in making better decisions about use of he th care resources resulting in: a) improved clinical efficiencies, b) higher patient sati faction, c) enhanced health outcomes and d) reduced health care costs. Clinical efficiency. Physicians bllieved that clinical efficiency would result from a reduced volume of phone calls an medically inappropriate visits. As it stands now, reducing the phone call volume. That's the biggest problem in my office and probably most offices. We keep adding phone lines and people are answering them and they just keep cdming up. A lot of it is, hopefully, to reduce the phone volume the patients that are having chest pain, or coughing up blood can get through. That's a lot of it, is trying to get a better ha4dle on handling that phone volume. In the long run, I don't know if it's going to have much impact on anything else. (Int4 From the doctor point of'?ew, to the extent that it educates patient's and may perhaps save them time If patients are being educated by this ser+ce, and come in and actually the doctor spends less time with them, and the time is a higher quality time in the sense thltt they get right to the issues that 238

PAGE 252

are pertinent to that patient, that would clearly be a benefit. Obviously, both to the pa1 ients and to the doctors. (Ophthalmologist) Patient satisfaction. Physicians sserted that a more informed, self-directed patient would result in higher patient sati faction. I would like to see happie and more self directed patients. Would like to see less cal s for self limited disease. I would really like to bump-up the sophistication and tenacity of selfmanagement in patients "th chronic disease. To me that's the largest opportunity here. (Family Practice) more uniform health and medical inform ion would be a more informed patient which in tum would foster better illness manalent and improved health outcomes. We're hoping as patients nderstand both their minor and their major illnesses bett they will make better use of the medical care system, and oe less demanding of inappropriate services and more able to I manage their illnesses to provide better outcomes and less domplications. (Pediatrician) For those physicians who L.t providing their patients the information, it might uhprove health outcomes . But if there's no other access tb that information, then it would improve health outcomes. 1The patient would be more satisfied. (Internist) Health care costs. Physicians reiterated that more informed and involved patients would result in improved health Jare cost management without sacrificing quality of care and patient satisfaction. 239

PAGE 253

I would be happy if the bfnefit that we saw was a cost neutral improvement in P!ltient satisfaction. rd think it was a very successful prdgram if it didn't cost us money, and people were happier] I would also consider it a great benefit if we improved clst outcomes. (Internist) But if what they are doing is to support the doctor/patient relationship, then, by getting patients more involved in their own h th care could improve outcomes, and thus also, obviously prove costs.(Intemist) Doctor-patient relationship_ P,sicians held conflicting views about the potential benefits to the doctor-patient rionsbip. Some physicians were of the opinion that patient access to outside their personal physician, including a telephone-base decision support service, rould potentially disrupt the doctor -patient relationship while others held an equally strong opinion that patient access to information would support and stimulate higher quality physician-patient communications. Anything like that, any flow of information outside of the venue of physicians, I'm rlot saying we're a deity, but it can effect that relationship bedause the patients going to say to himself, 'I got this from Internet. I got this from Prevention Magazine. I got this from the decision support service.' The information fge, I think, in lots of ways can be the physicians' greatest enemy or greatest ally. It depends on how it's going to be (Internist) At least I was somewhat ebused about the service. I think it will help open up between physicians and patients. It's not going to tnake somebody who's totally non assertive and unable to the simplest question of their physician into a strong person that comes in and says, 'here, I want to have a distussion about what's wrong with 240

PAGE 254

me today.' They'll still be timid and meek and difficult to I deal with and easily rolle9 over. But, sometimes there are alternative ways of getting information. People may get information in that route. I It may tum a very ineffectual physician/patient relatioruJhip into a very good one. I'm very positive about it. ( oftroenterologist) I think that if the patients to understand that this is something that can be valf able to them they can sort of insist that their physicians become better listeners. It isn't a power struggle of who' going to instruct whom, but in fact, the patient walks in fd says, 'doctor, I'm not critically ill but I have certain concerns about this area and I want to pay for your apt attention \for about 7 minutes, not interrupted by phone calls or anything else.' The doctor is going to respond to that. The docter's not going to immediately jump up and say, 'I can't fulfill tease expectations.' But on the other hand if you go in and watch a doctor in the course of a day you wonder if they evrr think of that in the whole process. And I would hope that one of the benefits might be that the physicians will come to thoughtfully reflect a little bit more on the quality of communications that they receive and partici pate in. Let's say that I several patients who have made phone calls, and had absolutely excellent information and have come away with the perception that whoever was on the other end of that line really was to what I had to say and gave me their best effort, tp.oughtful individual, consideration. And they go in and they tell their doctor that. That would be incredible communication. The doctor would be unlikely to ever have it considered that the patient got a better quality communi cation on a phone line to a decision support service than right here in his office. So I would think that the possibility exists that there may be ... that patient's may drive more interest and receptivity from the pHysicians in what's really a very important two-way commJnication. (Nephrologist) Summary. Physicians indicated thjt the most significant benefit to the community would be increased consumerJr atient access to good quality, uniform 241

PAGE 255

infonnation about health and medical concerns. Physicians purported that increased infonnation access would support patils in making better decisions about use of health care resources resulting in: a) imJroved clinical efficiencies, b) higher patient satisfaction, c) enhanced health outcomjs and d) reduced health care costs. In contrast, physicians presented mixed vieLs regarding the benefits to the doctor-patient relationship. Closing Questions The last two questions of the interview were intended to provide one final opportunity for physicians to provide additional thoughts and opinions that may not have come forth during the intervie"l 11. What else would vou like me to know about your thoughts regarding a more active involvement ofpatients in health imd medical decisions? The issue that physicians most fr1quently mentioned was the fact that medicine had entered into a new era. It was their ew that this new era is one that is more consumer centric. They emphasized that this new era of consumerism will be characterized by advancing sophistication and rapid growth of communications and infonnation management technologies. cbnsumers will accelerate the access of health and medical information from various soirees such as the Internet. A majority of the h dhi d I ed =-c. ed d p ystctans vtewe t s as postttve an SURPOrt a more uuorm an active pattent while recognizing that informed and actiJely involved patients may 42

PAGE 256

require them to make changes in the wa they manage the doctor-patient relationship and how they approach physician-patieJ communications. Specifically, many believed that J would become increasingly important for them to hone their physician-patient colunication skills. Many talked about the need to take the time necess8I}' to have 1 higher quality dialog with patients that would include discussion about not only the biological aspects of health and disease, but also about the social and culturaJ Jors affecting patient's health and health care utilization. Many conceded that with smJe patients they will need to become less paternalistic in their approach. Physicianl will have to become more sensitive to patient preference for involvement in h.Jth and medical care decisions and adapt their style to that preference. They acknowl ged that with many patients, this may require patient relationship. The dilemma they regularly spoke of was how to do all of this in practice environments and health care that are becoming more volume dependent and productivity-driven every Ly. Second, they reasoned that with the advancing sophistication of communi4n technologies, providers will be required to expand their communication modalities :6:-bm the traditional face-to-face encounter to include interactions with patients via the Jntemet, telephone, and voice mail. One of the most :frustratinl things to me is the patient who comes to see me and rays, 'I have this pain, what is it?' Its like they don't understand the necessity of the questions and answers, gi ng as much information as 243

PAGE 257

possible. It's :fiustrating in that they seem to be, the expectation is, 'if I tell y9u I have a pain, obviously this pain can mean only one thing.' Where as, it's only through the process ef communicating, asking the questions, listening carefully to the answers and then feedback my analysiJ of the infonnation that I I can really help a person. jT o me, communication, active patient participatioh is the only effective way to practice medicine how a days. (Internist) I really think tbat I'm modem in my view that patients have a right to kdow what's going on in their medical life. And I think that it's our job to provide good education or at lead have literature and/or things that we can talk about to tnake sure that they're educated. So, I clearly think that the!1 days where the patient comes in and gets a prescription and doesn't even know what the prescription is, are gone. [hat's the way is was when I was growing up. (Pediatribian) And it's the same when thl bring something in to me. The latest newspaper this week was on this lipo protein subset A, or it was. I said, 'gee, you know as usual the newspapers got it before JAMA does.' I don't know about that ndw. Thirty years ago it would have been hard for me to 'I don't know.' I would have probably whiffed wafted and said, 'oh, it doesn't amount to anything.' But I think it's important that they understand that we don't tdnow a lot of this stuff either. But we're going to find out and we're going to do the best that we can to advise the,. (Internist) It should be about a broad [array of things that are far beyond the fact that doctors JUSt use symptoms as the excuse that gets the patient in that you can do what you want, what you need to d9 with a patient. All of those need to be expanded. expansion is, well, how do you communicate? Do yol communicate in person? Do you do it by phone? Is it by TV? Is it by videos? Where is it? How is it? So, I'd be r open minded and experimental I 144

PAGE 258

with regards to that. Different ones are great for different people. Think that there brobably won't be one modality that's going to work. If ytl>u want to talk to my father it's going to be through his cbmputer and not through the tele phone because he'll be on! his computer as many hours as he can. And he hates to on the phone. He would never call someone on the telephone. But if he can down load information about a health care issue he has he would love to do that. It should be multi-modality. (Family Practice) The way that we used to lransmit information was one to one about one issue. The tloctor to the patient about the symptom. That's the The doctor. The patient. The I symptom. We've got to eJg>and all of those. It doesn't have to be the doctor. are a lot of other possibilities than the doctor. It doesn't have to be face to face. There I are a lot of possibilities otrer than face to face .. (Family Practice) 12. What else would you like me to know about your thoughts regarding the value of the telephone-based decision support serwice to the community o(Rochester? The responses to this question +minimal and did not include any substantial content. Physicians responded to this question in two principal ways. Most indicated they had no additional coLents as they felt their opinions about this had already been captured in other parts Jfthe interview, especially question 10 regarding the "most significant benefits fjllowing the implementation of telephone based decision support. For those few lo did provide minor additional comments, their inputs were integrated into the analylis and summary of question 10. 245

PAGE 259

A major purpose of this analysis as to present and organize key informant responses in a way that provided more depth insight to each question as well as a more clear overall pattern of emerging t emes and issues. This summary provides a synopsis of common themes that emergeCi across all questions in the analysis. Where appropriate, each thematic section comples and contrasts physicians' general opinions and perceptions about that issuJ with their opinions about that issue in relation to the implementation of a decision support (TBDS) service for patients. Theme 1: Support Idea of More lnformeCi Active Patients Physicians fundamentally support d and actively endorsed patients' and consumers' desire for more information out health and medical issues. They acknowledged that there is a growing co sumer trend characterized by patients who have a preference for more information rather than less. Physicians principally believed that a more informed patient woJld result in better health and medical care decisions leading to improved quality of cLe, health outcomes, patient satisfaction, clinical e:fficiencesand doctor-.patient relatJonships. 246

PAGE 260

Similarly, physicians held a fundamentally positive opinion about patients and consumers wanting more active particip tion in the medical care decision-making medical decisions would be have a great r sense of control and self-confidence culminating in greater compliance with r upon diagnostics and therapeutics. This would result in improved health outcomei and higher overall patient satisfaction with care. \ TBDS: Physicians indicated that the most significant benefit ofTBDS services to the community would be increased coJsumer/patient access to good quality, uniform information about health and mjical concerns. They suggested thai increased access to this type of informati1 would support patients in making better decisions about use of the health care Theme 2: Improved Health Outcomes: Chronicl!vlajor Dlness Physicians believed that, in gen,, a more informed consumer/patient would be more likely to make better decisions usually resulting in superior quality, more efficient care and improved health outcomls. They believed this to be particularly true for patients with chronic illnesses. Also, jost physicians contended that outcomes would improve when patients are involved in decisions about major medical issues.

PAGE 261

They explained that the mechanis responsible for improved outcomes was related primarily to elements of self-efficlcy: improved patient self-confidence and control over body, life, and disease leaJg to improved compliance and overall satisfaction. Physicians clarified that evel if outcomes are less than what was expected, a patient who has a higher undlrstanding of their illness and who more actively participates in treatment decisionl and self-management will still have a higher satisfaction level, and, will be less likely tllitigate despite poor outcomes. Physicians reiterated that patient Jreference for active involvement in medical decisions was a key factor influencing oLes. They contended that the probability of improved health outcomes would be .Jgmented if physicians were sensitive to patient preference for active involvement Ld matched their response to that I preference. If physicians failed to match their response to patient preference, then health outcomes and overall patient sati+on may be diminished. Several physicians noted that patient involvement in decisions about treatment fur minor health issues would have neith, a positive nor an adverse effect on outcomes. Many physicians reasoned that most minor health issues were self-limiting, ld th. al dh I. d .. wou run err natur course, an t at patient partic1patton m treatment eas1ons would have no significant effect one way dr the other on clinical or functional status outcomes.

PAGE 262

TBDS: Physicians purported tha one of the benefits of patient access to more I uniform health and medical information t ough the TBDS service would be a more informed patient. In fact, they believed t at this was one of the most significant strengths of the TBDS service. They pos ated that this would favor better patient d .. db illn lui ... d. c.. d ecisions an etter ess management res tmg m Improve patient satis1act10n an health outcomes. Theme 3: Improved Patient Satisfaction (Patient Preference) Physicians acknowledged that a ore active informed patient in decisions about minor, major and chronic illness tould be a more satisfied patient overall. Consonant with health outcomes, they ,lained that the mechanism responsible for higher patient satisfaction was related primarily to elements of self-efficacy: improved patient self-confidence and control over bldy, life, and disease leading to improved compliance and overaD satisfiu:tion with ttment. Physicians reiterated that patient prence for involvement in varying types of medical decisions also influenced patient satisfaction. That is, if clinicians failed to be sensitive to individual patient differences l decision-making preferences for minor, d hr il1n edicald I h di d maJor, an c orne ess m ecrsions, p ysician-patlent scor ance may occur, resulting in lower patient satisfaction. Thi\y also believed that satisfaction was affected by the patient's perception about the phys ian being "indecisive", "unconcerned" or r

PAGE 263

"uncaring" if the patient was encouraged to take a more active and prominent role in the decision-making process as describ in Theme 4. TBDS: Physicians indicated tJ the most significant benefit ofTBDS to the community would be increased consumJ/patient access to good quality, uniform information about health and medical coLems. Many believed that increased access to this type of information by patients J indeed translate to higher patient satisfaction with health plans and medical services in general. Theme 4: Improved Doctor Patient Relationship (Patient Preference) Physicians acknowledged that a more active involvement of patients in the medical decision-making process is a trend in today' s care environment and one that physicians need to be prepared to deal with in the dynamic of the doctor-patient relationship. Physicians contended that, generally, a more informed active patient in decisions about minor, major, and chronic illnesses would enrich the doctor -patient relationship and Ice patient satisfaction, if clinicians were sensitive to individual patient differences in preferences for involvement in decision making as previously described. Although physicians were mainly about patient involvement in decisions, some expressed concern that mLe informed actively involved patients have the potential for negative impact on the ddctor-patient relationship. Some physicians 250

PAGE 264

expressed an apprehension about proactively involving patients in decisions and giving them a more prominent role. They percer that from the patient's viewpoint, the physician may appear to be "indecisive," !'uncaring" or "unconcerned," especially if s/he referred the patient to a TBDS servibe. One physician compared it to coming across like "an MCI commercial." TBDS: Physicians held conflictin views about the potential benefits ofTBDS to the doctor-patient relationship. Some physicians were of the opinion that patient access to infonnation outside their persor physician would potentially disrupt the doctor-patient relationship. They were clncemed that the information provided by the TBDS service could potentially be in conflict with their viewpoint about a given medical issue. Many physicians contendt that if the infonnation dispensed by the TBDS service was in agreement with their opinion, they would be inclined to support I the service. If the information dispensed by the service was not in alignment with their view about a given issue, they would not for the service. Second, as described above, phyj didn't want to be perceived by patients as "indecisive", "uncaring" or "unconcerned" if slhe referred the patient to a TBDS Thi . .lh h h h sernce. s vtewpomt was m concert Wit t e contention t at one way to strengt en I physician support and utilization of the TBDS service was to provide physicians with assistance in how to "position" the servi, to avoid alienating the patient within the context ofthe doctor-patient relationship (Theme 11). 251

PAGE 265

Other physicians were of the eq ally strong opinion that patient access to telephone-based decision support wouldlstimulate higher quality physician-patient communications. They were comfortabll knowing that patients are accessing infonnation from a variety of sources (nlwspapers, self-help materials, friends, Internet) and that the infonnation being 1spensed by the 1BDS service was at least uniform, standardized, high quality inforon. These physicians optimistically viewed patient inquiries about infonnatii they had obtained elsewhere as opportunities to open constructive dialo and to enhance communications within the doctor-patient relationship. Theme 5: Improved Clinical Efficiencies for Minor and Chronic mnesses l Physicians contended that more a, tive patient involvement in decisions about minor and chronic disease conditions woLd definitely have a positive impact on clinical efficiencies and cost management sacrificing quality of care. Out of the three categories for patient involvl. (minor ilhtess, major illness, and chronic illness), physicians emphasized that involvement in the management of chronic disease held the most value and would hate the most significant overall impact on health care system quality, cost, and efficilncy factors. Most were of the opinion that the number of medically inappropriate vi) s to their office and the volume of unnecessary phone calls would be reduc 252

PAGE 266

However, conflicting opinions w re expressed regarding whether or not involving patients in decisions about maj!l r medical issues would be more efficient. The gamut of opinions ranged from thos who believed it would definitely be inefficient and would take more time to who believed it would definitely be more efficient. Others were of the opinion that involving patients in these types of decisions may take more time initially, but in the long run it would be more efficient and would result in higher patient satisfaction. TBDS: Physicians were uncert and differed in their opinions about the impact and benefit they thought the implementation of a telephone-based decision I support service would have on clinical ejciencies. Some stated that clinical efficiencies would be improved through a eduction in unnecessary calls and office visits. Other physicians contended that thte was no such thing as an unnecessary call or visit. Furthermore, these same physicii viewed calls and visits for simple, selflimiting, self-care, minor issues as to enhance the doctor-patient relationship. One of the major driving forces fo The Rochester Community Initiative for Demand Management and the eventual TBDS service was an identified need by formal physician leaders to reduce the mnnber of Ledically inappropriate and wmecessary "nuisance" calls and visits to their practicJ because of increasing pressures to manage costs and optimize productivity. 2153

PAGE 267

As far as a demand mana ement issue that, certainly we've been talking about for yers. We need to change patient demand or it's going to control costs. A well-informed patient who demands appropriately, that's going to tum out to be a good trurlg. (Internist) However, it was evident that th+ was conflicting opinion and lack of consensus about the specific need and there was lack of clarification regarding exactly what a majority of practicing physicians from a TBDS service in regard to improved clinical efficiencies. On the onl hand, many physicians believed that a TBDS service would help manage patient demJd and appropriate use of the health care system. This would allow physicians to janage costs and improve productivity. On the other hand, there was a group of phyll icians who contended that the TBDS service was unnecessary because physicians shol!lld be their own "TBDS supplier" for their own patients (ThemelO). I do think it does need to be clear what the physician organizations want from this program. I see my perceptions may be than what the physician organizations actually askJd for in this. Because I do lmow that many in the community have been clamoring for a long time how to get a reduced number of phone calls, redhcing the amount of telephone based management. Once \again there may be mixed messages about that. It's Kind of, 'we don't want the volume of phone calls, but\we also don't want faceless strangers dispensing information to our patients. And yet a service like this is just that. The physicians have a conflict here of how to effibiently manage patient care and not be tied down to the telbphone day and night, and yet giving patients access to all the information. (Internist) 254

PAGE 268

Several physicians pointed out that what exactly a majority of practicing physicians wanted from a TBDS service was key to physician endorsement and use of the service (Theme 11 ). Theme 6: Reduced Health CareCosts for Minor and Chronic Conditions Congruent with improved clinical fficiencies, most physicians were of the opinion that involving patients in miTKIT ih issues and chronic conditions would help reduce health care costs without compromising quality of care. They believed I this to be particularly true for chronic coniitions. However, physicians expressed uncertainty and mixed opinions about the effect of involving patients in decisions about major medical issues on health care bosts. At one end of the continuum were those who were certain that patient invowlnent would reduce health care costs because patients would employ more appJpriate use of medical care services. On the other end of the continuum were those whl held an equally determined opinion that patient involvement would increase health 1are costs. Some cited that more informed active patients may actually request and inJ st on more services rather than less. TBDS: A small number ofphysici s expressed hope that the TBDS would have a positive impact on health care costs or the community. A predominant number of physicians were either uncertain or belie ed there would be no impact on health care costs for their particular practice. 255

PAGE 269

Theme 7: Reduced Potential for Medic Liability Physicians affirmed that medical bability would definitely be reduced by appropriately involving patients in decisi I ns about major medical issues and in chronic disease management. However, many expounded that "involvement" meant being able to establish a genuine relation hip with a patient that would allow the physician to understand, communicate, and relate to the patient and their illness in a more holistic approach. To emote and e I pathize with them as well as educate and inform about treatment options, benefits, d risks. Physicians contended that with this type of physician-patient communica on, medical liability is reduced, even when the outcomes are not as positive as expeJed. Patients who have a better understanding of their disease and particitte in diagnostic and therapeutic decisions are less likely to sue when things do go Jong. TBDS: A predominant number o physicians indicated that either they were "not sure" or that there would be "no irnplct" of the TBDS service on reducing the "al.C'. di alliabili" h .L.l potentl 10r me c ty m t etr practice. 256

PAGE 270

Theme 8: Self-Efficacy Improved with ore Informed Active Patient Physicians frequently attributed ,proved doctor-patient relationships health outcomes, patient satisfaction, clinical em!ciencies and cost management primarily to I the mechanism of enhanced patient self-efficacy. They explained that when a patient refers to take a more informed active role lin decision making and in the self management of disease, they feel more in Lntrol of their lives and the disease, more confident to deal with the disease a po) tive fashion. Summary: Themes 1-8 Physicians fundamentally support the concept of more informed, actively involved patients in health and medical carl decisions. Physicians were of the opinion that a more informed, actively involved palients would result in better health and d.cal d 1 d. I ali d me 1 care ec1s1ons ea mg to unprovements m numerous qu ty, cost, an utilization variables. Physicians' opinions Lding the potential impact of more informed actively involved patients in medibal care decisions are summarized in Table 4. 4. Physicians frequently attributed the uJderlying mechanism of improvement for these potential impacts to enhanced patienJ self-efficacy. 2fJ7

PAGE 271

Table 4.4. Summary of physicians' op ons about the potential impact of more informed active! involved atients in various es of medical decisions. Variable Type of Medical Issue Minor Major Chronic Health outcomes ..1 t t Patient satisfaction t t t Quality of doctor-patient relationship t t t Volume of unnecessary calls J.. ? J.. Number of inappropriate visits J.. ? J.. Length of clinical encounter J.. ? J.. Health care costs J.. ? .J, Medicalliability ./ .J, .J, t = increase, .J, = decrease, ./ = neutrai!no impact, ? = uncertain Theme 9: Variance in Physician and Patil Preferences Physicians differed in terms of t1ir preference fur involving patients in decision making about minor as well as major health issues. For minor and chronic health issues, the tendency was primarily lowards the patient having a more prominent role in the decision-making process. In lntrast, preference for decision making fur major health decisions was primarily incled towards the physician having a more prominent and directive role. l They reasoned that two primary :fi ctors contributed to the need for physicians to have a more prominent and directive Jle in major medical decisions. First, many physicians believed that severity of illness is a variable influencing the degree of patient and physician involvement in major medichl decisions. They tended to equate severity of illness with the need for a higher level f clinical knowledge about the condition; 258 I

PAGE 272

therefore, the perceived need for a highe level of physician involvement in influencing and guiding decisions about treatment. Physicians said that as patients grow sicker, they generally tend to place more trust in reir physician and tend to want the physician to take a more prominent and directive role in decision making because of his/her clinical expertise. Second, as physicians varied in their preference for involving patients in decision making, physicians recurrently elphasized that patients also differed in their preference fur involvement in the decisiimaking process. That is, some patients may prefer a more passive role ("I prefer that my doctor tell me what to do") while some may prefer a more collaborative approach ("I prefer that my doctor and I make the decision together''). Several physicians ccmceded that if clinicians failed to be sensitive to individual patient differences il decision-making preferences for minor and llll!ior medical decisions, physician-patient rscordance may occur, resulting in lower patient satisfaction and a less than optimal (ioctor-patient relationship. In gl .. I b J:'. J:'. terestm y, m contrast to opm10ns a out patient pre1erence 10r actlve d d' a1 d I h mvo vement m mmor an maJor me IC ecistons, p ysiCians were not as mtense nor as incessant in their opinion about the need to recognize individual differences in patient preference for active involvement in the management of chronic illness. Only h .. d. ablhib' J:'. two p ystCians rna e mmor comments out t s emg a concern 10r patient involvement in chronic illness. For a majojty of the physicians, they recognized that 259

PAGE 273

differences in patient preference for acti e involvement in managing chronic disease existed. However, regardless of patient preference, it was almost as if extensive patient involvement in treatment decisioJs and self management was a given, an uncompromising and necessary componr\ t in the successful management of chronic disease. That to achieve better outcome patients had no choice but to become more actively involved in decisions about and for chronic conditions. Worth repeating here is how one summarized the essence of how important it is for both physician and paJent to be aware of this issue in the doctorpatient relationship. I think, and this is part ofl:hat I think is what the art of medicine is. And that is, is the physician appropriately reading the patient's desire for inv lvement, and matching his response to that desire. And if the hysician is able to do that, then that should enhance the relatio'Fhip. If the physician and patients aren't communicating that need appropriately, then there will be problems. Let's say, ifthe/1 atient really wants to be more actively in decision making, and th physician doesn't pick that up and adopts a more paternalistic\ approach, then the patient's not going to be satisfied. U: on the hand, the physician feels that it's important for the patient t9 be completely involved in the decision making and starts providing the patient with information and expecting patient's to make decisions, and they're not ready to do that! and they don't want to do that, then that's a in the other direction and that could cause problems J.lso. So, I think, the key here is for physician and patient to\sort ofbe able to read each other, and communicate to each other where they are in that desire to be involved in decision nWang, and then there would be a good fit and a good match. (Pediatrician) 260

PAGE 274

The take home message for clinicians is that for any given patient at any given time, there may be a variety of sociodemigraphic, intra-personal, and situational factors that determine a patient's preference for more information and/or more active involvement in different types of medicallare decisions. The most prudent clinical approach for the physician is to be sensitile to each individual patient's preference for I information and to the types of decisions for which patient involvement is most likely to be preferred and beneficial. Failure to 1o so on the part of the clinician may result in physician-patient discordance and a IesJ than optimal doctor-patient relationship with ensuing negative impacts on patient jarisfuction, health outcomes, and medical liability. Theme 10: TBDS -Uncertain or No Impact on Physicians' Practices and Patients Physicians differed in their opinions about the impact they thought the I implementation of a telephone-based decision support service would have on their patients and practice. A majority (n=18) physicians were either uncertain or I believed there would be no impact of the 1BDS service on their practice. Many noted they have not had enough experience with lhe service yet to really know what type of I impact it might have on their patients and their practice. Others candidly stated they weren't familiar enough with the service to know whether it would have a substantial impact on their practice. 261

PAGE 275

Another group of physicians said that the service would have "no impact" on their patients and their practice because t ey personally fulfilled the role of the service by providing their patients with the info arion and counseling they needed to be more active, involved consumers. They themselves, as one physician put it, as a "TBDS supplier'' for their own patients.l Many physicians were somewhat decisive and contradictory in their views about the impact of the TBDS service wt it came to speculating about the impact of the service on their practice. Many physicians stated that either they were unsure of the impact or that there would be no imp \ct ofTBDS services on their particular practice. Theme II: TBDS Lack of Support by Practicing Physicians Will Diminish Its Value Even though there was near unanimous support and endorsement for the concept of a more informed and active coAsumer of health care services, there were mixed opinions about the actual implemenlation of a community telephone-based decision support service for patients/consuLers. Three, eleven, and seven of the key informant physicians held generally neutral, and generally positive opinions about such a service, respectively. Even tllough only three physicians held an explicitly negative view about the service, seemed to be a prevailing attitude of 262

PAGE 276

"let's wait and see" skepticism or uncertainty even among the more positive physicians. It was the resolute opinion of these physicians that extensive physician endorsement and support fur the mos tee bad not been obtained. Because of this, they asserted that the two most signijficant and related challenges facing the implementation of the TBDS service werl: 1) physician endorsement and utilization in the everyday practice of medicine and 2) latient/consumer utilization, primarily due to lack of endorsement by the practicing phjsician community. The primary reasons cited fur a lack of widespread physician 1y-in and support included: 1. Lack of extensive participation in the design, development, and decision to implement the consumer decision support service. 2 lnsufli :_c. . d \ d" b th . etent uuormatton or mtsun erstan mg a out e sernce. 3. Inadequate experience with the servicJ 4. Unfamiliarity with the protocols used the telephone-based nurse counselors. 5. Potential conflicts between the inform1on provided by the telephone-based service and the physicians' opinion abo'ut appropriate utilization of medical services. 6. Suspicions regarding the primary aim or the service. 7. Uncertainty regarding its value in tenJ of impact on patient health and overall cost-benefit. 8. Uncertainty regarding how to "position" the service to avoid alienating the patient within the context of the doctor-patient relationship. 263

PAGE 277

Practical suggestions for ampl.ifyih.g physician endorsement and support included: 1. Working through the existing formal and informal network relationships within the existing physician organizations, e] ecially the IP A subcommittee structure and key physician opinion leaders. 2. Establishing a physician oversight rttee ofkey physician leaders. 3. Educating as many physicians in the f.eld as possible about the nature, purpose, benefits, and value of the service to fhem and their patients. 4. Arranging opportunities for key to "experience" the service first hand and providing examples of how certain calls from patients would be handled. 5. Affording key physicians and the opportunity to review, approve, and even participate in developing thJ 1 top I 0-20 protocols that would be most frequently used by the nurse counselor 6. Assisting physicians in how to the decisions support service to avoid alienating patients in the doctor-patie11t relationship. I 7. Providing frequent physician-specific (eedback about the performance of the decision support service in terms of mhnber of patients, type of call, and patient satisfaction with the service. 8. Rendering physicians and the opportunity to participate in approving and developing the counselihg protocols and the opportunity to influence the training and education o1the nurse counselors providing the service. 9. Involving and educating nursing and office staffs who play a key role in patient education. 10. Clarifying what exactly it is physicians want from such a service. 11. Completing an opinion survey of all field physicians regarding their perspectives about patient decision support serviceJ and suggestions for successful implementation prior to program deve opment. 2'4

PAGE 278

12. Assisting physicians and employers f forming alliances and partnerships that will optimize success for these kinds of community initiatives. 13. Developing a local consumer/patienl decision support service franchise that is physician owned, directed and The primary reasons cited for a pbtentiallack of patient/consumer utilization included: 1) lack of personal physician utilization, and referral to the service, 2) lack of full understanding by patients kbout what the service is, and 3) introducing a new mindset and a new behavior to pa1ents in regards to accessing health and medical information from someone other than their own physician. Suggestions for increasing patient utilization included: a) lnhancing physician support and integration of the service into the day-to-
PAGE 279

c \TER5 DISISSION This chapter presents a discussion of key quantitative and qualitative results including practical implications and limijons. Recommendations for future research and conclusions are provided. Key Quantitative an Qualitative Results Using a multi-method (quantitative Ld qualitative) approach, the specific aims of this study were to: I. Ascertain primary care and sp,alty physician opinions about more informed actively involved patients as partners in health and medical care decisions. 2. Assess primary care and specialJ physician opinions about the impact they believe consumers/patients being more info ed and taking a more active partnership role in health and medical care decisions will have on: (a) the quality of the doctorpatient relationship, (b) clinical efficiencies, outcomes, (d) patient satisfaction, (e) health care utilization, (f) h th care costs, and (g) malpractice liability. 3. Assess primary care and specialty physician opinions about the incorporation of TBDS services into care management and the impact they believe a 266

PAGE 280

TBDS service will have on: (a) the quali of the doctor-patient relationship, (b) clinical efficiencies, (c) health outcomt (d) patient satisfuction, (e) health care utilization, (f) health care costs, and (g) malpractice liability. Ph sicians' 0 inions about Shared Decisit Makin Physicians in this study fundamen y supported the idea of more informed and actively involved patients in the health medical decision-making process. Overall, approximately 90% of physicians rieved that patients generally desire more information about health and medical issue!; and 80-90% of physicians believed that patients generally desire more active invo/v, ment in the medical decision-making process. Regression results indicated that tLse opinions were consistently strong across all physician and practice characteris1cs. Unknown is to what extent these I beliefS translated practice behaviors i what these opinions were concordant With patient preferences for more information and/or active mvolvement m the medical decision making process. Exlation of physician-patient concordance variables is suggested for future studies. Not only did physicians believe that latients generally desire more information and more active involvement, this group of physicians endorsed the fact that more informed and actively involved patient decisiln making was good for patients and for medicine. Specifically, they believed that a ore informed patient would result in 267

PAGE 281

better health and medical care decisions and a greater sense of control and selfconfidence fur the patient ultimately 14 to improved quality of care, enhanced health outcomes (for chronic conditions Jd major medical issues), higher patient satisfaction, enriched doctor-patient relati1nships, potential reduction in medical liability; and especially for minor illnesses and chronic conditions, improved clinical efficiencies (reduction in the nwnber ofm+cally inappropriate visits and volume of unnecessary phone calls) and cost controllable 4.4). Physicians' opinions from both the quantitative and qualitative results of the present study substantiated that patient involvement in the self-management of chrorc disease was absolutely essential for optimizing health outcomes, enhancing patir satisfaction, enriching the doctor patient relationship, improving clinical efficiencies, cost management, and reducing the potential for medical liability. Physicians' opinions about the poten "al impact of more informed and actively involved patients are in agreement with Jous studies that have illustrated the positive impact of more informed and involved patients in decision making (Ong, DeHaes, Hoos, & Lammes, 1995; DiMatteo, 1994; Lerman, Brody, Caputo, Smith, Lazaro, & Wolfson, 1990; Lorig, Mjonson, & Holman, 1993; Brody, Miller, Lerman, Smith, & Caputo, 1989; Greenfield, Kaplan, Ware, Yano, & Frank, 1988; Wasserman, Inue, Barriatua, Carter, & Lippincott, 1984; Vickery, Kalmer, Lowry, Constantine, Wright & Loren,1983). 268

PAGE 282

When physicians were asked mor specifically their opinions about patient involvement in decisions regarding uea4ent, 56% and 68% of the physicians were of the opinion that the physician and the patibnt should collaborate and take a shared d . aki h :fi d \ edi al. I ecision m ng approac or nunor an maJor m c Issues, respective y. Additionally, 66% of physicians were of ,e opinion that their patients should take a greater role in the self management of chronic illnesses. However, both the quantitative and qualitative findings of the study corroborated that physician preference varied for involving patients in specific types of decisions and that physicians varied in their opinions about th patient's preference for decision making. For minor health issues and chronic conditions, there was a tendency for physicians to believe that patients should take a more predominant role. Although physicians were generally positive about palient involvement in decisions regarding minor health issues, an intriguing qualitativ, finding in the present study was physicians' perceptions about how patients rght view doctors who gave them a more prominent role in decision making and the perceived subsequent negative impact on h d I hi S h \. d h . t e octor-patlent re ations p. orne p ysicians perceive t at patients may VIew doctors who give them a more prominent roie in health and medical care decision making or who encourage them to use altemltive information sources such as TBDS as "indecisive", "uncaring", "unconcerned", t a "wet noodle". Unknown is to what extent these perceptions are actually held by \ atients. It is suggested that this variable 269

PAGE 283

be examined in future research regarding f.hysician-patient communications and patient preference for information and active involvement in decision-making processes. Secondly, these observations suggest that some physicians may benefit from enhanced physician-patient communication skills. srch training should assist physicians in "positioning" appropriate patient involveJent in decision making without alienating the patient within the context of the relationship. For major health issues, there was a tendency for physicians, especially older physicians, to believe that physicians shoul1 take a more predominant and somewhat paternalistic role in the decision making pcJcess. Two primary factors contributed to the perceived need fur physicians to believ, that they should have a more prominent and directive role in major medical decisioni. First, many physicians believed that severity of illness is a variable influencing ttJe degree of patient and physician involvement in major medical decisions. T ly tended to equate severity of illness with the need for a higher level of physician invo vement in influencing and guiding decisions about treatment. Physicians said at as patients grow sicker, they generally tend to place more trust in their physician t tend to want the physician to take a more prominent and directive role in decisi, making because of his/her clinical expertise. These opinions were concordant with previous research (Beaver, Luker, Owens, Leinster, Degner & Sloan, 1996; Nle and Brooks, 1995; Ende, Kazis, Ash, 270

PAGE 284

and Moskowitz, 1989) indicating there wr less patient preference for active involvement in decision making with incr \ sing severity of illness. Second, as physicians varied in their preference for involving patients in decision making, both qualitative and qujtitative analyses substantiated that making. In the qualitative portion of the p esent study, physicians frequently pointed out that there is a range of medical decisiormaking preferences by patients. Some patients prefer a more passive role ("I prefer that my doctor tell me what to do") while some prefer a more collaborative approach "I prefer that my doctor and I make the decision together''). In the quantitative po on of this study, physicians' opinions were varied regarding or not "patients r their io make decisions fur them." These observatiOns are consistent 1th numerous studies documenting variation in patient preferences for information and involvement in decision making (Cassileth, Zupkis, Sutton-Smith, and 1980; Ende, Kaxis, Ash, and Moskowitz, Sutherland, LlewellynThbmas, Lockwood, Tritchler, and Till, 1989; Nease and Brooks, 1995, Bradley, Zi and Hamilton, 1996; Hack, Degner, and Dyck, 1994; Davison, Degner, and Morgan, 995,) as well as variation in physician perception about patient desire for information and involvement (Stroll, Lo, and Charles, 1984). 271

PAGE 285

Physicians in this study noted that the most sensible approach for the clinician is to be sensitive to each individual patien 's preference for information and to the types of decisions for which patient invol }ement is most likely to be preferred and beneficial. Several physicians acknowledrl !bat if clinicians fililed to be sensitive to individual patient differences in decision making preferences, physician-patient I discordance may occur resulting in a less than optimal doctor-patient relationship and low patient satisfaction with ensuing nega,ve impacts on heahh outcomes and medical liability. These observations are consistenl with previous studies (Beaver, Luker, Owens, Leinster, Degner & Sloan, 1996; Davison, Degner & Morgan, 1995; Hack, Degner, and Dyck, 1994;Ende, Kaxis, Ash, and Moskowitz, 1989; Stroll, Lo, and Charles, 1984) suggesting that physicians jve explicit discussions with individual patients regarding information and decisionlmaking preferences. With the continued increase in the utilization of alternative medlcine (Eisenberg, Davis, Ettner, Appel, Wilkey, Rompay & Kessler, 1998), these are also consistent with recommendations about the importance of shared decision making and physicians I having explicit formal discussions ofpatientls preferences and expectations about the use of alternative therapies (Eisenberg, 1991. Not surprisingly, physicians viewed themselves as the best source of health and medical information for their patients. Prese1ng the physician-patient relationship in an era of managed care has become a central concern for most physicians (Emanuel 27

PAGE 286

and Dubler, 1995). Traditionally, the do or-patient relationship has been predominantly one in which the patient has assumed a very passive, dependent and coopera!ive/acquiescent role while the phtcian has assumed a very active, directive, autonomous, and professionally dominant1ontrolling role. However, physicians in this study acknowledged that medicine hal entered into a new era one that is more consumer centered and characterized by a!ancing sophistication and rapid growth of communications and information managem nt technologies. It was recognized that consumers will accelerate the access of heJth and medical information from various sources such as the Internet and that this in tum may require physicians to make changes in the way they mange the doctoratient relationship and how they approach physician-patient communications. Specific y, they were of the sentiment that it would become increasingly important for ph! sicians to hone their physician-patient communications skills. Many conceded that with some patients they will need to become less paternalistic in their approach. rey will need to become more sensitive to patient preference for involvement in heal h and medical care decisions and adapt their style to that prefurence. The observati+ from the present study -that physicians believe patients desire more information, thalphysicians support and advocate a more collaborative doctor-patient relationship, tha physicians view themselves as the best source of health and medical information to patients for shared decision-making, and h h . d ak hift" \. . I h t at p ystctans may nee to m e a s m p :tient commurucattons sty e prompts t e 273

PAGE 287

question, "Are contemporary physicians dequately equipped to communicate and collaborate effectively with the new bree of consumer (Herzlinger 1997) to preserve the physician-patient relationship in an era of consumerism and managed care, to enhance health outcomes, to optimize + satisfuction, and to ultimately sustain competitive advantage in the marketplacet Observations from the present study and current research in physician-patient communications indicate they may not be rg, DeHaes, Hoos, & Lammes, 1995; DiMatteo, 1994; Beckman & It is recommended that physicians objectively assess their strengths and Ts in patient communications and "position" themselves for the demands of tie emerging consumer-driven marketplace by improving physician-patient communica1on skills. Physicians should benefit from patient communications training that emph1izes active listening, dealing with patient exploring expectations/idt effects of problems (Frederikson, 1995), usmg open-ended quef1ons, alloWing the patient to talk uninterrupted at the outset, repeating and sJmma.mmg patient's problems, avoiding I the use of medical jargon (Levinson, Stiles, Inui, & Engle, 1993), orienting the patient to the flow of the visit, addressing emotions td psychological issues, exploring patient beliefs, checking for understanding, asking if the patient has additional questions, providing a summary of the visit, Lmpleting a final check for under-274

PAGE 288

standing, clarifying the treatment plan, pr viding encouragement/reassurance, and consolidating rapport/caring (White, Levinson, & Roter, 1994). Contrary to expectations, a majorit of physicians in this study indicated that they had enough time to answer patient's juestions about medical care and to fully involve patients in the decision-making pr cess. In a largely productivity-driven, predominantly fee-for-service marketplace, we expected these physicians to provide the regression analyses highlighted a statist cally significant linear relationship between the percent of capitated patients and physicians' opinions about having adequate time I to answer patient's questions. With increasing capitation, there was an increasingly higher percentage of physicians agreeing wi1h the statement that "physicians don't have adequate time to answer patient questilns about medical care. Also, even though there was not a statistically significant relatilnship, with increasing capitation there .. "glhigh fh\.. "hh h was m mcreasm y er percentage o p ylictans agreemg Wit t e statement t at "physicians do not have adequate time to fully involve patients in the decision-making process." There may be several factors that Ler into possible explanations for this finding. First, physicians practicing under managed care contracts with capitated panels are definitely under pressure to see patients i a timely manner. Certain perfonnance standards and guidelines for patient access d for seeing patients in a timely manner 275

PAGE 289

have been instituted by most managed care plans to "protect" the consumer/patient from potential access issues inherent in a system where the physician does not have to see the patient for reimbursement. Health plans and IP As monitor physician performance on these access standards d report results to patients and consumer groups. Therefore, physicians with capita ed panels may have the tendency to "squeeze in" more patients, especially in a mixed fee-for-service and capitated reimbursement environment. Second, capitated patients under care plans can tend to be more transient and more demanding than the more well-established long term patients a practice may have. When patients under aged care contracts are capitated to a given provider, they have no choice but to access care in that office if they want to be covered under the health plan; whereas, a ee-for-service patient can just go to another practice. Therefore, when capitated patienL are assigned to a given provider with whom they are not familiar and have not established a quality doctor-patient relationship, perceptions of restricted access and scheduling delays augments suspicions that they are not getting fair vall. under the plan. Thus, even though one would surmise that the fee-for-service pati+ would have better service at a physicians' office, they may be "shut out" by an already full schedule of a more highly regulated capitated group of patients the ph sician is also responsible for. 270

PAGE 290

Third, some physicians still hold a d set that capitated patients are poor reimbursement patients ($5/$10 co-pay fo the visit). Subsequently, this may create the sense that one cannot spend as much tLe with them considering the reimbursement immediately realized. In cJntrast, a fee-for-service patient requiring more time can be "billed" accordingly for J higher acuity level visit. Thus, a physician may feel more justified in billing a patient lore because they spent more time with them, but could not do so with a capitated batient. Further research is warranted to examine to what extent reimbursement + serve as incentives or disincentives to physicians talking with patients and the iffect on physician-patient communications, the doctor-patient relationship, patient healih outcomes, and physician productivity. Physicians' Opinions About Telephone-Bl Decision Suooort Physicians were asked specific about their opinions of a 24-hour telephone-based decision support (TBDS) sLvice that had just been implemented hi h "d .\ h c: b :_c: ed Wit n t e community to provt e pattents Wit a resource .tor econung more nuorm actively involved consumers in medical care becisions. The information regarding physicians' opinions about TBDS services is among the first to be reported in the research literature. Results from the present study are co sistent with findings from adoption rate studies in the d.i:ffusion of new innovations in health care (new drugs, findings from 277

PAGE 291

large clinical trials, national consensus treatment guidelines). In previous studies, rates of physician awareness with a "new inno1on" ranged from 44 94% approximately one year following their release (Peay & 1988; Lomas, Anderson, Domnick-\ Pierre, Vayda, Enkin, & Hannah, 1989; Hill, Levine & Whelton, 1988; Grilli, Apolone, Marsoni, Nicolucci, Zola & Liberati, 1991r.ln the present study, physicians were asked to report whether or not they had heard of the TBDS service (awareness) and to what extent they were familiar with its and available services. In the quantitative portion of this study, tr fuurths (73.7"/o) indicated that they were aware of the service. Of those reporting that they were aware of the service, 31.3% indicated that they were familiar wi hits purpose and available services while 42.4% reported that even though they had eard about the service they knew very little about its purpose and available services. Of the 21 physician key informants who participated in the qualitative portion ofthl\ study, 18 (86%) reported being familiar with its purpose and services. Given the fo sed intensity of the formal communications fur the physician communil inunediately prior to implementation of the TBDS service, it was anticipated that pJysician awareness levels would be high. Explanations for the finding that about one third who had reported being aware of the service were familiar with its plrpose may be partially explained by the fact that even though a very comprehensive Ld intensive formal physician communication plan was deployed, largely t \ ough mass media channels, more 278

PAGE 292

informal physician-to-physician communirtion and discussion necessary for higher levels of understanding and consensus abo t the service may not have occurred. That is, even though some studies in the diffusiJn of health care innovations have shown that the most frequently cited channel of lareness were staff conferences, Hill Levine & Whelton; Lomas, Anderson, omnick-Pierre, Vayda, Enkin & Hannah, 1989; ), others have suggested that information about a new innovation and receptiveness to change may be spread prilarily by word of mouth through informal Menzel, & Katz, 1966; Kosecoff, Kanouse, Rogers, McCloskey, Winslow & Brook, 1987; Lomas, Anderson, Domnick-Pierre, ayda, Enkin & Hannah, 1989; Greer, 1994). A higher rate of physician familiarity with and understanding about the TBDS purpose and specific services may have bee achieved through strategies that focused on the more informal interpersonal diffusion networks and consensus-seeking processes as described by Greer, (1987, 199 ) and Rogers (1995). Even though there was strong support and endorsement for the concept of a more informed and actively involved patient in the medical care decision making. process, physicians' opinions about the irnpl,mentation of a TBDS service to provide decision support services for patients were vred. Depending upon awareness and familiarity with the service, most held a neutral opinion ( 42-48%) with the remainder 279

PAGE 293

divided between generally positive (21-31%) and generally negative (15-26%). Comparable results were obtained in the qualitative portion of the study with neutral, generally positive, and generally negative expressed by 52'Yo, 33o/o, and 14% of the key informants, respectively. In cdnducting the key informant interviews, a observation of this investigator that even though only three physicians held an explicitly negative view about the rDS service, there seemed to be a prevailing attitude of uncertainty or "let's rand see" skepticism among even the physicians with positive and neutral opinions. It is believed by this investigator that the high percentage of physicians a "neutral" opinion about the TBDS in the quantitative portion of this study reflects tJe prevailing attitude of uncertainty revealed h d th k !-C. \ m t e more m-ep ey uuormant mtervters. The convergence of quantitative anl qualitative opinions of neutrality about the TBDS services may be somewhat explamed by the principle problem of uncertainty. That is, the physicians in this Ldy were exposed to an innovation (TBDS services) that they were uncertain iout. Rogers (1995) has emphasized that uncertainty is generated by innovations (defined as an idea, practice or object that is perceived as new by an individual or an orgLation) and that this uncertainty motivates individuals to seek information ablut the advantages and disadvantages of the innovation in order to reduce the uncert1nty. Furthermore, he adds that: 280

PAGE 294

Information about an innovation i 1 often sought from near-peers, especially information about theirbjective evaluations of the innovation. This information exc ge about a new idea occurs through a convergence process in lving interpersonal networks. The diffusion of innovations is essJntially a social process in which subjectively perceived information rbout a new idea is communicated. The meaning of innovation is thus gradually worked out through a process of social construction (Rogers, 1995; p. xvii). Likewise, Greer, (1994; p. 434) hal emphasized that "In discussion with colleagues, novelty is tamed, made tractab l. and joined to a familiar framework . .. The problem of uncertainty is resolved by new consensus .... In the diffusion of new innovations, members of a social syste create and share information with one another and pass through an innovation-d1sion process characterized by five stages. In the innovation-decision process, individuals pass from first awareness of an innovation (knowledge) to forming an o,n or attitude toward the innovation (persuasion), to a decision to adopt or reject (decision), to utilization of the new idea (implementation), and finally to evaluation jfthe decision (confirmation) (Rogers, 1995a). It may have been possible that the p ysicians in this study were somewhere in the continuum between the first two stages of the innovation-decision making process where there was still much uncertainty abou1 the mos and opinions (positive/negative) were still being formed. That is, the time period between the introduction of the TBDS service and the of data for this study may not have been sufficient time for significant info \ arion exchange among the interpersonal 281

PAGE 295

diffusion networks of physicians and a su sequent convergence and social construction of either a definitive positive or negative linion about the 1BDS services. Unfortunately, the design of this study did not allow us to examine this possibility. Researchers and practitioners interested in the diffusion of TBDS services need to examine more closely the innovation-decis1on process and its effect on physicians' opinions over time. Regression results indicated there were no significant differences of opinion about the 1BDS service related to physiaalpractice cbaracteristics. Interestingly, in the classic diffusion investigation studying le nature of diffusion networks for the spread of a new drug among physicians, Katz & Menzel (1966) found that various indicators of network communications and interconnectedness with a1 k I d. f d mterperson networ s were more important pre 1ctors o pos1t1ve opmmn an subsequent adoption of the new drug tban Le individual physician/practice characteristics. Specifically, affiliation with I hospital as a regular staff member, more frequent attendance at hospital staff meetinJs, sharing an office with one or more doctors, being named sociometrically as a Jurce of information and advice by other doctors as someone with whom they discusL their patient's cases, and as a best the new drug. The present study was not d signed to explore these issues. It is recommended that future research examine uch network interconnectedness factors 282

PAGE 296

as independent variables in predicting phy icians' opinions about TBDS and/or the adoption ofTBDS services. l Uncertainty may have also played\ partial role in explaining physicians' opinions about the potential impact ofTBDS on selected medical care delivery variables. In the key infonnant interviews, lpproximately 43% of the physicians indicated that they were ''not sure" or werl "uncertain" about the impacts ofTBDS. Many physicians emphasized that they had not had enough experience with TBDS services to really know what type of impa it might have on their patients and their practice. Supporting this qualitative resulj as the finding that a consistently predominant percentage (51.5-77.4%) ofp ysicians in the quantitative portion of the study were of the general opinion that TB t services would not have an impact on selected medical care delivery variables. Two possible explanations might be considered for these findings. First, the consistently predominant p\ercentage of physicians believing that the TBDS services would not have an impact selected medical care delivery variables may be a reflection of the uncertainty about lhe impact of TBDS services revealed in the more in-depth key infonnant interviews.lThis uncertainty may have once again resulted from a lack of clear understanding a out the purpose and the services of the TBDS; and, insufficient time for significant exchange among the interpersonal diffUsion networks of physici \ for convergence and social construction 283

PAGE 297

of a definitive common opinion (positive or negative impact) about the impact of TBDS. Supporting this possibility was e quantitative finding that a majority of the physicians did not believe that TBDS se ces would have an impact on patient demand for services, the number of inapp opriate visits, or the volume of calls. Ironically, one of the major driving fur the implementation of the TBDS service in the community was an identified need 9Y physician leaders to reduce the number of medically inappropriate visits and unneceJsary phone calls because of increasing pressures to manage costs and optimize p \oductivity ("For a Healthy Future Together," RCIP A Provider Newsletter, 1995). It appears that there was uncertainty within the physician community about exa ly what it was physicians wanted from a TBDS service. The second possible explanation is fhat physicians truly believed that the TBDS services would not have a significant impadt (positive or negative) on the selected medical care delivery variables. Evidence ipporting this possibility was gleaned from the key informant interviews. These interviews revealed that some physicians truly believed that the TBDS service would havj no impact on their patients and their practice because they believed that they plnally fulfilled the role of the 1BDS service by providing their patients with the and counseling they needed to be more active, involved consumers. They L1 ewed themselves, as one physician put it, as a "TBDS supplier" for their own patients. In either case, each of these possible

PAGE 298

explanations holds significant relevance f4 r the practitioner interested in implementing TBDS services. These conflicting views lack of consensus, once again, seem to point to uncertainty about the TBDS puJose and available services. If uncertainties are prevalent regarding physicians' opiniols about the purpose and potential impact of TBDS, then it is highly probable that the lervice will not be well-received and that physician resistance will deter and ultimately optimal utilization. Although studies to investigate the true Jpact of TBDS on selected medical care quality, cost, and utilization variables may be difficult to conduct, this may be an area fruitful for further study. It is worthwhile to note that even tiough an overall majority of physicians believed there would be no impact on pati nt demand for service's, the number of visits/week and the percent of capitated +ents were significantly related to physicians' opinions about the potential impact ofTBDS on patient demand for services. Physicians with a greater numberlof visits per week and physicians with greater than 50% capitation believed that t e TBDS service would have an impact on patient demand for service. Consistent w4 this observation was the finding that a greater percentage of physicians with the hikest percentage of capitated patients (> 50%) when compared to physicians with a lower percentage of capitated patients believed that the health plans implemented the TBDS to more efficiently care for capitated patients. It may be that the experiLces of physicians with high volume

PAGE 299

practices and a higher percent of capitatioi have led them to believe that TBDS services may indeed assist them in being eren more productive while maintaining high quality patient care. In an era of managed care where physicians are being asked to assume more financial risk in the care of al defined population of patients, a critical issue for future studies is assessing the trul impact of TBDS services on utilization, physician productivity, and costs. I It is also worthwhile to note that J. only exception to a majority of physicians believing that there would be no impact frbm the TBDS services was "patient satisfilction with medical care." In fact, oL one third of the physicians believed there would be in increase in patient satisfaction medical care because of the TBDS service. Supporting this result was the finding that approximately 60% of the physicians believed that the health plans TBDS to improve member satisfaction. These quantitative results werJ consistent with findings from the key informant interviews whereby physicians inkcated that the most significant benefit of the TBDS service for the community woulJ be increased consumer/patient access to good quality, uniform information about htth and medical concerns. As a result, many believed that increased access to this type of information by consumers/patients may indeed translated to higher patient with health plans and medical services in general. These opinions are conJistent with anecdotal reports that TBDS services enhance patient satisfaction (Barn \ 1995). This observation also provides a 286

PAGE 300

key implication for practitioners interestea in successfully implementing TBDS services. That is, TBDS services may be Lore acceptable to and endorsed by physicians if the primary purpose of the Jrvice is to enhance patient satisfaction. The primary reason physicians belited that the health plans implemented TBDS services were for reasons of cost c ntrol in response to the interests of the business community, as opposed to reaso1s related to clinical quality as measured by patient health outcomes, quality of care, malpractice (Table 3.6). This was not surprising in view of the fact that during j time of this study, a significant amount of attention was being given in this communi to employer concerns about rising health care costs. These observations are consiste t with Patrick and Wickizer's ( 1995) contention that the socioeconomic context of the local healthcare marketplace can significantly influence opinions and subsequent behaviors related to the introduction of new health care strategies. l Additionally, the large range and v ety of opinions regarding why the health plans implemented TBDS seems consistenth the theorem of uncertainty described earlier. Several physicians in the key info ant interviews pointed out that clarifying exactly what a majority of practicing physiclans wanted from a TBDS service was key to physician support and endorsement ofthl service. The qualitative portion of the study revealed that sufficient physician buy-Land support of the TBDS service had not been obtained. Most of these reasons wlre related to a lack of physician

PAGE 301

participation, information, experience, faririliarity, and overall uncertainty regarding the primary purpose, value, and impact of the TBDS service. These reasons, along with the practical suggestions given for amp g physician endorsement and support, have significant implications for practition rs faced with the daunting challenge of implementing TBDS and optimizing value. It is suggested that practitioners familiarize themselves with for diffilsing innovation in the medical conununity and assisting physici,s in dealing with uncertainties that are inherently involved in making decisions ab ut new alternatives. It is the opinion of this investigator that particular attention shoul be given to: (a) attributes of innovations that effect their rate of adoption, (b) varia es influencing the innovation-decision process, and (c) techniques for identifying d working with physician interpersonal communication networks. A significant finding of this study as that of all the predictpr variables, logistic regression determined "specialty" to be +cantly related to he greatest number of physician opinion variables. Specifically, pFary physicians (general practice, family practice, general internal medicine, ahd general pediatrics) were more likely than specialists to: 1) agree that their patiets should take a more active role in treating their own minor conditions; 2) agr1 that their patients should take a greater role in self-managing chronic conditions; 3) be familiar with and knowledgeable about the TBDS services; 4) have a more positiv opinion about TBDS services; 5) believe 288

PAGE 302

that the health plans implemented TBDS o respond to interests of the business community, to more efficiently care for capitated patients, improve quality of care, and to improve member satisfaction; 6) believl that TBDS services would decrease the number of medically inappropriate visits I\ d the volume of unnecessary calls. These results are somewhat di:ffic It to explain but may be rooted in the relationship between the innovation-deciJion making process and the very nature of primary care. In the innovation-decision Rogers (1995) has pointed out that social, cultural, and physical proximity and homophily (the degree to which two or education, and social status) are more lik ly to result in: (a) more frequent transfer of ideas and (b) more effective communicatiol. "When they share common meanings, a mutual sub-cultural language, and are alikd in personal and social characteristics, the gain, attitude formation and change, and o ert behavior change" (p. 19). It may have been plausible that this group of primary c e physicians in this particular community comprised a specific homophilous interperJonal social network of physicians whose frequency and means of interpersonal cojunications augmented the consolidation of consensus opinions and enhanced the probJbility of significant relationships to a number of physician opinion variables. FutLe studies should be designed to investigate this possibility.

PAGE 303

Second, and closely related to the concept ofhomophily, is the probability that by the very nature of their training and th ir personal decision to become a primary care physician, primary care physicians wire more interested in and possessed a more fuvorable predisposition towards the co+ of more infonned actively involved patients and 1BDS. Indeed, Donaldson, Yordy, Lohr, and Vanselow (1996; p. 80-81) have emphasized six distinguishing attribules of primary care that would tend to support this contention: 1. Excellent care is grounded in both the biomedical and fhe social sciences. 2. Clinical decisiory making in primary care frequently differs from in referral specialties. 3. Primary care haf at its core a sustained personal relationship benreen patient and clinician. 4. Primary care dofs not consider mental health separately from physical health. 5. t9 promote health prevent disease are mtnns1c to care practice. 6. Primary care is ihformation intensive. Furthermore, they go on to emphasize that the social sciences and humanism are core to primary care and that "beyond the knowledge of disease is knowledge of the patient as a human being" (p. 81). Thus, the very lature of primary care physicians and their training may have contributed to a very holophilous, interconnected group of physicians who possessed a more favorable predisposition towards the concept of more informed actively involved patients d TBDS. 2 0

PAGE 304

Rogers (1995a) has noted that indlvidual predisposition influences behavior toward communication messages, often r suiting in individuals exposing themselves to ideas that are in accordance with their oj interests, needs, and attitudes. When compared to specialists, primary care phylicians may indeed have taken the time to keep themselves more infonned about the !concepts of active patient participation in medical decision making and TBDS servils. It would behoove practitioners to recognize the distinguishing chantcteristi1 of primary care, to be cognizant of the concepts ofhomophily and the implicatioJ for working with physicians in the implementation of social science and humanities-based innovations in health care like TBDS. Researchers interested in the dilfuion of social science and humanitites-based innovations in health care like TBDS may rnt to investigate how and why the differentiating characteristics of primary care vs. referral specialties and homophily effects the innovation-decision making proless and the subsequent rates of adoption for these innovations. Commentary on Implications for Working with Physician Organizations This study bas attempted to idenf 1 and to provide insights on physicians' opinions about patients as partners in medidal care decisions; the use of TBDS technology to support a more informed, aJively involved patient; and the impact of patient involvement and TBDS on various \ ealth care delivery parameters. In 2 1

PAGE 305

consideration of these opinions, I would like to take the opportunity to provide the reader with some personal reflections ab+t what I think these findings mean in relation to the health care marketplace and some implications for working with physician organizations. In brief, it was my overall impressJon that even though physicians in this community genuinely supported more infc rmed and actively involved patients as partners in medical decision making, most physicians: a) were ill-equipped to convert these beliefS into meaningful clinical Ices and behaviors; b) did not have a sense of urgency about this being a priority for responding to the marketplace; c) were resistant to change and skeptical about the motivatirns driving the implementation of 1BDS, d) were very uncertain about 1BDS ha, any significant value for them. For those practitioners working with physician organizations to implement new "innovations" like patient involvement and supporting terologies like 1BDS, I would thus like to share several key insights that might be heT:ful in these endeavors. 1. Recognize that timing and local\ market dynamics can potentially have a significant impact on the attitudes, belii and opinions of physicians. During the time of this study, the local community going through a significant period of change and turmoil within their entire health care delivery system. A health . d "d I . I d hi .c. th comnusston was activate to provt e cont,umg commuruty ea ers p 10r e planning, promotion, and development of tile local health care system. The whole

PAGE 306

concept of community rating, which had ieen in existence for the past forty years was being questioned and challenged. Major rmployers who were faced with increasing cost management pressures in a competitil e global marketplace and who could no longer afford to play the role of the benevrlent industrialist were changing health care benefit plans from largely fee-for-servicedlto more tightly managed physician panels comprising preferred provider, point of service, physician-hospital and self-insured insurers and physicians were battling over ssues of control, finance and clinical autonomy. Hospital consolidation had begun to occur. A more competitive health care market place was beginning to take ,rm. For practitionen; who want to iduce demand-side strategies like consumer involvement and TBDS into physician organizations within the context of a changing and dynamic local marketplace, there must be clear and consistent information and communications regarding the primary objective( s) and intent of such strategies as well as clear indicators of achievement for lhe objective(s). This communication and information should flow not only to the folal executive leadership of the physician organization, but should also be strategijy targeted to the key informal physician opinion leaders and subcommittees as wen.l 2. Particular attention should be gijen to the art and the science of innovation diffusion. First, practitioners should consil mapping physician interpersonal 2 3

PAGE 307

communication networks within the phy ician organization. The process of mapping influence networks will assist the practitilner in identifying key physician opinion leaders and influential subcommittees wiL the IP A or group practice structure who can be specifically recruited, trained, and \mobilized to serve as influential change agents when introducing new innovatioj into the organization. Second, in working with influence networks, practitioners sbluld pay attention to the attributes of an I innovation that effects their rate of adop,on. Practitioners should recognize that physicians go through several stages of decision making in deciding whether to adopt an innovation or not. Specific, targeted, strategies should be developed for moving physicians through Lch of these stages toward implementation. Practitioners should recognize thJ the perceived attributes of a given innovation effect the rate of adoption. Jh a new innovation like TBDS, practitioners should work with identified influence ne,orks and key physician opinion leaders to: a) show how TBDS provides relative advantage over the current process for informing patients, b) how compatible the technoloJ can be in supporting physicians as their own best ''TBDS supplier" for their patit c) provide the opportunity for key rank and file physicians to trial run the TBDS, and d) provide key influential physicians with the opportunity to observe how TBDS cJ work for them. The initiative for introducing TBDS into the physician of this study fuiled to undertake a methodical, systematic diffiJsion approach }or introducing TBDS as a new innovation. 2 4

PAGE 308

One wonders if such an approach had bee1 taken, whether or not the results of this study would have been different. 3. Most, not all, but most physici s will state that they genuinely believe patients generally desire more infonnatio, about health and medical issues; and, that many patients will prefer a more collaborative decision making approach in the doctorpatient relationship. However, when it coLes to converting these beliefs to changes in bh d. I Ibli practtce e aVIors an mvestment m resources as a top pnonty, e eve most physicians and physician organizations, exlept for the most enlightened ones, will I continue to struggle with this issue and will continue to be somewhat reticent about addressing it. Why? Because most provit are currently entangled in a web of change, turmoil, instability, and financial cnallenge. They are uncertain and confused as to what kind of priority to give to this ifue, how to position themselves for the new consumer, and what kinds of direct and investments to make in resources and technology. They are primarily consumed with a focus on internal administrative and financial issues rather than on external forces that are and will continue to shape and effect their eventual survival. It is the opinion of this investigator\ that, at best, most physician organizations are only in the very early stages of really anticipating, and preparing for the new emerging marketplace. It is my belief that the new emerging marketplace is 2 5

PAGE 309

being shaped by three primary forces thai that have previously shaped other sectors of the U.S. economy: consumers, technoloi, and changing organizational structures. Specifically, the baby-boomer ge eration that has shaped virtually every other sector of the U.S. economy will emerge l. the new consumer, eventually shaping the health care industry as well. This new cd sumer of health care will be more informed and will want even more information abo t quality and costs. Access to health information on the internet will continue o support this new era of consumerism. Collectively, the top five most-visited health-related Web sites are recording nine to ten million "hits" per month. This new consumer will be more ,emanding about access (Including access to alternative medicine), convenience, and service quality; will be more assertive in the doctor-patient relationship; and in general will want to take charge and be more autonomous and independent. Pending legislation regarding patient protection and the loss ofERISA protection in combination more and more companies moving to defined contribution plans rather than deflhed benefit plans may mean that more corporations may be turning over the direr purchase of health care to individual employees. With the latest predictions that annual increases in health care costs are going to return to at least the 7-10% level for most employers, coupled with the recent backlash against liMOs, market pressures r continue to encourage insurers to get out of the business of managing healthcare and return to providing traditional r

PAGE 310

insurance functions like marketing and unrerwriting. If these bold events indeed it will only serve to fuel the emerglnce of the new well-informed, more assertive consumer. As provider organizations responi to marketplace demands fur access, convenience, cost-effectiveness, service Jd clinical quality, we will begin to see more and more provider organizational structur s move to horizontally integrated structures similar to the horizontal integration that hL occurred in other sectors of the U.S. economy. These horizontally integrated Jructures will allow provider organizations to focus multidisciplinary efforts on core bress processes fur improving clinical and financial performance in specific disease states and/or procedures. The continued evolution and deployment of communicatilns technologies involving various b . f h l h II . . com mat10ns o t e mtemet, te ep one, te eVIston, computers, mteracttve vmce d d will I d response, an mteracttve VI eo contmue to encourage an support a more informed patient/consumer and shared dec ton making between patients and physicians. It is the opinion of this investigator that most physician organizations are at best in the very early stages of understandiBg and preparing for such a market. I believe it would behoove those working wilh physician organizations to clearly articulate what this new emerging marketprl ce might look like, to create a sense of urgency in preparing for these changes, an to provide the means by which physician 297

PAGE 311

organizations can be "early movers" in tegrating appropriate actions for providing information and active involvement of clnsumers/patients in the medical care process. In doing this, it is my opinion that they Jn be positioning themselves for competitive advantage in this new marketplace. 4. Physicians contend that their esire to heal and to do what's in the best interest of the patient is at the heart ofwlily they practice medicine. Twenty-five years of research on the effects of physician-palient communication has suggested that ffi . dd .\ . .. infl e ecttve commurucatton an octor-patlent mteractlon exerts a pos1t1ve uence not only on the emotional health of the pati, but also on symptom resolution, pain functional and physiological healr status. There is so much supporting and emerging psychosocial evidence for the tlierapeutic effects of the provider within the provider-patient relationship, but so little lwareness, understanding, and practical application in the real world of traditional medicine! It is the opinion of this investigator that we must do a better job ir informing and educating physicians that there is a significant amount of research sJpporting the notion that the primary detenninants of health, disease, health ices utilization, and health outcomes have their basis in the social, cultural, psychological, and psychosocial domains of the patient. This type of research data and Jrmation needs to be integrated more frequently and more in depth into non-,tional practice-based continuing medical education (CME) for physicians. If physicians are genuine in stating that their desire 298

PAGE 312

to heal is at the heart of why they are in rcine, then it will be very difficult for them to ignore the information supporting the r lative advantage that the therapeuticallyenhancing doctor-patient communications can have on the healing process. However, since there is little evide ce that traditional CME delivery methods have direct impact on improving professional practice (Davis, Thomson, Oxman, and Haynes, 1995), education and information alone will not be sufficient to change physician behavior. The practitioner may ni. to view this as a cillfusion of innovation issue with specific, systematic practice-based strategies for predisposing, enabling, and reinforcing physician behavior sJce excellent primary care is grounded in both the biomedical and the social sciences diffusion strategies that include working first and foremost with key influential care physicians should be seriously considered. 5. Even though physicians acknow edged that patients have different preferences for involvement in decision making and that physicians should be sensitive to these preferences, it is the opinion of thiJ writer that many physicians are not that well-equipped to: a) adequately address the patient preference issue and b) effectively and efficiently provide therapeutically-enhancing interactions within the doctor-patient relationship. The therapeuti1y enhancing strategies that I'm talking about include: asking a wide range of not only about the disease, but also about the illness experience as related to the patient's feelings, concerns, and

PAGE 313

understandings of the problem; encourar patients to be active participants in care, encouraging patients to ask questions; ploviding clear verbal, written, and emotional support in response to patient's questionk; sharing power; and providing psychosocial support in collaborative decision making] I believe that the annual continul medical education (CME) requirements for physicians, at a minimum, should include a physician-patient communication training component that sensitizes physicians to i.e patient preference issue, informs them of the research about therapeutically-enhancr connnunicarions with patients, and gives them the skills and training that enables tliem to put it into practice. This type of training would also address the concern t some physicians have expressed about appearing to be "uncaring" or a "wet noodle" in the eyes of the patient when the physician attempts to involve them as a in the decision-making process. I've often wondered, what might be the of organizations like the American Medical Association or the American Aca(iemy of Family Physicians requiring so many hours of non-traditional practice-based C [ in "therapeutically-enhancing" doctorpatient communication skills? 6. Physicians in this study viewe, themselves as the best source of medical information and the "mDS supplier of chLce" for their patients. It is my perception that some physicians responded this way bLuse they genuinely believed that they do a good job in informing and actively involing their patients while other physicians 300

PAGE 314

responded this way because they were skrptical of a mDS system that they had no ownership in creating. Nonetheless, I tbif that most physicians realize that they are limited in how much time they can spend in direct contact or in phone contact with patients and that if they truly want +nned, actively involved patients they must look outside themselves for resources theJ are comfortable with. I want to be clear that TBDS doe not necessarily need to be the technology of choice. We need to keep in mind that it's not a specific technology per se that is the most important issue here. Rather, physiaan organizations need to understand that informed consumers are not going to go aLay. The new consumer wants access and convenience to health and medical infoion and they will continue to access information from various sources. They i be utilizing those sources that are of the highest quality and that are the most time and cost-effective. To this end, I would I suggest that provider organizations take a leadership role in responding to the new consumer by working with purchasers, insters, and consumer advocacy groups in structuring an information chassis fur medire that strategieally integrates and bundles various fonns of information and communications for patients using various b . f h I h I . com mat1ons o t e mtemet, te ep one, mteract1ve vo1ce response, computertelevision, and clinical support groups that LY positions them as the "supplier of choice" for decision support with their patir. I believe that the physician organizations who are truly passionate and enlightened about this issue will find 301

PAGE 315

creative ways to differentiate themselves in the marketplace by proactively establishing technology linkages with their patients t at will allow patients an "interconnectivity'' to them and health/medical information J4-hours a day. 7. Last but not least, it is the opt on of this investigator that there has not yet been a strong business case for both a more informed, actively involved consumer and/or TBDS in health care. A strong buliness case would include several components. First, there would be ditrable increases in patient health and/or satisfaction with concomitant decreases in costs and resource utilization. Second, purchasers would be willing to pay a hi+ premium and insurers would be willing to pay higher fees to physicians who were alle to demonstrate improved quality and decreased costs using these strategies. Third, increased market share would go to h I d d d ed p ys1c1an orgaruzat1ons m response to Improve qu 1ty an ecreas costs usmg these strategies. Given the financial pre+ that physicians are currently undergoing, it seems to me that Sl.ich a business case must be created before physicians will passionately embrace these strategies. I bllieve that physicians' skeptical opinions about TBDS and it's potential impact as in this snuly were also influenced by the fact that not all of the components for l strong business case were in place. As previously described, at the time of this Jdy, physicians were operating in a marketplace that was heavily involved in eLning supply-side cost control strategies. It is my opinion that many physicians resistd the TBDS service and were uncertain 302

PAGE 316

about its value because they believed tha TBDS seiVices were just another strategy being implemented by business and insur rs to control costs their own costs without concern for patient or physician welfare. In retrospect, the Rochester community initiative failed to fully develop, impleme t and effectively communicate all three components of the business case above. Also, the community initiative fell short in assuring that the most significant larts of the "business case" that had been developed for the initiative, with signifil input from the fonnal physician leaders, I had been shared with the most influential informal physician opinion leaders. What's needed is a shining examp e of a care management system that demonstrates what it means to have more informed, actively involved patients working with providers who structure therapeutically-enhancing doctor-patient relationships. With the previously described forces that ke shaping the new healthcare marketplace, I would encourage physician organizationr to heed the call for leadership in this arena by developing some beta sites and pilot programs that truly demonstrate the clinical and financial efficacy of this approach. Thl timing is ripe for such clinical process I improvement initiatives. Reiser p. 016) captures the essence of this important time in history by stating: As we reach the last decad of the 20th century, it is possible to recognize the emergenc9 of a medical era that engages the perspectives of patients lin the transformation of health care. In many guises and Pflicies, our experiences with illness are becoming keys to changing the health care system . . the passive roiJ of patients and consumers in 3 3

PAGE 317

the health care system sh uld be replaced by an active one .... a patient and coysumer-based perspective that draws on personal experi9nce with illness that can improve health care practice, education, research, and policy, and demonstrates the developfuent of a new stage in health care the era of the patie t. It is my hope that within this new era of consumerism, physicians will recognize the leadership opportunity the have to launch the single best platform we have for advancing a more effective, and humane paradigm of health and health care --one that is built on the foWJ.dation of an integrated biomedical social science approach. If physicians take the lLd in collaborating with each other and with patients/consumers to take full advantage of the historical window of opportunity that has been granted, early indicators signal a great opportunity for structuring the future logic of health care reform; and in shapin the next, and possibly the most impacting and rewarding social transfonnation of tcan medicine and public healt:lt LimitatiOnS Interpretation of the results of the study warrants caution. Our response rate for the Physician Opinion sJrvey (POS) was not high ( 42% ). Unfortunately, low response rates have beln historically ubiquitous in surveys of physician populations (Cykert, Hansen Lison & Joines, 1997). As with any population survey, factors affecting self se ection may have played a disproportionate 304

PAGE 318

role. Demographic information obtained from the IP A organizations involved in this study indicated that there was no dice among responders and non-responders in the distribution of primary care versus specialist physicians and male physicians versus female. However, since information aboJ the remaining non-respondent practice characteristics was unavailable, response bias may have affected the results. Even though we have no reason to believe that survey respondents were any different than non-respondents, the interpretation of an specific trends warrants prudence until consistent results can be obtained in adJonal study groups. Even though the development of Juestions for the POS was guided and informed by the substantial body development and patient research in this area (Cassileth, Zupkis, Sutton-Smith, Mlch, 1980; Strull, Lo & Charles, 1984; Ende, Kazis, Ash & Moskowitz, 1989; Suther! d, Llewellyn-Thomas, Lockwood, Tritchler, & Till, 1989; Brody Miller Lerman Smith Caputo, 1989; Beisecker and Beisecker, 1990; Neufield, Degner & Dick, 1993; Darson, Degner, & Morgan, 1995), measures of test-retest reliability and validity were nbt conducted on the POS. A I b f h .I :_c. . arge num er 0 cases Wit nussrg uuormatton can create maccurate or biased results. The results related to pract1ce visits/week and % of capitated patients should be interpreted with extreme cautioJ since each of these POS variables had a high percentage (19%) of missing values. 305

PAGE 319

The use of qualitative methods lay be controversiaL Patton ( 1990, p. 461) has emphasized that the credibility issue rr qualitative inquiiy is dependent upon how well the following elements have been rsed: What techniques and were used to ensure that rigorous techniques and Dfethods for gathering high quality data that were carefully analyzed, with attention to issues of validity, reliabtlity? What does the researcher bring to the study in termk of qualifications, experience, and perspective? What pradigm orientation and assumptions undergird the study? Steps taken in this study to address each of these elements and associated limitations will be discussed. Patton (1990) and Kuzel ( 1992) ie provide some guidelines and criteria for sample selection and sample size when conducting qualitative research. Patton (1990) has pointed out that, "Qualitative inquiry lypically focuses in depth on relatively small samples, even single cases, selected pwp,sefully .... The logic and power of purposeful sampling lies in selecting information-rich cases for study in depth" (p. 169). To enhance sampling rigor and identification of information-rich cases for in-depth study, the present study empltyed a combination of purposeful sampling techniques: maximum variation sampling and snowball sampling. "Maximum variation sampling occLs when one seeks to obtain the broadest I range of information and perspectives on the subject of study. Guba and Lincoln (1989) claim this is the preferred strategy ;or qualitative inquiry'' (Kuzel, 1992, p. 37). 306

PAGE 320

Snowball sampling is an approach for lo ating information-rich key informants by asking well-positioned people within the opulation under study "who knows a lot about ?"and "who should I talk to?" In most groups of people, a few key names are mentioned repeatedly. Kuzel 1992) has noted that when conducting maximum variation sampling, subject selection continues to the point of saturation or redundancy with 12-20 data sources reco ended. perspectives on physicians' opinions abol' TBDS services and patient involvement in decision making. That is, through snowb sampling, peer physicians purposefully identified seven key infurmant physicians from those groups of primary care physicians and specialist who reported being positive, neutral, or negative about TBDS I services. This provided a total sample of 21 key physician informants for the qualitative portion of the study. One poteltial limitation of this sample may have been gender bias. Out of the 21 key physician lronnants, only one (5%) was female. This is in contrast to the overall physician popjation for this study which was comprised of approximately 25% females. Future stud+ should be undertaken to determine if female physicians hold significantly differelt opinions and perspectives than male physicians on the issues of concern in this tudy. Such findings could have significant ami.fi h I. d h ed d akin r cat10ns 10r p ystctan-patlent comm cat10ns an s ar ectston m g. 3(i)7

PAGE 321

A standardized open-ended inte ew was designed to collect the qualitative information for this study. A standardizj open-ended interview was selected as the instrument of choice for this study because of it's inherent strengths: comparability of responses are increased because respondfts answer the same questions, reduces interviewer effects and bias, permits othel researchers to review and use the instrument, facilitates organization and of the data (Patton, 1990). Like the POS, the question development for the st j dardized open-ended interview was guided and informed by the substantial body of s rvey development and patient research in this area (Cassileth, Zupkis, Sutton-Smith, March, 1980; Strull, Lo & Charles, 1984; Endc; Kazis, Ash & Moskowitz, 1989; srland, uewellyn-Thomas, Lockwood, Tritchler, & Till, 1989; Brody Miller Lelan Smith & Caputo, 1989; Beisecker and Beisecker, 1990; Neufield, Degner & 1993; Davison, Degner, & Morgan, 1995). Patton (1990;p. 347-348) has stated that: No matter what style of is used, and no matter how carefully one words interview questions, it all comes to naught if the interviewer rails to capture the actual words of the person being interviewetl. The raw data of interviews are the actual quotations spokef by interviewees. There is no substitute for these data .. \the purpose of each interview is to record as fully and fairly as possible that particular interviewee's perspective, Some method for recording the verbatim responses of people being interviewed is, therefore, essential. For this study, the suggested standard for r cording the interview data as described by Patton (1992) was addressed. All intervie s were audiotaped for subsequent

PAGE 322

transcription, coding, and Additionally, Patton (1990; p. 349) notes that, "Because the raw data of interviews are uotations, the most desirable data to obtain would be full transcription of interviews.' All interviews in this study were transcribed verbatim. Methods for enhancing the rigor f qualitative analysis were addressed in this study. The qualitative data were coded + analyzed utilizing a template analytic approach (Crabtree and Miller 1992). 'J1fs involved beginning with a basic set of a priori codes based on theoretical and previous research and then expanding on these through iterative rea gs and interpretation of the text. Crabtree and Miller (1992; p. 108) have stated that one of the limitations with the template analytic approach is the "potential for mit information, especially if the codebook is produced in a completely a mannr analyst runs the danger of not looking beyond the codes." For this steps were taken to "look beyond the codes" and to addresJ this potenttallimitatton. First, using an initial set of a priori [codes, the primaiy investigator in this study expanded on the initial codes through an itl.nv. process of coding, analyses, inteq>retation, and recoding of salient pattLs, themes and categories. As a part of this iterative process, a second investigato1 independently expanded and refined the set of codes established by the primary Collaboration and consensus between the two primary investigators detJ ed the "final" codebook for the text 3@9

PAGE 323

analysis and interpretation. Overall, the iferative process encompassed ten passes of coding, analysis, interpretation, and recoding for the final codebook. Second, analysis of the data included testing for rival expllations (Patton 1990) and negative case I analysis (Erlandson, Harris, Skipper & 1 en, 1993) which involved examining alternative interpretations of the data that might refute the initial interpretation Third, a question-by-question analysis was cond cted to delineate directionality of opinion or how the key infOrmants felt about a topic (positive, negative, netural) as well as why they felt they way they did. Fourth, the two primary investigators independently coded a sample of questions to test for intlrcoder reliability. The intercoder reliability was 93%. Finally, a funn of modified m+er checking was conducted. "Member checking provides for credibility by allowing members of stake holding groups to test categories, interpretations, and conclusionl (Erlandson, Harris, Skipper & Allen, 1993 p. 142). Notwithstanding, a limitation of this study was that various logistical, resource and financial constraints did not allow us Jo conduct final member checking with the key informllllt physicians who were a part rfthis study. However, we were able to conduct a modified form of member checking or peer debriefing (Erlandson, Haris, Skipper & Allen, 1993). A copy of the pr liminary analysis and interpretation was provided to three independent physician re ewers who were not a part of the study 310

PAGE 324

population for critique and feedback. ior adjustments and revisions from the independent physician reviewers were into the final analysis. Patton (1990; p. 187) has stated that, "One way to strengthen a study design is through triangulation, or the combination of methodologies in the study of the same phenomenon or programs." This study u ilized methods triangulation to verify the consistency of and to validate the finding generated by the quantitative and qualitative portions of the study. Throughout the and discussion sections of this study, we have attempted to provide a comparative 8na.lysis of the consistencies and discrepancies that existed between the quLtitative and qualitative portions of the study. l Even though the primary investiga or of this study has a very strong appreciation for naturalistic inquiry and qjalitative methods and has had experience in conducting physician interviews, focus grl\ups, and action research, another limitation of this study may be the inexperience of th primary investigator in conducting formal qualitative research. A second is related to the fact that during the data collection phase of this study, the p+ investigator was a strategist with the local health care community working closely with employers, providers, and insurers in the development of health care cost coJainment and managed care strategies. He had intimate knowledge of the organizatioJal, cultural, economic, and political issues surrounding the significant transitions the e\ ire medical delivery system in this 3 1

PAGE 325

community was undergoing at the time. n the one hand, this intimate engagement with the local health care community mal have provided this investigator with increased access, as well as vital insights egarding the issues associated with the aims of this study that would not be apparent r available to researchers external to the system. On the other hand, unknown is tolwhat extent this involvement may have biased the data collection, analysis and int rpretation. Acknowledging these limitations t the potential for introducing personal bias, every attempt was made to adhere tJo the standards, guidelines, and criteria for conducting qualitative research as outlin by some of the most prominent experts I (Patton, 1990; Crabtree and Miller, 1992; Erlandson, Harris, Skipper & Allen, 1993) in the field. Additionally, oversight of th1 primary investigator was provided by an independent team of seasoned researchers with experience in both quantitative and qualitative research. Independent reviews and critique of data collection, analysis, and interpretation were provided by this team as a standard component of the research process. And, as previously noted, this + utilized methods triangulation to verifY the consistency of and to validate the find!. gs generated by both the quantitative and qualitative portions of the study. Finally, s ould other investigators have an interest in testing the transferability, dependability, con:firmability of these results, all audiotapes, transcipts, and records are avai able for conducting an independent audit (Erlandson, Harris, Skipper & Allen, 1993). 3 2

PAGE 326

The present study is limited to mostly middle-aged male European American volunteers in a solo or single-specialty grJup IP A practice structure with a mix of fee for -service and capitated reimbursement Jom a single urban community in one region of the country. Generalizability to male or female physicians in other ethnic groups, practice structures, compensation arrangerents, or rural environments is not known and should be done with caution. Patton (1990; p. 471) has noted that, "The importance of reporting both methods an results in their proper context can not be overemphasized." It is worthwhile to note that data 1 llection for this study occurred during a time period when this particular health care community was in significant turmoil and transition. Specifically, considerable was being given to health care cost containment by the employers. Physicians and insurers were in the midst of reacting to cost containment pressures through the de elopment of new provider arrangements and insurance products. For the first time 1n four decades, this community was experiencing significant change and upbeat in its entire health care delivery system. It is plausible that physicians' opinions ma!have been influenced by the socioeconomic context of this time period. Even though the socioeconomic. conditions of this healthcare marketplace y have paralleled those in other communities, generalizability of results to I\ther time periods and contexts should be done with caution. Future studies should e amine to what extent the context and the 3 3

PAGE 327

socioeconomic dynamics of the local h thcare marketplace effect physicians' opinions about shared decision making d TBDS services. Recommendation for Future Research This research was designed to berl rimarily descriptive and hypothesis generating. The results of this study hav suggested numerous recommendations for further research and evaluation. Ho4 it is the opinion of this investigator that in order to advance a more prominent role !r patient involvement in health and medical decisions into 21st century health care, to priority should be given to the following two types of studies: l I. Determine if there is a "busine s case" for more informed, actively involved consumers/patients in health and medical care decisions in both fee-forservice and capitated reimbursement ("Business case" is defined as having three components: a) demonstrabl increases in patient health and/or satisfaction with concomitant decreases in costs and_ resource utilization, b) purchasers would be willing to pay a higher p emium and insurers would be willing to pay higher fees to physicians who were able, demonstrate improved quality and decreased costs utilizing patient preference and involvement strategies, c) increased I market share WQuld go to physician organizations in response to improved quality and decreased costs using patient preference \ d involvement strategies). Given the 314

PAGE 328

financial pressures that physicians are ,ently facing, if at a mininrum we can not demonstrate that more informed and actilely involved patients can indeed have a positive impact on clinical quality while rdducing costs and resource utilization, then I believe that providers, purchasers, and pa ers will be less than passionate about embracing this philosophy as a sustainabl strategy. 2. Even after we demonstrate the j fficacy and/or effectiveness of more informed, actively involved patients in hr and medical care decisions, the challenge that still remains before us is how to accelerate the adoption of more informed, actively involved patients as a "best practiJe" innovation. The gap between medical science and medical practice (Greer, as well as the gap between best achievable practices and actual practice (Berwick, 19b8) is well-established. Many medical innovations may take years or even decadjs to reach widespread adoption. Therefore, I believed we need to conduct diffusion rerh to assess how the structural, organizational and systems characteristics of physicians' practices effect the relative speed at which a formal focus and emphasil on patient preference and involvement in decision making are adopted. Particular a,emion should be given to: a) how the perceived attributes (relative advantage, compatibility, complexity, trialability, observability) of more informed and involved patients effect the relative speed of adoption; b) how this perception is convr through physician interpersonal communication networks and the effect on adoption rates; and c) what characteristics 3 5

PAGE 329

ofkey influential physician opinion leadr are the most predictive in determining adoption rates. It would also be ofvalu1to: a) conduct comparative analyses of successful vs. unsuccessful diffusion, b) tilize systems vs. individuals as the unit of on diffusion rates. For those more interested in the DS services as a consumer health management and decision support per se, it is the opinion of this investigator that top priority should be giren to the previously described types of studies as well. However, it is worthwhile to point out that with the advent of widespread access to health infonnation i the internet, TBDS as cwrently structured may become somewhat obsolete. It may he of more value to conduct comparative studies that assess the clinical performancl improvement and resource effectiveness of TBDS vs. other forms of consumer decisitn support technologies such as the internet, interactive voice response, and clinical support groups. Other recommendations for future research and evaluation that have emanated fiom this study include (not necessarily 'T ordered for priority): 1. Determine if the findings of this study are generalizeable to physicians in other communities and regions of the cojtry with the same physician/practice characteristics and with differing physici,practice characteristics (male vs. female; Caucasian vs. African American vs. Mexican American, vs. Asian; multispecialty 316

PAGE 330

group practice vs. single specialty group ractice vs. IP A:, inner city vs. suburban vs. rural). 2. Investigate the epidemiology o discordance between patient and physician preferences for information and active invllvement in medical decisions. Evaluate the effect of physician-patient discordance ont patient satisfaction, patient health outcomes, patient retention, physician resource utilization, costs, clinical efficiencies, the doctor-patient relationshi,, and malpractice. 3. Utilize qualitative inquiry to determine if female physicians hold significantly different opinions and perspectives than jale physicians about patient involvement in medical decisions and TBDS for patients. 4. Compare physicians' opinions bout more informed actively involved patients to actual physician behaviors and wractice style within the doctor-patient relationship. j 5. Examine the effect of physici opinions about patient involvement in medical decision making and associated prlctice behaviors on: patient satisfaction, health outcomes, utilization, costs, clinicallfficiencies, physician productivity, h f1 al I d h d p ysiClan satis action, m practice, patient etent10n, an t e octor-patient relationship. 3 7

PAGE 331

6. Explore how patients view piysicians who give them a more prominent role in the health and medical care decision L or who encourage them to utilize alternative sources of information such a;s TBDS and the internet. 7. Examine to what extent structures serve as incentives or disincentives to physicians talking with l\ tients and the effect on physician-patient communications, the doctor-patient rela onship, patient health outcomes, and physician productivity. 8. Assess to what extent physicians' opinions about patient involvement in medical decisions and TBDS ovJ time with each stage of the innovationdecision making process. 9. Determine what patient infonnation and decision-support technologies (TBDS, internet, clinical support groups, personal coaching, etc.) are the most effective and efficient in altering physician behaviors relative to patient preference and involvement in medical decisions. j 1 0. Using the data from this stud conduct a path analyses utilizing physician/practice characteristics and op4on scores to further develop, refine, and explain the theoretical and logical relatioJhips among these variables. 318

PAGE 332

Concl sions Multi-method triangulation corroJorated a number of key findings in this study. A majority of physicians believed patients generally desire more information about health and medical issfs and that patients generally desire more active involvement in the medical decisioj"making process. They believed that a more informed patient would result in better health and medical care decisions and a greater sense of control and self-confidence fur thl patient, ultimately leading to improved quality of care, enhanced health outcomes (for chronic conditions and major medical issues), higher patient satisfaction, enricheti doctor-patient relationships, potential reduction in medical liability; and for minor illnesses and chronic conditions improved clinical efficiencies (reduction in the number of medically inappropriate visits and volume ofunnecessary phone calls) ,d cost control. However, physicians varied in their preference for involving patients inl decision making and, believed that patients have varying preferences fur involvement r decision making. Even though there was strong and endorsement for the concept of a more informed and actively involved patieht in the medical care decision making th . d dl . d" h "al process, ere were uncerta.J.nt1es an vane opm10ns regar mg t e purposes, potent1 impacts and reasons for why health plans hL implemented a TBDS service for nh . -C. d I I d" al d . e ancmg patient 11uormat10n an mvo ve ent m me IC ecisions. 3 9

PAGE 333

Primary care physicians (general 1 ractice, family practice, general internal medicine, and general pediatrics) were mrre likely than specialists to: 1) agree that their patients should take a more active r 1 le in treating their own minor conditions; 2) agree that their patients should take a feater role in self-managing chronic conditions; 3) be familiar with and knowledgeable about the TBDS services; 4) have a more positive opinion about TBDS servils; 5) believe that the health plans implemented mDS to respond to interesl of the business community, to more efficiently care for capitated patients, impii ove quality of care, and to improve member satisfaction; 6) believe that TBDS service would decrease the number of medically inappropriate visits and the volume of+ary calls. To what extent are to physician organizations IS unknown. Further quantitative and qu,tative research IS warranted by those interested in consumer health shared decision making, physician-patient communications, TBDS, and the diffusion of health care innovations. 320

PAGE 334

APPEND A PHYSICIAN OPINIO SURVEY (POS) 321

PAGE 335

PHYSICIAN qPINION SURVEY PATIENT INVOLVEMENfl IN MEDICAL DECISIONS and DECISION SUPPORT The purpose of this survey is to collect lormation about RCIP A and RIP A physicians' perceptions and opinions patient involvement in medical decisions and about a 24-hour telephone-based decfsion-support service that begins on June 1 for 80,000 families in Rochester. This counseling service is a non-directive, non diagnostic program that provides decisioJ-support, self-management support, and information to patients. Please respond tb these questions as candidly and directly as possible. Your responses are confidenti and will be known only to researchers at the University of Colorado, who will interpre the aggregated results. Please return the sunrey by June 7. Thanks for your help. A. Background 1. Gender: [ ] Male [ ] Female 2. Year of birth: __ 3. Year completed medical or osteopathic training: __ 4. Where did you do your residency traioiog? [ ] I am not residency trained [ ] Rochester [ ] Other; Specify: _______ ____,,___ ____ 5. Within the IP A structure are you considered: [ ] A Primary Care Provider _Family Medicine Internal Medicine OB/GYN General Pediatrics _Other, please specify-----+-------[ ] A Specialist What is your medical specialty?--+---------

PAGE 336

6. Please indicate practice structure: a. b. c. Solo practice Group practice, single specialty Group practice, multi-specialty Corporation (hospital-owned) [ ] Yes [ ] Yes [ ] Yes [ ] Yes []No [] No []No []No d. e. Other, specify:--------+-------7.Are you: [ ] Salaried 8. Excluding daily hospital rounds, approximat ly how many visits does the entire practice conduct in a typical week? __ 9. Approximately what percentage of your pa nts come from a capitated plan? __ % B. General Opinions and Perceptions The following section includes questions on youm general opinions and perceptions about patient in volvement in medical care. 1. Who would you want to make decisions abou treatment for your patient's medical issues? Only your Mostly your You and your Mostly you with patient patient-patient input from your Only you with input together patient from you A. For a minor 1 2 3 4 5 Health ISSUe B. Fora major I 2 3 4 5 Health ISSUe 323

PAGE 337

2. Please indicate the extent to which you agree or disagree with the following statements: Circle the number that best describes your views. I Strongly I No I Agree O(!inion Disagree Patients want more involvement in their health and medical care 1 3 4 5 6 7 decision making process. I Patients should be provided with a 1 2 3 4 5 6 7 range of treatment options available to them. Patients should be encouraged to 1 1 3 4 5 6 7 ask more questions about their medical care. Physicians do not have adequate 1 2 3 4 5 6 7 time to fully involve patients in the decision-making process. Patients expect their physicians to 1 3 4 5 6 7 discuss the risks and benefits of all available treatment options. Patients want their physicians to 1 3 4 5 6 7 make decisions for them. Physicians do not have adequate 1 2 3 4 5 6 7 time to answer patient questions about medical care. Patients want more information 1 2 3 4 5 6 7 regarding health and medical issues. Medical information obtained by 1 2 3 4 5 6 7 patients through newspapers, televisions, and lay literature is reliable. Patients who are fully informed 1 2 3 4 5 6 7 about the risks and benefits of all treatment options generally choose options that are less invasive, less risky, and less expensive. 324

PAGE 338

3. The following questions ask you to indicate t e extent to which you agree or disagree with the following statements: Circle the number that best describes your views. I would like my patients to take a greater role I Strongly No I in.-Opinion Agree Dill agree the medical decision making process. 1 2 3 4 5 6 7 asking questions about their care. 1 2 3 4 5 6 7 treating their own minor conditions. 1 2 3 4 5 6 7 seeking available information about a medical 1 2 3 4 5 6 7 problem or condition. self managing their chronic conditions (e.g. 1 2 3 4 5 6 "7 diabetes, asthma). C. Opinions About Informed Consent For the following two questions please check w ich of the following statements best describes your views about informed consent. 1. Check one of the foDowing: [ ] Informed consent is mostly for the protection fthe physician. [ ] Informed consent is mostly for the protection lfthe patient. [ ] Informed consent is for the protection of both l physician and the patient. 2. Check one of the following: [ [ ] Informed consent means the physicim descri s all potential risks of a chosen treatment. [ ] lnfonned consent means the physician the patient oftre-alternative. [ ] Informed consent means discussing with pati ts all treatment alternatives md allowing them to state their treatment preference. 3(25

PAGE 339

D. 0 inions and Perce tions About Tele hone-Based Decision Su Services The following section includes more specific questions about your opinions and perceptions of a 24-hour telephone-based decision-support service that begins on June 1 for 80,000 families in Rochester. This counseting service is a non-direc:tive, non-diagnostic program that provides decision-support, self-management support, an information to patients. 1. Please check which of the following statemen s best descn"bes your knowledge and awareness of the TBDS Service. [ ] I have never heard of the TBDS service. [ 1 I have heard of the TBDS .ervice but mn:f little about its pmpose and available services. [ ] I have heard of the TBDS service and am ar with its pWJ>ose and available services. 2. Please check which of the foUowing statemen s best describes your opinion about the TBDS senice. [ ] Generally Positive [ ] Neutral [ ] Generally Negative [ ] I have never heard of the TBDS service. 3. In this next section we would like you to indi te the extent to which you be6eve the following are reasons why Rochester health plans are implem nting the TBDS senice. Circle the number that best describes your views. TBDS is being implemented to ... No I Opinion Dasagree Improve the quality of care. 1 2 3 4 5 6 7 Improve member satisfaction. 1 2 3 4 5 6 7 Reduce health care costs. 1 2 3 4 5 6 7 Respond to interests of the Rochester 1 2 3 4 5 6 7 business community. Improve patient health outcomes. 1 2 3 4 5 6 7 Reduce number of medical liability 1 2 3 4 5 6 7 claims. More efficiently care for capitated 1 2 3 4 5 6 7 patients. r

PAGE 340

4. In this section we ask your views about the impact of the TBDS service in several areas. Circle the number that best describes your views. I I I No I Change following as a result of implementing Incjase the TBDS service? Volume of calls to your practice 1 2 J 4 5 6 7 Number of patient questions 1 2 3 4 5 6 7 Number of inappropriate medical visits to 1 2 J 4 5 6 7 your practice Patient demand for services 1 2 J 4 5 6 7 Number of patients avoiding necessary 1 2 J 4 5 6 7 treatment Number of medical liability claims 1 2 J 4 5 6 7 Average length of clinical encounter 1 2 J 4 5 6 7 Physician reimbursement 1 2 3 4 5 6 7 b. What do you think will happen to the I+ No foUowing as a result of implementing the change Worsen TBDS service? Health outcomes 1 2 3 4 5 6 7 Quality of physician-patient relationship 1 2 3 4 5 6 7 Continuity of patient medical care 1 2 3 4 5 6 7 Patient satisfaction with medical care 1 2 3 4 5 6 7 Retention of members within the health 1 2 3 4 5 6 7 plan E. Additional Comments =...,_ ..... ottbecballeogeo yoo urfollowing .. implelaeal-oflheTBDS 2. Please describe some of the benefits you antici ate foUowing the implementation of the TBDS service. 3 7

PAGE 341

APPENDIXB PHYSICIAN IN'IERVIEW (POI) 28

PAGE 342

Physician Interview Guide Patient Involvement in Medical Decisions Telephone-Bas Decision Support Introduction to Interview/Opening Statements a. Overview/background b. Purpose of interview c. Process and interview content d. Tape recording e. Confidentiality; use of information Informant questions/concerns A. Consumer/patient Information and volvement General I First, I would like to ask you some generai questions regarding your opinion about patient involvement in health and medical r decisions. I. What do you think about patients wanting more information regarding health and medical issues? 2. What do you think about patients wanting more active participation in the medical care decision-making process? 3. What do you think will be the impact Oljl health status/outcomes of those patients who actively seek out information and more actively participate in medical decision making (better, same, worse)? [Probe rationale] B. Consumer/patient Information and InvolvementSpecific Decisions We've been talking generally about patient involvement in health and medical care decisions. Now I'd like to ask you questions that are more specific in regards to patient involvement in health and medical care decisions. 329

PAGE 343

4. For a minor health issue (sprains, str s, cold/flu symptoms), who would you want to make decisions about treatmer? Only your Mostly your You an!d Mostly you Only you patient patient with your with input input from togeth \r from your _y_ou _patient 1 2 3 I 4 5 Please provide the reasoning your opinion. (Probe impact on: doctor patient relationship, clinical efficiendY, health outcomes, patient satisfaction, health care costs, medica/liability). 5. For a major health issue (hospitalizati?n, chemotherapy, surgery), who would yo want to make decisions about treatmert? Only your Mostly your You and Mostly you Only you patient patient with your pat ent with input input from together from your you patient 1 2 3 I 4 5 .I Please proVIde the reasorung supportmg your oplillon. (Probe unpact on: doctor patient relationship, clinical efficiencf,, health outcomes, patient satisfaction, health care costs, medica/liability). 6. What do you think about your patients taking a more active role in the selfmanagement of chronic disease conditions (e.g. asthma, arthritis, diabetes, hypertension, low back pain)? (Probe Jnpact on: doctor-patient relationship, clinical efficiency, health outcomes, pJtient satisfaction, health care costs, medica/liability). 3130

PAGE 344

C. Telephone-Based Decision Support Service (TBDS) The following section includes more specrfic questions about your opinions of the 24-hour telephone-based decision suppotf service that began on June 1, 1996, for 80,000 families in the community. This service is a non-directive, non-diagnostic program that provides decision-support, Jelf-management support, and information to patients. 7. Please indicate which of the following statements best describes your knowledge and awareness of the TBD : I have never heard of the service. I _I have heard of the TBDS se]'ce but know very little about its purpose and available services. _I have heard of the TBDS service and am familiar with its purpose and available services. 8. Please indicate which of the following est describes your opinion about the TBDS sernce: Generally positive Neutral _Generally negative 9. In this section, we would like to know our opinion about the potential impact of the TBDS service. Please describe hof you think the TBDS service will will impact your patients and your (Probe impact on: doctor-patient relationship, health outcomes, patient satisfaction, contiflllity of care, patient demand for services, flUJzber of inappropriate medical visits, volume of calls, average length of clinical encounter, medica/liability). 10. What will be the most significant challe ges you anticipate following the implementation of the TBDS service? 3 1

PAGE 345

11. How do you propose each of these challenges be addressed? 12. What will be the most significant benlfits you anticipate following the implementation of the TBDS service?. D. Closing Before we end this interview 13. What else would you like me to kno about your thoughts regarding a more active involvement of patients in heal and medical decisions? 14. What else would you like me to kno about your thoughts regarding the value of the TBDS service to the communi ? E. Demographics 15.1 Gender: Male Female 15.2 Year ofbirth: __ 15.3 Year completed medical or osteopat c training: __ 15.4 Where did you do your residency trai g? I am not residency trained. Rochester _Other (Specify): ____ -+----------3 2

PAGE 346

15.5 Please indicate your practice stru e: Solo practice 2-person partnership Family practice group Pediatric practice group Multispecialty group Corporation (hospital owned) Yes Yes Yes Yes Yes Yes No No No No No No Other (Specify):-----+------------15.6 Years of practice in Rochester: __ 15.7 Approximately what percentage fyour patients come from a capitated plan? % Oosing Statements a. Express interviewer's interest in other oughts, opinions, feelings infonnant would like to share following this interview (provide with telephone number and address). I b. Inquire about other possible key infonifants to interview. c. Review feedback process and d. Express appreciation/thanks to info t. 333

PAGE 347

APPEriXC QUALITATIVE AN YSIS CODEBOOK

PAGE 348

Node Location (1) (2) (2 1) (2 2) (2 4) (2 5) (2 6) (2 7) (2 8) (2 9) (2 10) (2 11) (2 12) (3) (3 1) (3 2) (3 3) (3 4) (3 5) (3 7) (3 8) (3 9) (3 10) (3 11) (3 12) (3 13) (4) (4 1) (4 2) (4 3) (4 4) (4 5) (4 8) (4 10) (4 11) (4 12 (4 13) (4 14) Node Titl s /CASEDA'l'A /MORE :IN!' mG.'l':ION /MORE INFQRMATION/DR-PATIENT RELATIONSHIP /MORE INF0RMATION/OUTCOMES & QUALITY /MORE INF0RMATION/PATIENT SELF-EFFICACY /MORE INFqRMATION/EFFICIENCY/APPROP /MORE INFORMATION/MEDICAL LIABILITY /MORE INFqRMATION/PHYSICIAN AS TEACHER /MORE INFqRMATION/CONSUMERISM /MORE INFdRMATION/CONSUMER INFORMATION /MORE INFgRMATION/BETTER DECISIONS /MORE INFORMATION/CHRONIC ILLNESS /MORE INFdRMATION/PATIENT PREFERENCE /P.AR'l':IC:IPJ'l':ION RELATIONSHIP OUTCOMES/QUALITY /PARTICIP$ ION/PATIENT SATISFACTION /PARTICIP ION/PATIENT SELF-EFFICACY AS TEACHER /PARTICIPAkiON/PATIENT PREFERENCE /PARTICIPAkiON/SEVERITY DECISIONS /PARTICIPAriON/ROCHESTER QUALITY /PARTICIPATION/INFORMED CONSENT /PARTICIPAtiON/CONSUMER CHOICE /HEALTH AND OUTCOMI!lS /HEALTH AND OUTCOMES/DR-PT RELATNS /HEALTH AND OUTCOMES/PT SELF-EFFIC /HEALTH AND OUTCOMES/PATIENT SATIS /HEALTH ANDOUTCOMES/MED LIABILITY /HEALTH AND OUTCOMES/DR AS TEACHER /HEALTH AND OUTCOMES/CIA MEDICINE /HEALTH STtTUS AND OUTCOMES/PT PREFERENCE /HEALTH AND OUTCOMES EFFIC/APPROP /HEALTH STATUS AND OUTCOMES/OUTCOMES /HEALTH ST TUS AND OUTCOMES/DISEASE DEP /HEALTH ST TUS AND OUTCOMES/CONSUM-CHOICE 33

PAGE 349

Node Location (5) (5 1) (5 2) (5 3) (5 4) (5 5) (5 6) (5 7) (5 8) (5 9) (5 10) (5 11) (5 12) (5 13) (5 14) (5 15) (5 16) (5 17) (5 18) (6) (6 1) (6 2) (6 3) (6 4) (6 5) ( 6 6) (6 7) (6 8) (6 9) (6 10) (6 11) (6 12) ( 6 13) (6 14) (6 15) (6 16) (7) (7 1) (7 2) (7 3) (7 4) (7 5) (7 6) (7 7) Node Tit.f.es I /MINOR DECXSXOHS /MINOR DECISIONS/DR-PATIENT RELATIONSHIP /MINOR DECISIONS/EFFIC/APPROPRIATENESS /MINOR OUTCOMES /MINOR DECISIONS/PATIENT SATISFACTION /MINOR DiCISIONS/PATIENT SELF-EFFICACY /MINOR DECISIONS/HEALTH CARE COSTS /MINOR LIABILITY /MINOR D CISIONS/PATIENT PREFERENCE /MINOR D CISIONS/CONSUMERISM /MINOR /MINOR D CISIONS/CONSUMER INFORMATION /MINOR DECISIONS/PHYSICIAN SUPPORT /MINOR DE\CISIONS/PHYSICIAN AS TEACHER /MINOR DECISIONS/CONSUMER CHOICE /MINOR CONSENT /MINOR DEEISIONS/REIMBURSEMENT /MINOR DECISIONS/CIA MEDICINE /MINOR DEFISIONS/SELF-MANAGEMENT DEbXSXOHS /MAJOR RELATIONSHIP /MAJOR DECISIONS/EFFICIENCY/APPROP /MAJOR DEtiSIONS/HEALTH OUTCOMES /MAJOR DEtiSIONS/PATIENT SATISFACTION /MAJOR SELF-EFFICACY /MAJOR CARE COSTS /MAJOR LIABILITY /MAJOR PREFERENCE /MAJOR CONSENT I /MAJOR INFORMATION /MAJOR DEdiSIONS/SEVERITY /MAJOR DEctiSIONS/PHYSICIAN AS TEACHER /MAJOR DEGISIONS/CONSUMER CHOICE I /MAJOR DEqiSIONS/INFORMED CONSUMERS /MAJOR DECISIONS/REIMBURSEMENT /MAJOR DEdiSIONS/CONSUMERISM /CHRONIC J:rSEASE SELF-Mmm.GEMEN'l' /CHRONIC giSEASE SELF-MGMT/DR-PT RELATION /CHRONIC giSEASE SELF-MGMT/EFFIC/APPROP /CHRONIC DISEASE SELF-MGMT/HLTH OUTCOMES /CHRONIC ti1ISEASE SELF-MGMT/PT SATIS /CHRONIC D 1 ISEASE SELF-MGMT/PT SLF-EFFICAC /CHRONIC D SEASE SELF-MGMT/COSTS /CHRONIC D SEASE SELF-MGMTT/MED LIABILITY 3 6

PAGE 350

Node Location (7 B) (7 9) (7 10) (7 11) (7 12) (7 13) (7 14) (7 15) (7 16) (8) ( 8 1) (8 2} (8 3} (8 4} (8 5} (8 6) (8 9) (8 10) (8 11} (8 12) (8 13) (8 14) (8 15} (8 16) (8 17) (8 18) (8 19) (8 20) (8 21) (8 22) (8 23) (8 24) (9) (9 1) (9 2} (9 3) {9 4) (9 5} (9 6} (9 7) (9 8) ( 9 9) (9 11) (9 12) Node Titl s /CHRONIC SELF-MGMT/PATIENT PREF /CHRONIC piSEASE SELF-MGMT/SOCIAL DETERM /CHRONIC piSEASE SELF-MGMT/DR AS TEACHER /CHRONIC SELF-MGMT/REIMBURSEMENT /CHRONIC !SEASE SELF-MGMT/SEVERITY /CHRONIC !SEASE SELF-MGMT/CONSUMER INFO /CHRONIC DISEASE SELF-MGMT/BETTER DECISNS /CHRONIC DISEASE SELF-MGMT/MAJOR IMPACT /CHRONIC IDISEASE SELF-MGMT/COMMUNITY /OPINiON iC /OPINION OUT IC/DR-PATIENT RELATIONSHIP /OPINION OUT IC/EFFIC/APPROP /OPINION IC/HEALTH OUTCOMES /OPINION IC/PATIENT SATISFACTION /OPINION IC/PATIENT SELF-EFFICACY /OPINION IC/HEALTH CARE COSTS /OPINION ASOUT IC/ACCESS /OPINION IC/CONVENIENCE /OPINION IC/CONFIDENTIALITY /OPINION ABOUT IC/INSUFF KNOWLEDGE IC /OPINION ABOUT IC/PHYSICIAN SUPPORT /OPINION ABOUT IC/CONSUMER INFORMATION /OPINION ABOUT IC/COST-BENEFIT /OPINION ABOUT IC/PHYSICIAN AS TEACHER /OPINION ABOUT IC/IC SERVICE QUALITY /OPINION ABOUT IC/PERFORMANCE FEEDBACK /OPINION ABOUT IC/PATIENT UTILIZATION /OPINION ABOUT IC/PATIENT PREFERENCE /OPINION ABOUT IC/BETTER DECISIONS I /OPINION IC/ROCHESTER QUALITY /OPINION IC/COMPETING SERVICES /OPINION OUT IC/PATIENT ED/AWARENESS / D.IPACT OF IC ON PATiENTS AND PRACTICE /IMPACT ON DR-PT RELATIONS /IMPACT ON ON EFFIC/APPROP /IMPACT ON OUTCOMES /IMPACT ON PATIENT SATISFACTION /IMPACT ON PATIENT SELF-EFFICACY /IMPACT ON COSTS /IMPACT ON MEDICAL LIABILITY /IMPACT ON CONTINUITY OF CARE /IMPACT ON CCESS /IMPACT ON PHYSICIAN AS TEACHER /IMPACT ON PHYSICIAN SUPPORT

PAGE 351

Node Location (9 13) (9 14) (9 15) (9 16) (9 17) (9 18) (9 19) (9 20) (10) (10 1) (10 2) (10 3) (10 4) (10 5) (10 6) (10 7) (10 8) (10 9) (10 10) (10 11) (10 12) (11) (11 1) (11 2) (11 4) (11 5) (11 6) (11 7) (11 8) (11 9) (12) (12 1) (12 2) (12 3) (12 4) (12 5) (12 6) ( 12 8) (12 9) (12 11) (12 12) (12 13) Node Titles I /IMPACT 9N SERVICE QUALITY /IMPACT PATIENT USE /IMPACT N INSUFF KNOWLDGE /IMPACT N PERFORM FEEDBACK /IMPACT N SEVERITY /IMPACT dN PT PREFERENCE /IMPACT dN UNDERSERVED POPS /IMPACT 1N COMMUN-MARKETING I CHALLENGES /CHALLENGES/COST-BENEFIT SUPPORT UTILIZATION SERVICE QUALTIY REIMBURSEMENT /CHALLENGfS/CONSUMER INFORMATION /CHALLENGES/PERFORMANCE FEEDBACK /CHALLENGfS/DOCTOR-PATIENT RELATIONSHIP AND APPROP /CHALLENGES/ROCHESTER QUALITY /CHALLENGES/INSUFFICIENT KNOWLEDGE IC /CHALLENGES/HEALTH CARE COSTS I /ADDRESSCFHALLENGES /ADDRESS qHALLENGES/PHYSICIAN SUPPORT /ADDRESS aHALLENGES/PATIENT ED AWARENESS I /ADDRESS qHALLENGES/COMMUNIC & MARKETING /ADDRESS qHALLENGES/PERFORMANCE FEEDBACK /ADDRESS qHALLENGES/LOCAL FRANCHISE /ADDRESS gHALLENGES/COMMUNITY ORGANIZ /ADDRESS dHALLENGES/IC SERVICE QUALITY /ADDRESS \HALLENGES/CONSUMER INFORMATION /BENEFI:TS OF :IC /BENEFITS b 5F IC/DOCTOR-PATIENT RELATIONS /BENEFITS [F IC/CLINICAL EFFIC AND APPROP /BENEFITS OF IC/HEALTH OUTCOMES /BENEFITS PF IC/HEALTH CARE COSTS /BENEFITS pF IC/CHRONIC DISEASE /BENEFITS 9F IC/ACCESS /BENEFITS bF IC/PERFORMANCE FEEDBACK /BENEFITS :F IC/BETTER DECISIONS /BENEFITS 'F IC/PHYSICIAN SUPPORT 33'8

PAGE 352

Node Location ( 12 14) (12 15) (12 16) (12 17) (13) (13 1) (13 2) (13 3) (13 4) (13 7) (13 8) (13 9) (13 10) (13 11) (13 12) (13 14) (13 15) (13 16) (13 17) (13 18) (14) (14 1) (14 2) (14 3) (14 4) (14 6) (14 8) (14 10) (14 11) (14 13) (14 14) (14 15) (14 16) (14 17) Node Tit es /BENEFITd OF IC/INSURER BENEFIT IC/CONSUMER /BENEFIT OF IC/UNDERSERVED POPULATIONS /BENEFITS OF IC/INSUFFICIENT KNOWLEDGE IC /ACTrvE OLVEMENT /ACTIVE IrvOLVEMENT/DR-PATIENT RELATIONS /ACTIVE IEOLVEMENT /HEALTH OUTCOMES /ACTIVE I OLVEMENT/PATIENT SATISFACTION /ACTIVE I OLVEMENT/PATIENT SELF-EFFICACY /ACTIVE INvOLVEMENT/ACCESS /ACTIVE INvOLVEMENT/PHYSICIAN AS TEACHER /ACTIVE OLVEMENT/SUPPORT GROUPS /ACTIVE I OLVEMENT/C/A MEDICINE /ACTIVE I OLVEMENT/CONSUMER INFORMATION /ACTIVE I OLVEMENT/UNDERSERVED POPLTNS /ACTIVE DISEASE /ACTIVE IVOLVEMENT/BENEFITS /ACl.IVE INVOLVEMENT /REIMBURSEMENT /ACTIVE IiJrfoLVEMENT/BUSINESS OF MEDICINE /ACTIVE I OLVEMENT/PREVENTION /VALUE OF /VALUE OF /VALUE OF /VALUE OF /VALUE OF /VALUE OF /VALUE OF /VALUE OF /VALUE OF /VALUE OF /VALUE OF /VALUE OF /VALUE OF /VALUE OF IC IC/INSUFFICIENT KNOWLEDGE IC IC/PERFORMANCE IC/PATIENT PREFERENCE IC/PATIENT IC/PHYSICIAN SUPPORT IC/COMPETING SERVICES IC/INFORMED-INVOLVED PATIENTS IC/IC SERVICE QUALITY IC/ACCESS C/PATIENT SATISFACTION C/DOCTOR-PATIENT RELATIONSHIP C/HEALTH CARE COSTS C/EVALUATION 3 9

PAGE 353

APPENDIXD VARIABLES WITH MISSING VALVES T

PAGE 354

Independent Age Gender Variable Location of residency Specialty Practice structure Practice visits/week % of patients in capitated plan Minor decisions (B 1A) Major decisions (B 1B) Patients want more involvement (B2A.) Patients should ask more questions Physicians can't involve patient (B2D Patients don't want to make decisions (B2F) Physicians can't answer questions Patients want more information (B2H) Patients need role in decision making Patients need role in asking questions 3B) Patients need role in minor conditions (B3C) Patients need role in seeking info (B3D) I Patients need role in chronic conds (B3E) Physicians' awareness ofTBDS (DI) Physicians' opinion ofTBDS (D2) TBDS to improve quality of care (D3A) TBDS to improve member satisfaction\ (D3B) TBDS to reduce health care costs (D3C) TBDS to respond to business (D3D) \ TBDS to improve patient outcomes (DIE) TBDS to reduce liability claims (D3F) TBDS to be more efficient (D3G) TBDS impact on phone calls (D4A1) TBDS impact on inappropriate visits (R4A3) 34 1% < 1% < 1% <1% <1% 190/o 19% 5% 5% 2% 2% 2% 3% 2% 2% 2% 2% 3% 2% 3% 3% 5% 7% 7% 7% 7% 7% 7% 6% 9% 9%

PAGE 355

Variables with missing values (Cont.) Variable 1 TBDS impact on patient demand (D A4) TBDS impact on liability claims (04 6) TBDS impact on clinical encounter 7) TBDS impact on health outcomes (D B 1) TBDS impact on phys-pt relationship (D4B2) TBDS im act on atient satisfaction 4B4 0/o Missing Cases gofo 9% 9% gofo gofo 9%

PAGE 356

RE :RENCES Barnett AA. Is knowledge really power for patients? Business and Health 1995; May:29-36. Ballard-Reisch, DS. A model of Pfticipative decision making for physician patient interaction. Health ComTicalion 1990;2:91-104. Battista RN, Williams n, MacFarlfe LA Determinants of preventive practices in fee-for-service primary care. Am J Prev Med 1990;6:6-11. Beaver K, Luker KA, Owens RG, Leinster SJ, Degner LF, Sloan JA. Treatment decision making in wo en newly diagnosed with breast cancer. Cancer Nurs 1996;19:8-19. Beckman H & Frankel R. The of physician behavior on the collection of data. Ann Intern Med 1984;101 :6\2-696. Beisecker AE & Beisecker TD. Patient information-seeking behaviors when communicating with doctors. Med Care 1990;28: 19-28. Berkanovic E, Telesky C, Structural and social psychologic factors in the decision to seek medical car1 for symptoms. Med Care 1981; 19:693709. Berwick DM. (1998). As good as il should get: making health care better in the new millenium. Washington, DC.: National Coalition on Health Care Report. Bradley, JH, Zia MJ, Hamilton N.1atient preferences for control in medical decision making: a scenario-based approach. Fam Med 1996;28:496-501. Brody DS. The patient's role in cJcal decision-making. Ann of Inter Med 1980;93 :718-722. Brody DS, Miller SM, Lerman CE, Smith DG, Caputo GC. Patient perception of involvement in medical care: rei tionship to illness attitudes and outcomes. J Gen InternMed 1989;4:506-511.

PAGE 357

Brook RH, Kamberg CJ, Mayer-Oakes A, Beers MH, Raube K, Steiner A Appropriateness of acute medical pare for the elderly: an analysis of the literature. Health Policy 1990; 14:.225-242. Brook RH, & Lohr KN. Will we to ration effective health care? Issues in Science and Technology 1986;l1-10. Brook RH, Lohr KN, Chassin M, Kosecoff J, Fink A, Solomon D. Geographic variations in the use of services: do they have any clinical significance? Health Affairs I 98413:63-73. Cassileth BR., Zupkis RV, Sutton-Smith K, March V. Information and participation preferences among ca!ncer patients. Ann of Inter Med 1980;92:832-836. j Coleman JS, Menzel H, Katz E. dical Innovation: Diffusion of a New Drug. Indianapolis: Bobbs-Merrili, 1966. Cook TD, & Campbell DT. Quas/Experimentation: Design and Analysis Issues for Field Settings. Boston, Houghton Mifflin Company, 1979. Crabtree, BF & Miller WL. A Tjlate Approach to Text Analysis: Developing and Using Codebooks.\. In: Crabtree BF, Miller WL, eds. Doing Qualitative Research. CA: Sage 1992:93-109. Cykert S, Hansen C, Layson R, J. Primaiy care physicians and capitated reimbursement. J Gen Intern Med t 997; 12: 192-194. Davis, DA, Thomson MA, Oxman Haynes RB. Changing physician performance: a systematic review or the effect of continuing medical education strategies. JAMA 1995;274:700-70,. Davison BJ, Degner LF, Morgan TR. Information and decision-making preferences of men with prostate ccilicer. Oncology Nursing Forum 1995;22:1401-1408. l Degner LF & Sloan JA Decision g during serious illness: What role do patients Really want to play? J Epi 1992;45:941-950. 344

PAGE 358

Delbanco TL. Enriching the doct r -patient relationship by inviting the patient's perspective. Ann oflntemMed Dentzer S. The allegory of the Jnveyor belt. Dartmouth Alumni Magazine. 1991; 5:23-25. DiMatteo :MR. The physician-patient relationship: effects on quality of health care. Clin Obstet and Gyn 1994;3 : 149-161. Donaldson, MS, Yordy KD, Lohr KN, Vanselow NA. (Eds.). (1996). Primary Care: America's Health n a New Era. (Institute ofMedicine, Committee on the Future of Care). Washington, DC: National Academy Press. Eddy DM. Connecting value and costs whom do we ask, and what do we ask them. JAMA 1990;264:1737-1739. EddyDM. Rationing by patientc oice. JAMA 1991;265:105-108. Eddy DM. Three battles to watch in the 1990s. JAMA 1993;270:520-526. Eisenberg DM. Advising patients ho seek alternative medical therapies. Ann IntemMed 1991; 127:61-69. \ Eisenberg DM, Davis RB, Ettner JL, Appel S, Wilkey S, Rompay MV, Kessler RC. Trends in alternative use in the United States, 19901997. JAMA 1998;280:1569-1575. Emanuel EJ & Dubler NN. Preservmg the physician-patient relationship in the era of managed care. JAMA 1995;j73:323-329. Ende J, Kazis L, Ash A, Moskowil MA Measuring patients' desire for autonomy: decision making and preferences among medical patients. J Gen Intern 1989;4:23-30. Engel GL. The need for a new med1cal model: a challenge for biomedicine. Science 1977;196:126-136. Equitable Securities Research. Access health, inc.: taking the next step. I Nashville, TN: Seth R. Frank. (Ap "122, 1996). 3 5

PAGE 359

Erlandson DA, Harris EL, Skipp r BL, Allen SD. Doing Naturalistic Inquiry. CA: Sage Publications, 1993. Association Provider Newsletter, anuary/February, 1995; 8:4, 7. Ford L, Kaluzny AD, Sondik E. 1iffusion and adoption of state-of-the-art therapy. Semin Oncol1990;17:48S-494. Fox SA & Stein JA. The effect of hysician-patient communication on mammography utilization by diffe ent ethnic groups. Med Care 1991 ;29: 10651082. Frederikson LG. Exploring info tion-exchange in consultation: the patient's view of performance and outcomek. Patient Educ and Coun 1995;25:237-246. Gilchrist VJ. Key Infonnant In: Crabtree BF, Miller, WL, eds. Doing Qualitative Research. CA: Sage Publications; 1992:70-89. Green LW, Eriksen MP, Schor EL Preventive practices by physicians: behavioral detenninants and pot1eial interventions. Am J Prev Med 1988;4:Sl01-107. Green L, Kreuter M, Deeds S, P "dge K. Health Education Planning: A Diagnostic Approach. CA: Mayfie!d Press; 1980. Greenfield S, Kaplan SH, Ware JE, Yano EM, Frank HJL. Patients' participation in medical care. J Ge. Intern Med 1988;3:448-457. Greer AL. The two cultures ofbioledicine: can there be consensus? JAMA 1987;258:2739-2740. Greer AL. Scientific knowledge an1 social consensus. Controlled Clin Trials 1994;15:431-436. Grilli R, Apolone G, Marsoni S, Ni olucci A, Zola P Liberati A. The impact of patient management guidelines on t 1 e care of breast, colorectal, and ovarian cancer patients in Italy. Med Care 991;29:50-63. 3 6

PAGE 360

Grilli R & Lomas J. Evaluating die message: the relationship between compliance rate and The subject bf a practice guideline. Med Care 1994;32:202-213. Hack TF, Degner LF, Dyck DG. Relationship between preferences for decisional control and illness infotmation among women with breast cancer: a quantitative and qualitative Soc Sci Med 1994;39:279-289. Hampton J, Harrison M, Mitchell J, Prichard J, Seymour C. Relative contributions ofhistory-taking, physician examination, and laboratory investigation to diagnosis and m11 agement of medical outpatients. Br Med J 1975;2:486-489 Herzlinger, R. Market Driven Health Care: Who Wins, Who Loses in the Transformation of America's Service /Jidustry. MA: Addison-Wesley Publishing Company, Inc.; 1997.1 Hill MN, Levine DM, Whelton P11-. Awareness, use and impact of the 1984 Joint National Committee report on high blood pressure. Am J Public Health 1988;78:1190-1194. Jordan HS, Burke JF, Fineberg H, Hanley JA. Diffilsion of innovations in bum care: selected findings. Bums 1983;9:271-279. Keller VF & Carroll JG. A new midel for physician-patient communication. Patient EdJic and Couns !994;23 Kosecoff J, Kanouse DE, Rogers 'f.H. McCloskey L, Winslow CM, Brook RH. Effects of The National InstitUtes ofHealth consensus development I program on physician practice. JAMA 1987;258:2708-2713. Kreps GL, O'Hair D, Clowers M b influences of!nnnan communication on health outcomes. Amer Behav Scien 1994;38:248-256. Kuzel .AJ. Sampling in Qualitative\ Inquiry. In: Crabtree BF, Miller WL, eds. Doing Qualitative Research. CA: Sage Publications; 1992:31-44. Lawrence RS. Diffusion of the US Preventive Services Task Force: recommendations into practice. J en Intern Med 1990;5:99s-103s. 347

PAGE 361

Lennan CE, Brody DS, Caputo OC, Smith DG, Lazaro CG, Wolfson HG. Patients' perceived involvement ih care scale: relationship to attitudes about illness and medical care. J Gen lrltern Med 1990;5:29-33. Lennan C, Daly M, Walsh WP, Jesch N, Say J, Barsevick A, Birenbaum L, Heggan T, Martin G. betweenpatients with breast cancer and I health care proVIders. Cancer 1913;72:2612-2620. Levine PH & Britten AFH. Supek.ised patient-management of hemophilia. I Ann of Intern Med 1973 ;78: 195-!20 1. I Levinson W, Stiles WB, Inui TS, Engle R. Physician frustration in communicating with patients. Med Care 1993;31 :285-295. Lewis CS, Wells KB, Ware J. A L fur predicting the counseling practices of physicians. J Gen Intern Med 1r86;1: 14-19. Lomas J. Words without action? The production, dissemination, and impact of consensus Recommendations. Amk Rev Public Health. 1991;12:41-65. Lomas J, Anderson GM, DomnicJ-Pierre K, Vayda E, Enkin MW, Hannah WJ. Do practice guidelines guide bractice? The effect of a consensus statement on the practice of physicians. NEJM 1989;321: 1306-1311. Lomas J, Enkin M, Hannah WJ, VaydaE, Singer J. Opinion leaders vs. audit and feedback to implement practice guidelines. JAMA 1991:265:2202-2207. Lorig KR, Kraines RG, Brown BS, Richardson N. A workplace health education program that reduces oJtpatient visits. Med Care 1985;23: 1044-1054. Lorig .KR, Mazonson PD, Holman HR. Evidence suggesting that health education for self-management in patients with chronic arthritis has sustained health benefits while reducing healfh care costs. Arthritis and Rheumatism 1993;36:439-446. Lynch WD & Vickery DM. The pptential impact of health promotion on health care utilization: an introduetion to demand management. AJHP 1993;8:87-91. 348

PAGE 362

Mayo PH, Richman J, Harris HWj. Results of a program to reduce admissions for adult asthma. Ann of Inter M'ed 112:864-871. McPhee SJ, & Bird JA. lmplemeAtation of cancer prevention guidelines in clinical practice. J Gen Intern MeU 1990;5: 116s-122s. National Committee for Quality 1surance. Annual member health care survey manual (version 1.0). Washingtol1 DC: Author, 1995. Nease RF & Brooks WB. Patien desire for information and decision making in health care decisions: the autollj my preference index and the health opinion survey. J Gen Intern Med 10:593-600. Neufield KR, Degner LF, Dick Jk. A nursing intervention strategy to foster patient involvement in treatment ecisions. Oncology Nursing Forum 1993;20:631-635. Nunnally JC and Bernstein IH. P, chometric Theory. 3rd Ed. New York: McGraw-Hill, 1994, pp. 264-265L Ong LML, DeHaes JCJM, Hoos Lammes FB. Doctor-patient communication: a review of the li1erature. Soc Sci Med 1995;40:903-918. Osler, SW. Lecture to medical stuaents. Albany Med Ann 1899;20:307. Patrick DL & Wickizer TM. and health. In B. C. Amick, S. Levine, AR Tarlov, & DC Walsh 1 ds.), Society and Health ( pp. 46-92). New York, NY: Oxford University Pre s, 1995. Patton MQ. Qualitative EvaluatiJ\n and Research Methods. (2nd ed.). CA: Sage Publications, Inc., 1990. Peay MY & Peay ER The role of rmmercial sources in the adoption of a new drug. Soc Sci Med 1988;26:1183]1189. Pietrszk J, Ramler M, RennerT, Ford L, Gilbert N. Practical Program Evaluation. CA: Sage Inc., 1990. 349

PAGE 363

Prochaska JO & DiClemente CC. Stages and processes of self-change in smoking: toward an integrative model of change. J Consult and Clin Psych 1983;5:390-395. Reiser SJ. The era of the patient: using the experience of illness in shaping the missions ofhealth care. JAMA 1993;269:1012-1017. Reid AO. Computer managemJ strategies for text data. In: Crabtree BF, Miller WL, eds. Doing Qualitatite Research. CA: Sage Publications; 1992:125-148. I_ Roberts CS, Cox CE, Reintgen Baile WF, Gibertini M. Influence of physician communication on newtr diagnosed breast patients' psychologic adjustment and decision-making. Cancer 1994;74:336-341. Rogers EM. Diffusion oflnnovajons. (4th ed.). NY: The Free Press, 1995a. Rogers EM. Lessons for guidelinJs from the diffusion of innovations. Joint Comm J Qual Imp 1995b;21 :324J328. Ryan B & Gross NC. The of hybrid seed com in two Iowa rural communities. Rural Sociology Sandler G. The importance of the !hlstory in the medical clinic and the cost of unnecessary tests. Am Heart J 1980;100:928-931. Schwartz JS, Lewis CE, Clancy Kinosian MS, Radany MH, Koplan JP. Internists' practices in health prolotion and disease prevention: a survey. Ann /nternMed 1991;114:46-53. Speedling EJ & Rose DN. Building an effective doctor-patient relationship: patient satisfaction to participation. Soc Sci Med 1985 ;21: liS-Starr P. The Social of American Medicine. USA: Basic Books, 1982. Stross JK & Harlan WR. The dissemination of new medical information. JAMA 1979;241 :2622-2624. 350

PAGE 364

Stroll WM, LoB, Charles G. Do patients want to participate in medical decision making? JAMA 1984;25F:2990-2994. Sutherland HJ, LlewellynThomaJ HA, Lockwood GA, Tritchler DL, Till JE. Cancer patients: their desire for inlformation and participation in treatment I decisions. J Royal SocMed 1989;82:260-263. Szaz PS & Hollender MH. A colrution to the philosophy of medicine: The basic model of the doctor-patient telationship. Arch JntemMed 1956;97:585-592. Tabachnick BG & Fidell LS. Using Multivariate Statistics. New York NY: Harber & Row Publishers, 1983. Thompson SC, Pitts JS, L. Preferences for involvement in medical decision-making: situational and demographic influences. Patient Educ andCouns 1993;22:133-140. Vickery DM, & Fries JF. Take Care of Yourself. (5th ed.). Reading MA: Addison-Wesley, 1994. Vickery, DM, Golaszewski TJ, Wljight EC, Kalmer H. A preliminary study on the timeliness of ambulatory care utilization following medical self-care intervention. AJHP 1989;3 :26-31. Vickery DM, Kalmer H, Lowry D, Constantine M, Wright E, Loren L. Effect of a self-care education program on medical visits. JAMA 1983;250:2952-2956. Vickery DM & Lynch WD. Demand management: enabling patients to use I medical care appropriately. JOi 1995;37(5):1-7. von Friederichs-Fitzwater MM, Callahan EJ, Flynn N, Williams J. Relational control in physician-patient Health Communication 1991;3: 17-36. Ware, JE. SF-36 Health Survey Marrual and Interpretation Guide. Boston, MA: The Health Institute, New Enkiand Medical Center Hospitals, 1993. 351

PAGE 365

Wasserman RC, Inue TS, Barria a BS, Carter, WB, Lippincott P. Pediatric clinicians' support for parents m es a difference: an outcome-based analysis of clinician-parent interaction. PJfiatrics 1984;74:1047-53 Wennberg JE & Gittelsohn A. vJnatins in medical care among small areas. Sci Am 1982;246:120-134. White J, Levinson W, Roter D. "?h by the way ... ": the closing moments of the medical visit. J Gen Intern 1994;9:24-28. Wolfe JS & Smith DS. Practical iomedical tests: a guide to the selection of statistical tests. Urology 1996;47: -13.