Citation
Regulation of physicians' clinical decision-making behaviors

Material Information

Title:
Regulation of physicians' clinical decision-making behaviors an evaluation of the impact of federal cost-containment policies and other context factors on physicians' perceptions of autonomy
Creator:
Karshmer, Bernard Aaron
Publication Date:
Language:
English
Physical Description:
xiv, 238 leaves : forms ; 28 cm

Subjects

Subjects / Keywords:
Clinical medicine -- Decision making -- United States ( lcsh )
Medical care -- Cost containment -- United States ( lcsh )
Medical policy -- United States ( lcsh )
Medical logic ( lcsh )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Bibliography:
Includes bibliographical references.
General Note:
Submitted in partial fulfillment of the requirements for the degree, Doctor of Philosophy, Public Administration.
General Note:
School of Public Affairs
Statement of Responsibility:
by Bernard Aaron Karshmer.

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:
28863875 ( OCLC )
ocm28863875
Classification:
LD1190.P86 1993d .K368 ( lcc )

Full Text
REGULATION OF PHYSICIANS' CLINICAL DECISION-MAKING
BEHAVIORS: AN EVALUATION OF THE IMPACT OF FEDERAL
COST-CONTAINMENT POLICIES AND OTHER CONTEXT FACTORS
ON PHYSICIANS' PERCEPTIONS OF AUTONOMY
by
Bernard Aaron Karshmer
B.S., Columbia University, 1963
M.B.A., George Washington University, 1966
A thesis submitted to the
Faculty of the Graduate School of the
University of Colorado at Denver
in partial fulfillment
of the requirements for the degree of
Doctor of Philosophy
Public Administration
1993
lAij


1993 by Bernard Aaron Karshiner
All rights reserved.


This thesis for the Doctor of Philosophy
degree by
Bernard A. Karshmer
has been approved for the
Graduate School of Public Affairs
by
April drt> l *7 *)3
Date
JayiShafritz


Karshmer, Bernard Aaron (Ph.D., Public Administration)
Regulation of Physicians' Clinical Decision-Making
Behaviors: An Evaluation of the Impact of Federal Cost-
Containment Policies and Other Context Factors on
Physicians' Perceptions of Autonomy
Thesis directed by Professor Robert W.Gage
Public policy relative to health care financing and
delivery is changing rapidly. As the portion of the
nation's gross domestic product devoted to the provision
of health services grows, private and public concern for
mechanisms to control costs, enhance quality and increase
access also grow. Not surprisingly, the physician, seen
by many as the gatekeeper to the system, has increasingly
become the focus of a variety of regulatory programs
designed to accomplish these complex objectives.
Policy-makers have paid little attention to the
impact that increased regulation may have on the ways in
which physicians respond to regulatory policy which may
appear to limit their ability to exercise independent
professional judgment (autonomy). This research examines
the factors that led people to choose medicine as a
profession, and analyzes the ways in which certain
factors in the context of health care, especially
regulatory policy, affect their perceptions of autonomy.
Twelve context factors, including regulatory policy,
malpractice considerations, insurance regulations, and
patient expectations were assessed though the use of a
iv


questionnaire which was returned by 335 physicians from a
wide range of specialties and practices in all parts of
the United States.
Analysis of data revealed that autonomy was held to
be a leading concern for physicians in the selection of
their profession. The context factors evaluated were
deemed, to a lesser or greater extent, to have a negative
impact on how practicing physicians evaluated the levels
of autonomy that they felt that they could actually
exercise versus what they thought they would have when
they entered the profession. In spite of the negative
impact of federal regulation, insurance regulations,
etc., on perceptions of autonomy, the physicians surveyed
appear to be largely satisfied with their profession.
Interestingly, the entry expectations and context
factor impacts reported by female physicians
participating in the study appeared to be different from
those of their male counterparts. The scope and type of
research conducted, however, do not allow further
explanation of these findings and only point to the need
to more closely examine the ways in which physicians
respond to regulation.
The form and content of this abstract are approved. I
recommend its publication.
S igned_________________________
Robert W. Gage
v


ACKNOWLEDGEMENTS
At the risk of sounding trivial, predictable and
maudlin I must first acknowledge the numerous
contributions of my best friend (and spouse) Marcia. She
tolerated many years of trial and travail in the doctoral
program and endured the last two years, which can only be
described as intense. Actually, she did more than endure
the last two years, she encouraged me to continue in the
face of missed social engagements and many a night and
weekend with me planted firmly in front of the computer.
Marcia, I couldn't have done this without you.
Our children, Sandy, Hud and Beth, also deserve
recognition. They not only encouraged me to keep going,
they were wise enough not to remind me of my all too
regular pontification about finishing what you start.
Naturally, my mom and late father bear much of the
credit for this effort. They always believed that you
could never have too much education. Finally, Mom, it
can be told I agree. Pressure from my mother-in-law
Dora helped move me along; as she couldn't wait to tell
her friends in Florida that her son-in-law was a doctor.
Special recognition must go to my good friends, and
mentors, Bob Gage and Jay Shafritz. My time with them
was especially rich and exemplifies the best of what
doctoral education is all about. Bob is the consummate
teacher, and through his stewardship of my thesis, helped
me to bring all of those numerous loose ends together.
Jay, always the teacher and always the pragmatist, kept
reminding me that the goal of participation in a Ph.D.
vi


program was a Ph.D.; not just a long-running series of
Quixotic jousts with intellectual windmills.
My "brothers" Art Karshmer and Tim Kay provided high
levels of motivation. Art got me started on this quest
by earning his Ph.D. many years ago. When you're an
identical twin, competition takes on new meaning even
if it takes almost twenty years to catch up. Art's
unbelievable courage and good nature though the trying
times of the second half of 1992 set an example that I
hope I will be able to emulate throughout my life. Tim
was a mere lad when I started this process. You might
say that we "grew up together" and pushed each other to
complete our programs of study. Tim did win our wager;
finishing first. But, how can a thirty-one year old
Ph.D. be as happy as a fifty-two year old Ph.D.?
Special recognition must also go to the other
members of my committee. Sam Overman, Lou Diamond and
John Sbarbaro gave unselfishly of their time to help me
through this most rigorous and rewarding process.
Friends and associates Dave Scott and Peggy Krause
should be awarded medals for putting up with my
daydreaming, intellectual quests on company time, and
anxieties resulting from waiting to hear how the latest
of drafts, too numerous to mention, played downtown.
Three other friends, Will Dupree, Jonijane Paxton
and Greg St. John provided invaluable assistance. Will
is a master survey designer and gave me authoritative
insight into how to "ask the questions." Joni not only
helped to make sure that all "T"s were crossed and "I"s
vii


dotted, she gave candid insights into the clarity, or
lack thereof, of my work. Greg had the unenviable task
of attempting to fine-tune my statistical
"consciousness."
viii


CONTENTS
Tables...........................................xiii
CHAPTER
1. INTRODUCTION ...................................... 1
Overview .......................................... 1
Autonomy in Transition ............................ 5
Public Policy and Professional Autonomy .... 7
Policy Analysis as an Ongoing Process ......... 8
The Interface Between Policy and Health Care
Delivery.................................... 11
Conceptual Underpinnings ......................... 13
Open-System Strategy ......................... 14
Acceptance Theory .......................... 16
Motivation-Hygiene Theory .................... 17
Risk Shifting/Sharing .........................20
Relevance........................................ 22
Purpose of Study and Research Question . . .23
Purpose of Study............................. 23
Research Question ............................ 24
Organization of the Thesis.........................25
Review of the Literature Relative
to Health Care Regulation......................25
Review of Theory and Delineation
of a Model That Attempts to
Explain Changing Health Care
Financing Mechanisms ......................... 26
Collection and Analysis of Data...............26
Propositions ..................................... 27
ix


2. HISTORICAL OVERVIEW OF FEDERAL REGULATION AND
INSURANCE MECHANISMS AS THEY RELATE TO CHANGING
PATTERNS OF PHYSICIAN AUTONOMY..................29
Synopsis......................................... 29
Part I: Health Care Legislation -
An Overview....................................31
Introduction ................................. 31
The Effects of Planning and Regulatory
Mechanisms on Health Facilities and
Services Proliferation ....................... 35
Hospital Cost Containment Through
Reimbursement Regulation ..................... 38
Part II: Risk Shifting/Sharing.....................40
Overview ........................... ..... 41
A Model of Risk-Sharing/Shifting
in Health Care.................................44
Part III: The Role of Insurance Mechanisms
and Government Programs in the Changing Risk
Patterns in Health Care Delivery ............. 49
3. RESEARCH DESIGN....................................54
Introduction ..................................... 54
Propositions . ................................54
Proposition 1<0................................56
Proposition ^ .i...............................63
Proposition 2.o................................66
Proposition 2.i................................68
Proposition 2>2................................70
Variables..........................................72
Type of Research Conducted.........................75
x


4. METHODOLOGY
79
Overview...........................................79
Data Collection................................. . 81
Review of Statistical Methodology ................ 86
Overview.......................................86
Types of Data and the Nature of the Sample .86
Statistical Tests Used ........................... 90
Analysis of Demographic Data and Generalizability
of Data.........................................94
Overview.......................................94
Analysis of Demographic Data...................95
Generalizability of Data .....................112
Summary.......................................... 115
5. RESULTS...........................................116
Overview .........................................116
Open System Strategy..........................117
Acceptance Theory ........................... 118
Motivation-Hygiene Theory ................... 118
Risk Shifting/Sharing ....................... 119
Analysis of Data Relative to Propositions . .120
Proposition 1>0...............................121
Proposition 133
Proposition 2,o .......................141
Proposition 2.i...............................148
Proposition 2>2...............................157
Similarity of Results with Selected Studies . 165
xi


6. CONCLUSIONS, POLICY IMPLICATIONS, AND FUTURE
INVESTIGATION .................................... 169
Conclusions Relative to the Research
Question and Propositions........................ 169
Conceptual Underpinnings in the Setting of
Conclusions...........................170
Conclusions Relative to Erosion of Autonomy 170
Conclusions Relative to Choice of Medicine
as a Career...................................174
Conclusions Relative to Impact of Context
Factors.......................................176
Conclusions Relative to Constraint of Physician
Autonomy.................................... 177
Conclusions Relative to Overall Satisfaction
of Physicians.................................180
Summary of Conclusions........................... 182
Policy Implications of Research General . 182
Policy Implications of Research Autonomy
Specific.................................... 186
Conclusions Relative to Female Physicians . 187
Conclusions Relative to Generalizability
of Findings...................................191
Future Investigation ............................. 192
APPENDIX
A. RISK SHIFTING/SHARING MODELS ..................... 196
B. LIST OF CME ACTIVITIES.............................197
C. QUESTIONNAIRE COVER LETTER....................... 199
D. QUESTIONNAIRE......................................200
BIBLIOGRAPHY ....................................... 213
xii


TABLES
Table
2.1 Selected Bench-Mark Regulatory Programs ... 30
4.1 Classification of Survey Questions ...... 88
4.2 Comparison of Age of Study Group with
American Medical Association Data ............ 97
4.3 Political Party Affiliation of Study Group . 97
4.4 Political Party Affiliation of Study Group;
Re-aggregated ................ 98
4.5 Political Party of Study Group by Sex
of Respondent..................................99
4.6 Year of Graduation from Medical School . .100
4.7 Medical Specialty of Study Group
Compared to American Medical Association
Data..........................................102
4.8 Location of Study Group Practices
Compared to American Medical Association
Data..........................................102
4.9 Types of Practices Reported by Study Group . 103
4.10 Annual Net Income in 1989 ................... 105
4.11 Annual Net Income in 1989 Income Groupings
Listed Cumulatively ... ..................... 106
4.12 Change in Income over the Past Five Years . 109
4.13 Truncated Regional Distribution of Physicians 115
5.1 Mean responses to Questions 23 and 24 . . . .127
5.2 Mean Values of External Factors...............129
5.3 Rankings of Expected and Actual Impact
Mean Values...................................130
5.4 Mean and Ranked Responses to Question 18 . . .134
5.5 Mean Scores to Question 18 by Gender . . * . 137
xiii


5.6 Mean Scores and Ranks of Responses Relative to
General Attitudes Regarding Practice .... 142
5.7 Chi-Square Significance Levels ............. 145
5.8 Pearson's R and Chi-Square Values for Selected
Variables Affecting Negative Impacts on General
Attitudes (Factors) Regarding Practice ... 147
5.9 Context Factor Equivalents ...................151
5.10 Comparison of Rank-Order Results .............155
5.11 Career Preferences for Children ............. 159
5.12 Overall Satisfaction with Medicine .......... 160
5.13 Significance Levels of Crosstabulations . 161
5.14 Income Status Change Over Five-Year Period .163
5.15 Job Satisfaction as a Function of Age . . 164
6.1 Comparison of Rank-Ordered Context Factors .179
xiv


CHAPTER 1
INTRODUCTION
Overview
The health care delivery system is in the midst of
what Aluise (1987) refers to as changes "that will
revolutionize how medical care is provided and financed"
(1). Brown adds:
Notwithstanding the myriad of competitive and
regulatory innovations of the past twenty years, the
U.S. health care system remains astonishingly
laissez-faire. . The 1990s may well mark the
beginning of the end of the laissez-faire
indulgences that still pervade much of the system
(Brown 1992, 31-32).
Many of these changes are either designed to, or
will, affect, the ways in which patient care related
resource consumption decisions (decisions made largely by
physicians) are made.
While physician services account for only 22% of
health care expenditures (Aluise 1987), physicians play a
key, if not preeminent, role in the style and cost of
health care delivery. Approximately 493,000 of the more
than 600,000 physicians in the United States are involved
with direct patient care (Roback, Randolph, and Seidman
1990) and are responsible for the lion's share of the
resource consumption decisions of this industry which
accounts for nearly 12% of the nation's Gross Domestic


Product (Zinberg 1991). A number of pressures
are being brought to bear on this group of medical care
providers as the health care system evolves.
For one, the system is increasingly pluralistic,
with new players exerting new influences on physicians.
According to Starr, the
. . system has begun to slip from their
[physician's] control, as power has moved away from
the organized profession toward complexes of medical
schools and hospitals, financing and regulatory
agencies, health insurance companies, prepaid health
plans, and health care chains, conglomerates,
holding companies and other corporations (Starr
1982, 8).
Corporations and multi-institutional health care
providers vie for the attention and cooperation of physi-
cians. In order to participate, practitioners accept
new, and frequently different, roles and relationships
(Burchell, et al. 1988). These new patterns of affilia-
tion and association increasingly call for physicians to
reorient time-honored professional expectations of
independence and autonomy.
New patterns of organization and reimbursement, such
as Preferred Provider Organizations (PPOs), Health
Maintenance Organizations (HMOs), Diagnosis Related
Groups (DRGs), Resource Based Relative Value Scales
(RBRVS), etc., have attempted to reorient and relocate
the locus of financial risk in the health care system. A
2


significant focus of these efforts has clearly been
physicians, and, according to Grumbach (1988) such activ-
ity "threatens professional values with an unprecedented
administrative and commercial intrusion into the daily
practice of medicine" (210).
Additionally, a clear pattern of feminization is
taking shape in American medicine. This defacto shift,
which is the result of changing medical school admissions
policies, has eventuated in significant changes in the
numbers of women practicing medicine in the United
States. Between 1967 and 1986 the percentage of women
physicians under the age of 35 grew from 8.4% to 25.2%
and enrollment of women in some postgraduate training
programs increased to more than 50% (Moore and Priebe
1991). The American Medical Association reported that
the number of women physicians increased 287.6% between
1970 and 1989 (Roback, Randolph, and Seidman 1990). When
medical school classes achieve gender balance the ratio
of male physicians (84%) to female physicians (16%)
(Roback, Randolph, and Seidman 1990) is likely to change
to a physician work force which is about 50% female.
This transition may bring with it certain changes in
income expectations, autonomy needs, and work patterns.
Analysis by Kletke, Marder, and Silberger (1990), for
example, indicates that female physicians work fewer
3


hours per week, have different specialty interests and
higher retirement rates. These factors may affect the
availability of physician services as the ratio of female
to male physicians shifts.
Many other factors are influencing, or are
influenced by, the changing role of the physician in the
health care system. None-the-less, physicians will
continue to be a key to the changing nature, performance,
and costs of health care.
Thompson (1967) would describe medicine as one of the
technical "cores" of the health care system. The trends
described above tend to affect this core technology, and,
as such, may affect the quality and quantity of the
health care system's output.
These environmental, or contextual, factors are but
a few of the factors impacting physicians. This study is
concerned with met and unmet expectations of physicians
relative their ability to practice medicine in a satisfy-
ingly autonomous manner. Autonomy was selected as a
surrogate measure of the impact of certain contextual
factors on medicine; one of the "core technologies" of
the health care delivery system for:
4


Changes in the ways in which medical care is
provided and in the structure of medical practice in
organized settings have resulted in significant
modification and diminution of medical autonomy
(Astrachan and Astrachan 1989, 1509).
Autonomy in Transition
History and tradition, according to Hardy and Conway
(1978), Starr (1982), and Stoeckle (1988), have long
granted physicians the autonomy to control the processes
of health care delivery. During the past few decades,
however, that autonomy has been challengedand continues
to beas a result of public policy decisions, and other
factors in the context of health care. According to
O'Connor and Lanning :
We are witnessing the end to the remarkable absolute
autonomy that has traditionally been the cornerstone
of the American medical profession (O'Connor and
Lanning 1992, 70).
The effort, in this thesis, is to understand the
effects of policy decisions, and other forces, on the
notion, and reality, of professional autonomy and to
examine the relationship between professional autonomy
and physician satisfaction.
Self-regulated conduct is a hallmark of the
traditional professionslaw, the ministry, dentistry,
and medicine, the profession examined in this thesis
(O'Connor and Lanning 1992). Additionally, the term
"profession" implies that its practitioners possess
special knowledge, which may be rooted in precedent,
5


research (data derived experimentally), and/or recorded
experience (empirical data), and is learned in nature
(Becker 1962). Licensing laws, or similar regulations,
often prevent others from practicing behaviors included
within the scope of the profession in question.
The special knowledge professionals possess is made
available to clients/patients in a manner that matches
the practitioner's best judgment with the needs and
resources of the client. Professional judgment and/or
conclusions, on any given issue, may vary from
practitioner to practitioner, and is often difficult to
evaluate. The complexity and individuality of each
client-professional relationship is such that
practitioners believe their individual professional
judgment (autonomous professional behavior) should not be
casually altered and/or evaluated by others, especially
by those outside the profession (Haug 1989; O'Connor and
Lanning 1992).
This belief system can be called "professional
autonomy." To one extent or another, professionals
believe that the application of their special
knowledge/skills should be largely unencumbered by rules,
regulations, or administrative authority that may
interfere with, or limit, the exercise of professional
6


discretion. (The term "professional discretion" is used
interchangeably with the terms "autonomy" and "clinical
freedom.")
In medicine, the very essence of the practice, from
the physician's point of view, is the exercise of
independent professional judgment. Grumbach and
Bodenheimer equate clinical judgment/autonomy with
"clinical freedom."
Clinical freedom is the ability of the physician to
deliver medical care to a patient without the unin-
vited imposition of outside influences whose purpose
is not the optimal health of the patient. Clinical
freedom allows physicians to fulfill their role as
the patient's agent in performing those services
believed to be beneficial to the patient's well-
being (Grumbach and Bodenheimer 1990, 121).
As the concept is explored more fully, it will
become clear that professional autonomy, or the
restriction thereof, is a critical issue in medicine.
Public Policy and Professional Autonomy
Regulatory policy in health care has increasingly
focused on the physician because, according to Eiseman,
the physician alone is legally and professionally enabled
to dictate how health care services are delivered.
Eiseman and Stahlgren (1987) submit that "It is we
[physicians] who make the ultimate decisions when a
patient becomes ill" (x).
Inevitably, according to Navarro (1989), public
policy that addresses health care delivery issues, in an
7


effort to affect access, quality, and/or cost, must
either limit, or be perceived to limit, the autonomy of
those who make the vast majority of health care
consumption decisionsthe physicians.
If, as Dye (1975) suggests, public policy may be
viewed as "whatever governments choose to do or not to
do," (1) must consider all actions of government,
especially those actions taking authoritative form, as
public policy issues. By definition, "actions" are
designed to accomplish results; thus, public policies are
intended to cause change at any number of levels (i.e.,
behavioral, organizational, economic, etc.).
Policy Analysis as an Ongoing Process
Theoretically, analysis should take place before,
during, and after policy implementation and such analysis
should assess both the causes and consequences of policy
actions.
Public policy can be viewed as a dependent variable,
and we can ask what environmental forces and
political system characteristics operate to shape
the content of policy. Or public policy can be
viewed as an independent variable, and we can ask
what impact public policy has on the environment and
the political system (Dye 1975, 6).
This thesis will look from an historical perspective
at the first question; "What environmental forces and
political system characteristics operate[d] to shape the
content of policy," and from an analytic/ descriptive
8


perspective at the second question: "What impact [does]
public policy [have] on the environment and the political
system" of the health care field. Specifically, this
research examines the impact of public policy, and other
contextual factors such as malpractice and patient atti-
tudes, on professional autonomy.
While it has been suggested that "a commitment to
rational thinking, economic analysis, and scientific
investigation dominate[s] the policy analysis movement,"
(McCurdy 1986, 44) it is not at all clear that the actual
practice of policy analysis adheres to these values.
Lindbloom (1959) rejected the notion of the "rational-
comprehensive" model suggesting rather that most
decisions are "muddled through" on the basis of
"successive limited (incremental) comparisons" which are
less than comprehensive. This "incremental" approach to
policymaking is described by Shafritz (1985) "as
dependent upon small, incremental decisions that tend to
be made in response to short-term political decisions"
and further that "decisionmaking is controlled infinitely
more by events and circumstances. . (315). Perhaps
Etzioni's (1967) "mixed scanning" idea, which combines
incrementalism with the rational, comprehensive approach,
more adequately describes how policy-makers typically
address health care issues.
9


Nonetheless, during the information gathering and
analysis (intelligence) phase of policymaking, certain
dimensions of careful intelligence may be neglected or
disregarded. While the process of formulating policy may
be less than rational, the resultsthe policybecome
regulatory. Therefore, analysis of the potential impact
of policy on closely held medical values, such as
professional autonomy, is an issue that deserves close
attention.
Review of the literature indicates a paucity of
research on the autonomy question. This is especially
interesting in light of the direction in which health
care policy has been evolving: toward the limitation of
the autonomy of physicians. According to de Pouvourville
(1989) "one can predict that physicians will be reluctant
to consider any form of control of their professional
activity, using their expertise and political power to
fight back" (343).
What, if any, attention has been paid to the
professional autonomy issue during the formation or
analysis of public policy? The answer appears to be
"very little."
This research explores the importance of autonomy
and job satisfaction as public policy related issues.
10


Grumbach and Bodenheimer highlight this issue from a
physician's point of view.
As our nation continues to experiment with different
cost containment measures, we believe physicians and
policymakers should carefully consider factors such
as clinical autonomy when evaluating these measures
(Grumbach and Bodenheimer 1990, 125).
The Interface Between Policy and Health Care Delivery
While Federal regulations do not apply to all
patients, the spill-over effect of existing policies can
be seen in private regulatory activities as well. The
insurance industry, for example, has adopted a number of
government practices in an effort to control health care
costs. As such, regulations, or their externalities,
affect many groups of patients, not simply those covered
by Federal policies.
Federal regulatory policy vis-a-vis the financing of
health care (i.e., the Diagnosis Related Group [DRG]
amendments to the Social Security Act) were designed to
alter professional autonomy to a given, large group of
patients. While these limitations may not directly
prescribe diagnostic and therapeutic behaviors, they do
establish mechanisms that force physicians to justify
their professional behaviors; especially as they relate
to hospitalization and discharge of covered patients.
Additionally, purchaser controls (public and private) on
health care are primarily designed to reduce expenditures
11


1
by reducing utilization and/or cost of services. Zinberg
succinctly describes the relationship between quality of
health care and the costs of providing health care.
The resurgence of interest in the quality of health
care stems from increased government spending,
rising costs of medical care, and rapid advances in
medical science. There is agreement that the
current interest in quality is largely finance
driven . (Zinberg 1991, 723).
The term "purchaser" includes individual consumers,
corporations, groups (such as associations), insurance
companies, health plan managers, unions, government, etc.
Perhaps secondarily, such controls are designed to affect
quality and availability of care.
It is conceptually useful to envision physicians as
gatekeepers to the health care system for a variety of
reasons from legal to societal. Thus, they may be seen
as the only providers with the capacity to significantly
control health care expenditures.
The gatekeeper authority of doctors gives them a
strategic position in relation to organizations. In
effect, the profession's authority puts at its
disposal the purchasing power of its patients. From
the standpoint of the solvency of a health insurance
company, the authority to prescribe is the power to
destroy. So, too, the physician's authority to
decide whether and where to hospitalize patients
gives doctors great leverage over hospital policy
(Starr 1982, 26).
Accordingly, to be effective, cost/utilization
control mechanisms (policies) must be designed to guide
and/or motivate physicians with respect to their
purchasing, prescribing, and ordering patterns. They
12


must create an economic environment that encourages
physicians to examine time-honored patterns of behavior.
Conceptual Underpinnings
Several constructs have had a significant impact on
the formulation of this thesis. Four, in particular,
deserve mention here.
1. Open-System Strategy (Concept)
2. Acceptance Theory
3. Risk Shifting/Sharing
4. Motivation-Hygiene Theory
While all four constructs affected the formulation
of this study, only the fourth, Motivation-Hygiene
Theory, was used to partially explain physician responses
to the context factors examined. The first three
constructs are dealt with only in a developmental manner;
that is, they were critical to the formulation of the
research question, the assumptions and the propositions
only. These three constructs, however, will be discussed
in the conclusions drawn from the research.
Open-System Concept, Acceptance Theory, and
Motivation-Hygiene Theory are described in detail in the
literature and will be discussed briefly below. The Risk
Shifting/Sharing concept, however, is not described in
depth in the literature and will be more fully explored
in Chapter 2.
13


Open-System Strategy
Medical care delivery mechanisms are frequently
referred to as a "system." Cleland and King (1983)
define a system as an "organized or complex whole; an
assemblage or combination of things or parts forming a
complex or unitary whole" (17). Shafritz (1985) suggests
that a system is "any organized collection of parts that
is united by prescribed interactions and designed for the
accomplishment of a specific goal or general purpose"
(522). Terms such as "organized," "unitary,"
"prescribed," and "designed" permeate such definitions
and imply a closed environment of measurable and bounded
characteristics. While there is a certain level of
comfort associated with the notion of closed systems, the
reality, according to Thompson (1967), is that many
systems contain "more variables than we can comprehend at
one time, or that some of the variables are subject to
influences we cannot control or predict. ." (6). This
notion of the "open-system," helps focus attention on the
bipartite nature of complex organizations as "indetermi-
nate and faced with uncertainty, but at the same time as
subject to criteria of rationality and hence needing
determinateness and certainty" (Thompson 1967, 10).
The utility of the apparently counter-intuitive
open-system strategy becomes persuasive when coupled with
the realization that open systems themselves are subject
14


to "criteria of rationality." In a sense, this is the
systems equivalent of the Lindbloom's (1959) "rational-
comprehensive" vs. "successive limited iterations" ap-
proaches to policy analysis.
Approaching an understanding of the complex health
care "system" is enhanced by the "open-system strategy."
Analysis of any subsystem enhances understanding of the
entire system as, according to Cleland and King (1983),
all systems (including subsystems) are interdependent.
Medicine is clearly a pivotal subsystem of health care.
Physicians, the practitioners of medicine, play a crucial
role in the delivery of health care. Thus, understanding
the multitude of variables surrounding this group, helps
to reduce the uncertainty associated with the inherently
indeterminate "open" system called health care.
Medicine (and thus physicians) is a dominant segment
of what Thompson (1967) would refer to as the "core
technologies" of health care. Such core technologies are
critical to the success of open systems and must be
understood, and adjusted, to the extent possible, to
assure achievement of system goals. This analysis is
designed to enhance understanding of this core
technology, and in so doing to contribute to the
information available to policy-makers and analysts
15


concerned with input (resource) and output (services)
questions.
Acceptance Theory
The Acceptance Theory of Authority posits that the
validity of authority is a function of its acceptance by
those it affects. Simon (1947) referred to this
phenomenon as the "Zone of Acceptance," which he
described as the range of decisions made by others which
a person will accept before "disobedience will follow"
(12). Barnard, using the term "indifference" rather than
acceptance, suggested that:
If all the actions reasonably practicable be
arranged in the order of their acceptability to the
person affected, it may be conceived that there are
a number which are clearly unacceptable, that is
which certainly will not be obeyed; there is another
group somewhat more or less on the neutral line,
that is, either barely acceptable or barely
unacceptable. The last group lies within the "zone
of indifference." The person affected will accept
orders lying within this zone and is relatively
indifferent as to what the order is so far as the
question of authority is concerned (Barnard 1938,
168-169).
This construct provides a useful conceptual frame-
work for examining the response of physicians to
regulations (expressions of authority) because such
expressions may force physicians from the "Zones" of
"Indifference" and "Acceptance" into a Zone of
Nonacceptance.
The disposition of physicians toward regulatory
policies which appear to limit their autonomy is a
16


critical public policy question, especially to the extent
that such policies may cause physicians to become less
accepting of regulation or contextual factors.
Colombotos and Kirchner (1986) suggest that the effects
of public policy on the satisfaction or dissatisfaction
of physicians with their work is an important area of
exploration for the policy analyst since the physician is
the pivotal professional in the health care delivery
system. Fuchs notes that it is
impossible to understand the problem of medical care
without understanding the physician, and it is
impossible to make significant changes in the
medical field without changing physician behavior
(Fuchs 1984, 1572).
Colombotos and Kirchner add:
what physicians think and what they do make a
difference in how the system works and how it
evolves. To the extent that physicians' attitudes
are linked to their practice behavior, they may
spell the difference between success and failure in
the implementation of health care programs
(Colombotos and Kirchner 1986, 195).
Motivation-Hvaiene Theory
Herzberg, Mausner, and Snyderman (1959), in their
seminal work dealing with the Motivator-Hygiene Concept,
suggested that job factors were not necessarily
bi-polarthat is that the opposite of a dissatisfier is
not necessarily a motivator. While low pay may, for ex-
ample, cause dissatisfaction, higher pay will not auto-
matically result in satisfaction. Additionally, job con-
17


tent factors (motivators) appear to contribute to satis-
faction while elements in the context (hygiene factors)
such as external regulations, seem to contribute to dis-
satisfaction (Herzberg, Mausner, and Snyderman 1959).
Medicine is generally seen to be rich in job content
factors (i.e., achievement, recognition, and responsibil-
ity) which partially explains the appeal of medicine to
highly motivated, self-directed, focused people. This
property of medicine may also explain the apparent para-
dox in which physicians can find themselves highly satis-
fied with their profession in general and very critical
of many of its elements. Context, or hygiene factors,
tend to constrict autonomya professional imperative
that is an intrinsic value of medicine. It is assumed
that negative attitudes about context factors are partial
dissatisfiers. Reversal or elimination of these
dissatisfiers, however, may or may not enhance total sat-
isfaction with medicine.
Twelve (12) context factors were selected for analy-
sis:
A. Malpractice
B. Public view of the profession
C. Patient expectations
D. Role of non-physician health care providers
E. Requirements of insurance companies and other
third party payers
F. Government involvement in the delivery and fi-
nancing of health care
G. Competition in medicine
H. Quality assurance mechanisms
I. Alternative delivery systems
18


J. Aging of the population
K. Specialization in medicine
L. Pace of developments in medicine
Inferential and anecdotal data seem to indicate a
change in the satisfaction level of physicians. Policy
issues are certainly implicated in this regard and are a
major focus of this research. None the less, other
factors in the context of health care were also selected
for analysis. Some of these other factors are directly
related to public policy (i.e., requirements of
insurance, quality assurance mechanisms), while others
(i.e., malpractice, patient expectations, etc.) are not.
In short, it is clear that regulatory policy alone does
not adequately explain how physicians feel about the
important dimensions of their profession. The eleven
other factors were selected in an effort to distinguish
the relative impact of regulatory policies on physicians.
The examination of all these factors as they affect
physician satisfaction is the goal of this research.
The changing nature of medical care financing and
organization makes it essential that the level of
satisfaction of physicians with their work, and the many
factors that affect the way they do their jobs, be
assessed. Lichtenstein's research showed that the
correlation between job satisfaction and performance is
positive.
19


. . low physician job satisfaction may pose
serious costs to organized health delivery settings
in the form of low morale, high turnover, and,
perhaps, low productivity and quality of care
(Lichtenstein 1984b, 56).
The importance of the relationship between autonomy
and satisfaction was established by Schulz, Girard, and
Scheckler in their analysis of Dane County, Wisconsin,
physicians.
Perceived clinical freedom [autonomy] was a strong
and positive predictor of job satisfaction,
regardless of the dimension of satisfaction being
predicted (Shulz, Girard, and Scheckler 1992, 302).
Schulz, Girard and Scheckler (1992) add that
physician satisfaction is "not only a goal in itself, but
for the recruitment of the best persons to the field"
(304), and suggest that there is evidence that
physician satisfaction may also affect patient
satisfaction.
An improved understanding of the factors that affect
physicians' perceptions of the exercise of professional
autonomy should contribute to the quality of considera-
tion given during the formation and evaluation phases of
public policy development.
Risk Shiftinq/Sharina
Starr (1982) suggests that organizations, such as
hospitals and health insurers, generally have greater
control over the behavior of physicians, and thus the
entire system of production, by employing these providers
20


rather than allowing them to function independently
outside the organization. The history of the health care
system in the United States, however, reflects the exist-
ence of physicians as independent practitioners.
Hospitals and insurers generally allowed physicians
to remain independent entrepreneurs, though there
can be little doubt that leaving doctors outside the
organizational structure of these institutions
increased costs of medical care. ... As
independent entrepreneurs, doctors are unlikely to
be sensitive to any organizational interest in
conserving resources. . But doctors were able to
block this type of control, and the hospitals and
insurers instead developed financial arrangements
that allowed them to pass through the higher costs
that professional autonomy produces (Starr 1982,
26) .
Thus, one might conclude that physicians have been
insulated from the financial sequelae of their actions.
In a very real sense, doctors have used the professional
autonomy argument as a mechanism for remaining
independent of needs of the system to conserve and/or
better use resources. This may have been tolerable in an
era of apparently "unlimited" resources. In an
environment of limited means, however, detachment from
the financial consequences of medical care consumption
decisions may not make sense from an economic point-of-
view.
Financial risk is a fact of life in virtually all
economic systems. The ways in which such risk is posi-
tioned and/or distributed, however, may vary considerably
from situation to situation. The scope of such variation
21


in health care financing is described in some detail in
Chapter 2. In short, the importance of this construct to
understanding physician responses to changes in
regulatory policy, and other context factors, lies at the
heart of this research. The physician has become the
focal point of efforts to realign the health care
system's financing mechanisms in an effort to conserve
resources and, secondarily, to improve quality. This
attention has placed the physician in a position of risk,
a position that appears to have had a negative effect on
how medical practitioners view their own satisfaction and
effectiveness.
Relevance
Evolving regulatory policy, community values, and a
host of factors in the context of health care, have had
fundamental impacts on the delivery and financing of
health care. Beginning with the passage of the Health
Maintenance Organization Act of 1973. the physician has
become the focus (directly and indirectly) of some of the
more important of these policy initiatives designed to
influence the cost of, access to, and quality of health
care. Such focus on the physician is significant, as it
reflects a growing belief that this profession is the
linch-pin in the control of costs, quality, and access in
an industry that accounts for approximately 11.4 percent
22


of the gross national product (Zinberg 1991). The inter-
est in the physician, from a regulatory point of view,
stems from this professional's unusual role in control-
ling expenditure dynamics in the health care system.
The market for physicians' services does not satisfy
the conditions that define a reasonably competitive
market. First, widespread health insurance coverage
reduces patients' sensitivity to fees. Patients
generally choose physicians and services with little
concern for price. Second, patients often lack
adequate knowledge on which to base their choices of
medical services and judgments about technical
quality; they typically rely on their physicians for
advice in making medical decisions. Physicians are
often not subject to the checks and balances gener-
ated by traditional competitive forces. Finally,
legal restrictions specify who can provide medical
services, admit patients to hospitals, and prescribe
drugs. Although such restrictions protect patients
from unqualified providers, they also tend to grant
monopoly power to the medical profession (Hsiao et
al. 1988, 835).
Purpose of Study and the Research Question
Purpose of Study
Because of the central role of the physician in the
control of the medical care sub-system, these profession-
als have become the focal point of a complex set of
regulatory mechanisms; mostly governmental. In an
effort to shed light on the importance of policy changes
vis-a-vis the physician, this thesis addresses the
following issues.
1. To highlight professional autonomy as a
significant public policy issue.
2. To identify and evaluate the factors that had an
effect upon the selection of medicine as a
career.
23


3. To identify and evaluate factors in the context
of health care that affect physicians'
perceptions of autonomy.
4. To identify areas of importance to policy makers
concerned with the impact of their regulatory
decisions on physicians.
5. To suggest areas for further study.
The study intends to describe physicians' existing
values and the effect of public policy, and other
factors, on these values; primary emphasis is placed on
physicians' perceptions of autonomy.
Research Question
What is the relationship between public regulatory
policy, and other factors in the context of medical care,
and the perception of physicians regarding their
professional autonomy?
The time-honored tradition of physician autonomy
mandates that attention be paid to the response of physi-
cians to policies that may have a negative impact on this
valued dimension. Thus, the disposition of physicians
toward the practice of medicine in general, and the
acceptance of regulatory activities in particular, is
crucial to the potential success of programs designed to
reform the delivery and financing of health care. The
results of the analysis stemming from this question,
then, are descriptive and of value to the extent that
24


enhanced understanding of these factors may have an
impact on the formulation of public policy.
Organization of the Thesis
For the sake of description, this dissertation is
organized into three segments. The first reviews the
literature relative to health care regulation; the second
discusses theory and delineates a model of health care
financing which helps to explain the changing environment
of health care delivery and regulation; and the third
addresses the collection and analysis of data relative to
the research question.
Review of Literature Relative
to Heath Care Regulation
A necessary prerequisite to the analysis of the ef-
fect of regulation and context factors on physicians'
perceptions of autonomy is the description of major
changes in the health care delivery and financing
systems. The implementation of the Hospital Survey and
Construction Act (1946) was selected as a starting point
for this analysis as this Act marks a significant post-
World War II entry of the government into health care
delivery and financing. A special emphasis was placed on
the dynamics of health care financing following the
passage of the Medicare (Title XVIII) (1965) and Medicaid
(Title XIX) (1965) amendments to the Social Security Act,
and the Health Maintenance Organization Act of 1973.
25


These policy decisions set the stage, from a regulatory
perspective, for understanding the Federal government's
impact on the health care system in general and
physicians in particular.
Review of Theory and Delineation
of a Model That Attempts to
Explain Changing Health Care
Financing Mechanisms
The delineation of a model that describes
alterations in risk shifting and sharing in health care
financing provides a backdrop for understanding the
pressures that may affect the attitudes of physicians.
This model attempts to describe changes in the mechanisms
employed to finance the delivery of health care, with
special emphasis on how these changes have affected
physician autonomy. Of special interest is the
Diagnostic Related Group (DRG) health care financing
program. This policy may be seen as a philosophical
antecedent of the direction in which federal health care
financing policy, and eventually private financing, will
take to attempt to make physicians conscious of
cause-and-effect relationships in the provision of health
services.
Collection and Analysis of Data
Through the use of an original questionnaire, a
multi-specialty, nationally-distributed group of
physicians was studied. The purpose of this survey was
26


to characterize physicians' attitudes on certain widely
held professional values. Emphasis was placed on the
ways these values have been affected by Federal
regulatory activities and community and context factors
in an effort to identify those professional issues of
greatest importance to physicians as a result of a
changing regulatory and community environment.
Propositions
Two propositions (observed uniformities), and three
subordinate propositions, have been formulated. These
propositions purport that professional autonomy is a
critical component of the practice of medicine, and that
attacks on this value may have a negative impact on
physicians' perceptions of satisfaction and their will-
ingness to accept changes in the delivery system mandated
by public policy.
P1>0 Autonomy is a multi-faceted property that many
physicians believe has been eroded by a variety of
factors in the context of health care. That is,
they believe that the level of personal and work-
related autonomy that they actually have is somewhat
less than they expected to have as a practicing
physician when they entered medicine.
Pl.l Certain values (i.e., service, prestige, challenge,
autonomy, etc.) were critical determinants in the
physician1s choice of medicine as a profession.
P2.0 Several factors in the context of health care
(hygiene factors according to Eerzberg) have a
disquieting effect on physicians' attitudes about
their profession.
27


p2.1 Some factors in the context of health care are
perceived by physicians to be "autonomy
constraining*1 in nature. Forces that appear to
constrain the professional autonomy of physicians
have a negative effect on the vay in vhich
physicians perceive various dimensions of their
professional lives.
P2<2 The gestalt of medicine is, in and of itself, a
powerful "satisfier." Physicians can, on one hand,
see themselves as "satisfied" professionals, while
on the other hand they can also find aspects of the
profession less than satisfactory.
These propositions form the basis of the research
reported herein. The research design and implementation
are predicated on these propositions.
28


CHAPTER 2
HISTORICAL OVERVIEW OF FEDERAL REGULATION AND
INSURANCE MECHANISMS AS THEY RELATE TO CHANGING
PATTERNS OF PHYSICIAN AUTONOMY
Synopsis
Federal health policy has evolved in significant
ways. A review of Federal policies implemented during
the five decades since the end of World War II is
instructive, as it helps to set forth how Federal health
policy has evolved. During this period, other changes
have also occurred in technology, the economy, and social
values. Clearly, the emphasis of Federal policy and
changes in these factors are inextricably linked.
This chapter briefly describes the ways in which
Federal health policy has evolved and examines the forces
that have influenced these changes. The objective is to
indicate how policy emphasis has reached the point where
the physician has become a primary focus of attention.
This examination assumes that any attempt to make the
physician the focus of regulatory activity designed to
alter the ways in which health care is financed and
delivered must ultimately target patterns of medical
decision-making. It is held that to do so is to confront
the professional autonomy of the physician(s) whose
consumption decisions are seen as central to the issue of


cost-containment. Such a review is vital as this
research attempts to describe how evolving factors, such
as public policy and community values, have affected the
practice of medicine and the way physicians feel about
their professional autonomy.
Several Federal health care policies can be
identified as reflective of a changing regulatory
environment in health care delivery. While policies, too
numerous to mention, could be highlighted, seven programs
have been selected as reflective of the evolution of
Federal concern with health care delivery and financing.
Briefly, these bench-marks of evolving health care
policy, and their foci, can be summarized in the
following way.
Table 2.1 Selected Bench-Mark Regulatory Programs
1. Hospital Survey and Construction Act 1946
Focus: Hospital and long-term care beds, and
public health facilities. (Also called the
Hill-Burton Program)
2. Regional Medical Programs Act 1964
Focus: Redistribution/regionalization of
specialized services to hospitals and medical
schools. (Also called the Heart, Cancer, and
Stroke Program)
3. Title XVIII (Medicare) and Title XIX (Medicaid)
Amendments to the Social Security Act 1965
Focus: Direct payments to providers (hospitals
and physicians) guided by traditional insurance
mechanisms for eligible patients.
30


4. Comprehensive Health Planning Act 1966
Focus: Voluntary community planning for
facilities and services.
5. Health Maintenance Organization Act 1973
Focus: Attempt to gain voluntary acceptance of
altered patterns of risk shifting/sharing.
6. National Health Planning and Resources Develop-
ment Act 1974
Focus: "Public Utilities" approach to
regulation of facilities and services
proliferation; also called for "Certification
of Need" for the addition of new facilities
and/or services. The goal of the program was
the prevention of excess capacity.
7. Social Security Amendments Section 603 1983
Focus: Prospective Payment System (Diagnosis
Related Group financing) mandated changes in
risk shifting/sharing patterns for hospitals.
While hospitals were put in an "at-risk"
position vis-a-vis reimbursement, the admitting
physicians were targeted by the hospitals to
control costs.
This chapter presents a review of the literature
that delineates the policy transformation summarized in
Table 2.1, explores ways in which risk has been altered
as the result of these policies (Part II), and also
examines how risk factors have been affected by
government and insurance industry programs (Part III).
Part I: Health Care Legislation An Overview
Introduction
Early federal involvement in health care came in the
form of large-scale facility construction immediately
after World War II. To address the rather glaring, and
31


easily defined, needs for hospitals, the Hospital Survey
and Construction Act of 1946 (Hill-Burton) was passed in
an effort to get construction money into areas that were
undeserved, or isolated medically and/or geographically.
Large amounts of money were made available through out-
right grants, loans, and loan guarantees to construct
facilities that would bring hospital-based health care
services to all parts of the nation. Between 1946 and
1974, 400,000 beds and related facilities were
constructed with Hill-Burton assistance. In 1974, the
Hill-Burton Program was phased-out with the passage of
the National Health Planning and Resources Development
Act fP.L. 93-641).
The hospitals built during the Hill-Burton period
were independent, free-standing, and highly variable in
terms of scope and quality of services provided. Under
the Hill-Burton Program the primary purpose of planning
was to allocate funds equitably among states through the
use of formulae that were based simply on population and
per capita income (Lightle 1978). The absence of inter-
institutional planning and cooperation seemed of little
consequence at a time when third-party indemnification
was exponentiating and public confidence in the value of
health care was at an all time high. Raising funds for
hospital construction was a relatively simple matter and
32


each hospital became adept at developing, maintaining,
and satisfying capital needs through often fiercely loyal
constituencies. Hospital constituencies adopted the
attitude that their facility had to be the "very best"
and that nothing could interfere with its prosperity and
comprehensiveness. As a result, an industry developed
that has been characterized by Starr (1982), and others,
as an unlinked cottage industry.
The "cottage industry" nature of the hospital
segment of the health care system is perhaps best
reflected by data that indicate that in 1975 (these data
have not changed appreciably to date) the United States
had approximately 1,500,000 beds, with the vast majority
of these beds in hospitals having fewer than 200 beds
each. Nuveen relates that most of the hospitals included
in this group were nonprofit, being either church or
community based. Fewer than 2,000 of the total were
operated by units of state, county, or local government
(John Nuveen and Company 1975). These data describe a
large inventory of beds distributed across a significant
number of inefficiently sized, independent, largely
unlinked, and, until the 1960s, unregulated sponsors.
As technology and costs grew, and inter-city dis-
tances shrank, this mentality became obsolete. Nonethe-
less, the nation was left with a fragmented, highly du-
plicated, frequently inefficient, and unlinked system for
33


the delivery of hospital-based services. Perhaps more
important, the mentality of fierce chauvinistic
competition that had taken root was the most destructive
legacy of this "every man for himself" approach to
uncontrolled and unplanned development. The fires of
this approach were fueled by an expanding system of
third-party "cost-based" reimbursement. This
reimbursement methodology was fueled by government and
private insurance mechanisms, and according to Califano
(1988) and Studnicki (1983) virtually guaranteed the
unchallenged payment of all bills. Risk was successfully
shifted from the individual patient to impersonal,
distant third parties. The increased costs associated
with these third parties were barely visible to patients
and were predicated on the basis of incurred costs rather
than value.
If a hospital is reimbursed for whatever it spends
to provide services, plus a negotiated surplus
beyond that level, the basic financial incentive
will be to maximize costs. For hospital executives
in the era of negotiated surplus, the "philosophy of
abundance" (i.e., more is better) was also good
management (Studnicki 1983, 101).
This type of "removed risk" reimbursement philosophy
made it possible for so-called "community hospitals" to
define their communities in virtually any way they saw
fit. Thus, needs assessments that resulted in the
expansion of hospital facilities and services could rely
34


upon inexact or biased perceptions that frequently
considered the hospital's image as a symbol to special
constituencies and/or the medical staff.
The Effects of Planning and Regulatory
Mechanisms on Health Facilities and
Services Proliferation
The entry of the Federal government into the health
care reimbursement picture during the 1960s had a
significant effect on hospital finances. Medicare
(social insurance for the elderly) and Medicaid (social
welfare reimbursement for the categorically eligible)
greatly expanded the scope of reliable third-party
payment sources for most hospitals in the United States.
It has been suggested that with the advent of these
programs the responsibility for hospital receivables
shifted dramatically away from individuals, with
approximately 80-90 percent of hospital revenues coming
from third-party sources (John Nuveen and Company 1975).
Planning Mechanisms. The late 1960s was also the
legislative turning point for health planning in the
United States. P.L. 89-749 established the Comprehensive
Health Planning System. The "Heart, Cancer, and Stroke"
program (Regional Medical Program Act 1964) provided
impetus for decentralization of sophisticated facilities
and services into community settings that previously had
looked to tertiary care centers for such services.
Additionally, significant changes in the original Hill-
35


Burton Program were all intended to raise the horizons of
health facilities planning from "bricks and mortar" to
community need.
Lightle (1978) argues, however, that the provisions
of the Medicare and Medicaid reimbursement systems them-
selves eliminated the need for construction grant pro-
grams such as Hill-Burton by including allowances in
reimbursement rates for depreciation to cover future
capital needs. The synthesis of these and other programs
occurred in 1974 with the passage of The National Health
Planning and Resources and Development Act. P.L. 93-641.
In a very real sense, however, an unforeseen externality
of the Medicare and Medicaid systems doomed the health
planning initiatives to failure.
The provisions of P.L. 93-641 called for a
fundamental change in planning approach, from facilities-
based planning to population-based planning.
Legislatively, for the first time, the Federal shorthand
called for analysis of what consumers needed rather than
what hospitals and other providers wanted to provide.
This refocusing of attention attempted to force hospitals
to redefine their community of interest and raise their
sights beyond what had previously been expected. In a
sense, P.L. 93-641 asked health care institutions to be
regional rather than neighborhood facilities. While the
36


intent was to refocus hospitals' "area-of-interest
constructs," nothing was done to realign the risk system
that removed institutional decision-makers from the
realities of the competitive world.
P.L. 93-641 mandated universal application of
"certification of need" for a wide range of capital and
highly skilled, labor-intensive projects. The theory
behind certification of need was that change for the sake
of change, or community pride, was unacceptable if the
value to the public interest could not be demonstrated.
Financial feasibility, in an artificially risk-free
context, became the battle cry of the seventies.
Hospitals that previously had been essentially isolated
from economic realities faced by other purveyors were
suddenly confronted with demonstrating the relationship
between dollar-intensive projects and the needs of the
people they served.
While it may seem that these factors would have a
depressing effect on ambitions for expanding hospital
service and facilities, they appear to have had the oppo-
site effect. In spite of the fact that rational planning
and the public interest were purported to be paramount,
legislative pressure for increased access and improved
quality resulted in the perceived need for new and ex-
panded facilities.
37


Rate Regulation. Despite the pressures described,
some of which may have had an idiopathic effect upon
health care facilities and service proliferation, only
minimal effort was given to implementing policy strate-
gies intended to regulate hospital operating and capital
expenditures. While the National Health Planning and
Resources Development Act fP.L. 93-641V did speak of
procedures to "de-certify" existing unnecessary, redun-
dant, and expensive services in an effort to contain
costs, the rules and regulations regarding this charge
were never implemented. Certain rate review experimenta-
tion specified in the law suffered a similar fate.
P.L. 93-641. for example, called for the issuance of
"standards respecting the appropriate supply, distribu-
tion, and organization of health services (National
Health Planning and Resources Development Act 1983)." As
one might expect, these standards represented the back-
bone of an act designed to implement a planning system
based on "population-based" criteria rather than "facili-
ties-based" planning.
Hospital Cost Containment Through
Reimbursement Regulation
Health economists, planners, and administrators
speculate that the health care system, as we know it, is
a reflection of the way in which institutional health
care services and facilities have been financed. This
38


argument suggests that the only effective way to control
institutional costs and growth is to alter how services
provided through these outlets are financed. Thus, if
significant changes are to be effected in the delivery
system, the financing system must be redirected. Recent
Federal health care initiatives attempt to address this
hypothesis.
The 1982 TEFRA amendments to the Social Security Act
called for a rather significant change in the way hospi-
tal services were reimbursed for patients covered under
Medicare and Medicaid. These amendments altered the bas-
ic cost framework for federal payments to hospitals, set-
ting "target rate increases" indexed to overall increases
in the Medical Care Consumer Price Index.
Amendments to the Social Security Act passed and
signed into law in April of 1983 (P.L. 98-21) carried the
TEFRA provisions still further and specified prospective
payments to hospitals on the basis of Diagnosis Related
Groups (DRGs). The DRG notion, which was tested in New
Jersey, established urban and rural area rates for
Federal payments covering a wide range of
hospitalizations for many common diagnoses. The Federal
government established, in advance, the "value" of an
admission with all necessary diagnostic and therapeutic
procedures. The DRG program, whether successful or not,
39


marked a significant point in government involvement in
health care financing.
Part II: Risk Shiftina/Sharina
The preceding discussion delineated the evolution of
key health care policy pieces. In this section, a
"locus-of-risk" construct is used to depict how risk
shifting/sharing patterns have changed and suggests that
modification in risk shifting/sharing patterns has put
the physician in a position of having to justify and
rationalize medical care decisions. This pressure is
seen by doctors as a challenge to their professional
autonomy.
From a conceptual point-of-view, this model attempts
to furnish perspective to the ways in which changes in
health care financing have altered the risk-bearing
position of the performers in the health care arena. The
three mechanisms described are neither comprehensive nor
all-inclusive. They are presented, however, as
representations of important philosophical positions
which have affected the ways in which health care has
been financed and, consequently, delivered. From the
perspective of physician autonomy, this model depicts a
transition from virtually total professional autonomy in
Model #1 to significantly constrained autonomy in Model
#3. The third paradigm suggests an environment in which
regulations and rules promulgated by parties outside of
40


the medical community delimit the range of decision-
making available to practitioners.
Overview
The increased scope of involvement of third-party
payers (insurance companies and the Federal government)
has altered the at-risk positions of those who provide
health services, those who pay for health services, and
those who receive services. Other pressures, most
notably torts, rapidly expanding technology, and a
blurred picture of causal relationships between cost and
out-of-pocket expenditures, have contributed to
significant increases in health care expenditures.
"Competition" and risk shifting/sharing have become
rallying cries for those who believe that the means of
slowing down America's rising health care expenditures
lies in a reorientation of how health care services are
financed. Stockman suggests there is a need to develop
a:
marketplace system in which health care providers
would invest only in those facilities and equipment
that would contribute to their ability to enhance
patient care at reasonable costs (Stockman 1981, 9).
This notion suggests that a "marketplace system" does
not exist in the health care field today and that should
one be in place, provider consumption decisions would
change on the basis of altered perceptions of self-
41


interest. Two fundamental characteristics of a market-
place system are competition and risk (risk/reward).
The participation of government in health care
regulation has frequently been stimulated by concerns
(sometimes premature) about the availability and cost of
health care. Freeland, Catlat, and Schendler (1980), in
their projections of national health care expenditures
for 1980, 1985, and 1990, suggested historical trends "in
which health care expenditures have approximately doubled
every six years." They attributed this to two factors:
(1) The role of third-party payments in increasing
consumer demand for services; and (2) the associated
fee-for-service and cost-based reimbursement systems
which lack incentives to provide medical care in the
least expensive manner (Freeland, Catlat, and
Schendler 1980, 17).
The concern of the Federal government is expressed
in its efforts to control health care expenditures by
implementing "marketplace"-inducing policies, such as DRG
(Diagnostic Related Group) financing of inpatient serv-
ices, and cost-adjusting policies such as Relative Value
Schedule (RVS) payments to physicians (U.S. Department of
Health and Human Services 1987). A Relative Value
Schedule system of payment suggests that physician
services can be evaluated on the basis of benchmark
values that recognize such factors as intensity, time,
training, etc., rather than historical pricing
structures. These two policies serve as examples of an
42


intent to shift the emphasis of medical care
reimbursement. The preoccupation with introduction of
risk/reward strategies (such as DRGs) can be explained by
the fact that approximately fifty-four cents of each
health care dollar is spent on services rendered by
institutional providers (Grimaldi and Micheletti 1982).
Since hospital-based care is the single largest
category of health care expenditures, these costs have
become the focus of a great deal of national attention
directed at reducing total health care expenditures. The
"architecture" of this institutional segment of the
health care delivery system may be seen as the product of
policy initiatives in financing that encouraged the
proliferation of facilities and services.
The implementation of DRG-based prospective
reimbursement is, perhaps, the most significant effort by
the Federal government to compel changes in traditional
risk patterns in the financing of health care. While the
Health Maintenance Organization Act of 1973 attempted to
induce a voluntary response to risk shifting/sharing on
the part of physicians and hospitals, the DRG Amendments
to the Social Security Act were the first major
regulatory enactments designed to accomplish this change
in the responsibility patterns of physicians. The
prospective payment methodology places increased
importance on risk shifting as a cost containment
43


modality. While risk shifting itself is not a new idea,
its use to control health care costs has been limited.
The strategy, however, is gaining in acceptance from a
public policy point of view since many see the locus-of-
risk as a critical variable in the national cost
containment equation.
So complex and controversial is the role of the
financing system that some believe the structure of
health care delivery is little more than a reflection of
the financing systems. Others believe the financing
mechanisms simply mirror the structural elements of the
delivery subsets. Assuredly, this classic "chicken and
egg" argument holds interest for those inclined toward
such arguments. However, the fundamental causation
question, which addresses how different means of
financing may affect the delivery of health care, is more
pertinent to developing an understanding of the forces
that drive the multitude of delivery models embedded in
the health care delivery system.
A Model of Risk-Sharinq/Shiftina
in Health Care
Financial risk is a fact of life in virtually all
economic systems. The way in which such risk is
addressed, however, may vary considerably from situation
to situation. "Locus-of-Risk" models (See: Appendix A)
portray certain basic differences in the ways the issue
44


of risk have been addressed in health care financing and
delivery. These models are patterned after a dental care
financing model developed by Cormier and Levy (1981). In
the words of Cleland and King (1983), however, these
three models are "abstractions of reality" and thus are
neither comprehensive nor all-inclusive. They simply
describe general approaches to areas of risk taking, and
are used as a point of reference for analysis. The
locus-of-risk models attempt to depict the
relationship(s) of various elements of the health care
delivery systems with one another and in turn their
effect on the performance of the systems. This tripar-
tite representation is an incremental view of the financ-
ing subsystem of health care delivery and, most impor-
tantly, suggests ways in which modifications to this
subsystem have affected the "core technology" of medicine
and the decision-making of physicians. The act of re-
stricting decision-making by physicians, for economic or
technical reasons, may be seen as having an impact upon
the physician's professional autonomy.
MODEL #1. In a "first-party-at-risk" paradigm, one
of the two principals in the delivery equation, the
patient, or group of patients, is at risk for paying the
majority of costs incurred in the receipt of health care
services. This model assumes that the patient is self-
45


insured, and as such, pays, or promises to pay, for
needed services. While the provider (physician,
hospital, etc.) may be at risk to the extent of bad debt,
it is the patient (sometimes referred to as the consumer)
who bears the majority of financial risk. It is
perceived that this model fosters the most direct
attention to cost, level of care needed, and level of
care provided since the party receiving the benefits is
also the party directly responsible for paying for them.
MODEL #2. A third party that provides
indemnification or "protection against loss" for patients
takes on a significantperhaps major, and occasionally
totalresponsibility for payment for needed services.
While a variety of means may be used by the indemnifier
(or insurer) to force the patient to share in the risk
(i.e., deductibles, co-payments, limits, etc.) the major
share of the risk is borne by a third party. This third
party, being neither the provider nor the patient, is not
directly involved in the delivery or receipt of the
covered services. The presence of risk-reducing
modalities, such as those mentioned above, is clear proof
of the importance placed by insurers on the elimination
of first-dollar coverage. It is believed that such
first-dollar protection largely removes both the patient
and the provider from positions of direct concern for
cost.
46


It is also possible that third-party payments to
hospitals on the basis of retrospective, cost-based
reimbursement principles have caused hospital authorities
to lose sight of the utility, on a cost/benefit basis, of
many services. Since hospital care accounts for
approximately 40 percent of all health care expenditures,
the effect of third-party reimbursement practices on
hospitals has become the focus of a great deal of
attention from health economists and policy officials who
are attempting to gain control over the increasing costs
of health care. Studnicki points out that:
The growth of insurance has resulted in the "third
party" now imposed between the patient and the
hospital. Dollars do not flow from the patient to
the hospital for services rendered. Instead, var-
ious third parties collect dollars from citizens,
their employers and government, and redistribute
those dollars to hospitals on the basis of varying
contractual arrangements . [which] typically
involve the payments for hospital services based on
the cost of providing those services (italics mine)
(Studnicki 1983, 100).
MODEL #3. The second party in the health care
delivery equation is the provider. As the majority of
health care services consumption decisions are made, or
influenced, by physician providers, not patients, many
refuse to refer to the patient as a consumer. The
resource consumption mix characteristics may vary
depending upon the physician and the patient, but the
physician, with the concurrence of the patient, is
47


legally the only party able to implement most consumption
decisions. It is held that when the provider, the second
party in the delivery equation, is either directly at
risk for costs of care or kept aware of the ramifications
of consumption decisions by an at-risk patient, then the
provider will be more parsimonious with respect to the
level and intensity of services ordered. In the case of
Model #3 the provider (i.e., physician in an HMO system
or the hospital under DRG reimbursement) is at financial
risk to a significant degree. As such, the physician is
exposed to external scrutiny, often from non-physicians,
to make patient care decisions in a manner more
acceptable to that outside party. Thus, this evolving
shifting of risk has put the physician in the position of
control by corporations, the government, or other
interested parties, that Starr (1982) proposes had been
intentionally avoided earlier.
Hospitals and insurers generally allowed physicians
to remain independent entrepreneurs, though there
can be little doubt that leaving doctors outside of
the organizational structure of these institutions
increased the cost of medical care (Starr 1982, 26).
The reorientation of risk patterns represents a
significantly different way of bringing physicians into
the fold.
48


Part III: The Role of Insurance Mechanisms and
Government Programs in the Changing Risk
Patterns in Health Care Delivery
The health insurance market in the late 1970s was
composed largely of commercial insurers and Blue Cross-
Blue Shield plans. The Blue Cross-Blue Shield plans
enjoyed many advantages conferred upon them by state and
federal law and controlled somewhat clearly defined
areas. Freeh and Ginsburg (1988) suggest that the Blue
Cross plans received "significant regulatory and tax
advantages at the state and Federal levels, such as lower
premium taxation rates and exemption from local property
taxes and state and local income taxes" (280). As the
"Blues" were presumed to be operating in the public
interest, these advantages were rational. Interestingly,
though, the Blue Cross-Blue Shield plans, with their tax
advantages, did not seem able to penetrate a larger share
of the market than the commercial insurers. Perhaps this
was a function of the Blue Cross-Blue Shield posture of
requiring only minimal cost sharing in most of their
plans. These plans required few, if any, deductibles or
co-insurance for hospital care while the plans of commer-
cial insurers often had such risk-sharing mechanisms
built in, a parameter that can be partially explained by
the origins of Blue Cross and Blue Shield.
Historically, Blue Cross plans were controlled by
the hospital community and the Blue Shield plans were
49


dominated by the medical sector, particularly the
surgical specialties. The benefit structures of the
"Blues" plans then, might well have been designed to
encourage the utilization of hospital services and to
reward surgical procedures at a higher reimbursement
level. Commercial plans had no such "genetic" biases and
sought that sector of the health insurance market that
valued cost sharing as a way of reducing the cost to the
consumer, in this case the party paying for the health
insurance benefit, namely the employer.
From a public policy perspective the limited product
line of the "Blues" was perhaps the most important
concern. The "Blues" practice of selling only
policies with minimal cost-sharing led to a lower
level of cost-sharing in the over-all market (Freeh
and Ginsburg 1988, 280).
Blue Cross-Blue Shield plans (with reduced or nonex-
istent cost-sharing provisions), competing with plans
that called for higher levels of cost sharing, attracted
consumers, who found them more desirable in that they
offered "first-dollar" coverage rather risk sharing on
the part of the beneficiaries. Freeh and Ginsburg (1988)
hold that "because such practices led to less cost-
sharing in the aggregate, health care costs were driven
higher than they would have been otherwise" and that
health insurance data indicate that in states where Blue
Cross plans have a major market share "hospital prices
and costs were higher" (281). These non-cost-sharing
50


features of Blue Cross and Blue Shield programs, coupled
with the administrative slack such plans encouraged, may
have resulted in higher aggregate health care costs and
prices than would have been experienced had such
advantages not been available to the Blues.
A look back at government involvement in health care
financing shows that public sector resources were used
initially to expand a deficient hospital facilities in-
ventory and later to pay directly for services rendered
to persons eligible for an array of social welfare and
social insurance benefits. Public financing of health
care services, while not originally intended to do so,
enabled private sector institutional providers to en-
large their capital stock dramatically. The "Findings
and Purpose" section of P.L. 93-641. The National Health
Planning and Resources Development Act spotlights the
role Federal health care dollars have played in fueling
inflation in health care costs and utilization.
. .(2) The massive infusion of federal funds into
the existing health care system has contributed to
inflationary increases in the cost of health care
and failed to produce an adequate supply or distri-
bution of health resources . (National Health
Planning and Resources Development Act 1983).
Perhaps as a result of the private insurance world's
domination of government financing activities, the "cost
reimbursement" system that encouraged hospitals to bill
for the cost of providing services (rather than the value
51


in the marketplace) allowed hospitals to add new services
and facilities with little regard to their necessity or
economic viability. The "Washington Report On Medicine
and Health" put the issue of Federal policies that
eliminated delivery system risks into perspective.
Under the old Medicare retrospective cost-plus
reimbursement system, hospitals had no incentive for
efficiency. Medicare paid for all services unless
they were clearly unnecessary. It was to the
hospital's financial advantage to provide more
services and to extend length of stay (Brazda 1985b,
i)
The evolution of Federal public policy in the area
of health care delivery and financing reform appears to
reflect a trendthe shifting and sharing of risk for the
financing of health services. Risk patterns have been
altered in a number of ways that have affected the major
providers of carehospitals and physicians. While the
effects of policy on the institutional providers have
been studied extensively, little attention has been paid
to the scope or magnitude of the effect upon the
professional satisfaction of the physician providers.
While hospital costs accounted for 49% of health care
expenditures in 1990 (Aluise 1987), the physician holds
the key to controlling many of these expenditures.
Certainly, many factors, including community
expectations, advancing technology, and malpractice
exposure and associated costs, have an impact upon the
52


medical decision-making. None-the-less, a primary focus
of regulatory and commercial policy has been to attempt
to control the limits of medical decision-making..
There is also a growing concern with not only costs
but effectiveness of practice patterns. One potential
answer to the problems of inappropriate, confused and
even contradictory treatment practices is the development
of what Brook (1989) refers to as "Practice Guidelines."
Such guidelines, when developed, offer the prospect
enhancing the quality of care rendered and the
satisfaction of physicians rendering care covered by such
guidelines.
Various factors, including increased financial
pressures on the health care system, the rapidity of
the introduction of technology, and data showing
high levels inappropriate care, will coalesce into a
movement that will yield practice guidelines. If
the guidelines are developed with the aid of the
best methods and if they are applied constructively,
then the twin goals of increased health of the
American public and physician satisfaction can be
achieved (Brook 1989, 3027).
This notion may be slightly optimistic as such guide
lines may on one hand help to clarify medical decision-
making while on the other hand be seen as the imposition
of external controls on professional autonomy.
53


CHAPTER 3
RESEARCH DESIGN
Introduction
To study the impact of Federal regulatory policy,
and other factors in the context of health care, on
physicians' perceptions of autonomy several levels of
research (description) may be identified. They focus on
the overall relationship between Federal regulatory
policy in health care, and changing community factors
(such as patient expectations, malpractice, etc.), and
the perception of physicians vis-a-vis their professional
autonomy.
Propositions
The research question, assumptions, variables, and
propositions influence the approach and conduct of the
research. The propositions represent the specific items
to be analyzed and discussed. Two propositions, with
three sub-propositions, have been formulated. Following
the statement of each proposition, and sub-proposition,
is a discussion of its theoretical rationale with
associated citations from the literature.
These propositions suggest that professional
autonomy is a critical factor to physicians and that real
or perceived attacks on this value may negatively affect


physicians' perceptions of satisfaction.
The order of presentation of the propositions
follows a step-wise approach which begins with a
discussion of the matter of professional autonomy itself
(P^). The discussion which follows explains the issue of
professionalism and the place which professional autonomy
occupies for physicians.
Values which attracted people to the practice of
medicine are discussed in sub-proposition 1.1, as they
are seen, as the starting point from which one can begin
to assess ways in which the attitudes of physicians may
have been affected by the realities of context factor
impacts on their professional lives.
Proposition 2.o examines context factors as disqui-
eting elements in health care. Herzberg's Motivation-
Hygiene Theory is introduced as a construct which has
utility in the development of an understanding of the
extent to which factors in the context of health care
impact upon the attitudes of physicians about their
profession. Twelve (12) context factors are enumerated
and discussed.
Two sub-propositions (P2.I anc* p2.2^ introduce the
notions of autonomy constraint by context factors and the
seemingly contradictory notion that in spite of the fact
that certain factors in the context of health care may be
autonomy-constraining and thus have a negative impact on
55


physician's attitudes about their profession, they still
have largely positive views of their professional
satisfaction.
Proposition 1.0
P1>0 Autonomy is a multi-faceted property that many
physicians believe has been eroded by a variety of
factors in the context of health care. That is,
they believe that the level of personal and work-
related autonomy they actually have is somewhat less
than they expected to have as a practicing physician
when they entered medicine.
Overview. Reese (1985) proposes that autonomy may
be seen as "that which gives law to itself, or is its own
law" (77). Stamps et al. (1978) define autonomy as the
"amount of job-related independence, initiative, and
freedom either permitted or required in daily work activ-
ities" (339). These rather general definitions of this
professional property will be discussed in terms of the
attributes of professionalism and professional autonomy
as distinguished from other forms of autonomy endemic to
the medical profession.
Professionalism. Any discussion of the attributes
of the practice of medicine must inevitably address the
question of professionalism. While there is little
agreement on what constitutes professionalism, it is
widely held that medicine qualifies. Nonetheless, sever-
al common threads run through all definitions. Descrip-
tors such as intellectual, learned, service-oriented,
56


ethical, organized, self-directed, self-regulating, etc.,
permeate even the most widely divergent definitions of
professionalism.
This widely claimed appellation holds great
significance for society, even though there is no
universally accepted way to define it. Becker (1962)
holds that "'profession' is an honorific title, a term of
approbation ... a collective symbol and one that is
highly valued" (33). Becker further explains, in
discussing what he calls the "symbol" profession, that
professions possess a monopoly of some esoteric and
difficult body of knowledge that is essential to the
well-being and continued functioning of society.
Medicine, which is universally accepted as a
"profession," satisfies the requirement of being
essential to the well-being and continued functioning of
society. While medicine is of value to society, it
fulfills this responsibility one patient at a time.
Without question, the individual patient is the raison
d'etre of personal health care, that part of medicine
with which society is most familiar. Perhaps most
importantly,
. . what the members of the profession know and
can do is tremendously important, but no one else
knows or can do these things. . This knowledge
cannot be applied routinely but must be applied
wisely and judiciously to each case (Becker 1962,
35) .
57


This special knowledge, according to Becker, and the
proposition that only other physicians can judge the work
of physicians, has been used by the medical profession to
justify demands for autonomy.
Professional autonomy may be used strictly in the
interests of the client; in fact, it is likely that
without some measure of autonomy the client's
interests cannot be well served (Becker 1962, 39).
Professional Autonomy. Freidson and Rhea, in
discussing the fact that professionals typically are more
autonomous (freer from external controls than other
groups), explain that this freedom has been justified by
three claims:
First, their work entails such a high degree of
skill and knowledge that only fellow professionals
can make accurate assessments of professional
performance. Second, a high degree of selflessness
and responsibility characterizes professionals, so
they can be trusted to work conscientiously. Third,
in those rare instances in which individual
professionals do not perform with sufficient skill
or conscientiousness, their colleagues may be
trusted to undertake the proper regulatory action
(Freidson and Rhea 1965, 107-108).
The highly personal (patient-professional
relationship-centered) nature of medicine leads to the
expectation that the physician is "to make his own
decisions consistent with the norms of his profession"
(Engle 1970a, 33). Thus, autonomous behavior on the part
of physicians relative to their patients is the essence
of medical practice. Engle (1970a) contends that if the
"physician becomes less able to function in the manner
58


decreed by his profession, in this sense he could become
deprofessionalized" (33) .
Additionally, Goss (1961) suggests that "personal
responsibility" (47), on the part of the physician,
appears to be the hallmark of the patient-physician
relationship. Thus, anything short of autonomous
behavior in this relationship is eschewed by physicians.
Assumption of personal responsibility for patients
would seem to account in large measure for
physicians' unwillingness to take or give
authoritative orders concerning patient care. . .
These [professional] norms did not require, as is
sometimes thought, that each physician be autonomous
in every sphere of his activity, but only that he be
free to make his own decisions in professional
matters as opposed to administrative matters. Nor,
even in the professional sphere, did the norms rule
out the possibility of supervision; so long as
supervision came from a physician and took the form
of advice, it was within normatively acceptable
bounds for physicians (Goss 1961, 48).
Engle (1970b) suggests that professional autonomy
can be viewed on two separate levels: "(1) with respect
to the individual professional, and (2) with respect to
the occupational group or profession" (12). The concern
of this research^ much as the research conducted by
Engle, is with autonomy relative to the individual
professional. While Engle's work related autonomy to
bureaucratic organization, her division of "individual
professional" autonomy into "personal" and "work-related"
autonomy is nonetheless germane to this study.
Personal autonomy is freedom to conduct tangential
work activities in a normative manner in accordance
59


with one's own discretion. Work-related autonomy
for the professional is freedom to practice his
profession in accordance with his training. It is
this type of autonomy which appears to be important
for the professional, since a loss of work-related
autonomy or control to his client, or to any lay
individual or group, might reduce the quality of the
service he renders. The studies instigated by the
Human Relations School (Rothlisberger and Dickson,
1939; Mayo and Lombard, 1944), which stressed the
importance of personal freedom and morale to the
performance of the worker, have pointed out the
importance of personal autonomy (Engle 1970b, 12-
13) .
This "personal" and "work-related" autonomy
construct was used in the definition of question-sets for
this research as it underscored the bipartite nature of
the autonomy issue faced by physicians. Engel's work is
additionally instructive in its isolation of three
dimensions of work-related, professional autonomy. Engle
(1970b) suggests that the physician is expected to be
autonomous, relative to patients, in the areas of
"innovation, individual responsibility, and
communication" (15-16).
Autonomy with regard to innovation existed when the
physician instigated changes related to work tasks,
altered established work methods, and produced novel
ideas and/or methods. Autonomy with regard to
individual responsibility occurred when the
physician determined the uses to which his work was
put, was not subordinate to those less
knowledgeable, defined his own work goals, and was
permitted to act and think without interference. A
physician was autonomous with regard to free
communication when he had access to all vital
information, could communicate without interference
or obstacles, and participated in democratically
organized discussions (italics mine) (Engle 1970b,
16) .
60


Grumbach and Bodenheimer interchange the term
"clinical freedom" with professional autonomy.
. . clinical freedom is the ability of the physi-
cian to deliver medical care to a patient without
the uninvited imposition of outside influences whose
purpose is not the optimal health of the patient
(italics mine) (Grumbach and Bodenheimer 1990, 121).
Thus, a threat to the basic notion of autonomy may
strike at the very heart of the physician's perception of
how to best serve the interests of patients. Because
service to the patient is the driving force for many
physicians, individual practitioners jealously protect
their ability to act in the best interest of each
patient; thus, the perceived need for professional
autonomy. Intrusion from outside forces, such as
insurance companies, government, the judicial system,
etc., may be seen by physicians as compromising their
ability to provide the level of service deemed necessary.
Becker (1962) contends that "the symbol of the profession
is not merely the selfish propaganda; many propositions
contained in it [i.e.. autonomy] are in large part true"
(39) .
According to Starr, an attack on autonomy (perceived
or real) may strike at the core of why many men and women
choose medicine as a career.
The profession's success in maintaining autonomy has
had both material and psychological dimensions. Had
doctors been subject to the hierarchical control of
insurance companies, hospitals, and large medical
practice organizations, their income would probably
61


have suffered. They would have lost freedom in
choosing their hours, their clients, their fields of
specializationall the advantages that come with
"being one's own boss." For many physicians, these
concerns about autonomy have outweighed strictly
financial considerations (Starr 1982, 25-26).
Hughes further amplifies this notion:
The true professional ... is never hired. He is
retained, engaged, consulted, etc., by someone who
has need of his services. He the professional, has,
or should have, almost complete control over what he
does for the client (Hughes 1963, 256).
In an examination of three groups of health care
professionals (physicians, nurses, and support staff),
Stamps et al. observed:
For all three groups of professionals, autonomy is
the most important factor in terms of contributing
to the overall level of satisfaction in their jobs.
This is not at all unexpected. The level of
professionalization within the health field is very
high and the essential criterion for definition of
a "professional" is autonomy (Stamps et al. 1978,
342) .
Not unlike Starr's observation, the Stamps group
found that pay was a less important issue to physicians
in their study group. In fact, autonomy, "organizational
requirements," "interaction," "task requirements," and
"job status" all ranked ahead of pay as determinants of
satisfaction for the physician group (Stamps et al.
1978) According to Wholey and Burns (1991), "early
research evidence suggests that the ideology of
professional autonomy has strongly shaped physicians'
attitudes" (257).
62


The National Association of Managed Care Physicians
(NAMCP), an association of physicians who participate in
any one of a number of managed health care environments,
lists as its first priority the maintenance of "Physician
autonomy to treat patients in a manner consistent with
quality health care" (National Association of Managed
Care Physicians 1991, 1). The NAMCP claims that
approximately 350,000 of the 600,000 physicians in the
United States are involved in some form of managed health
care.
Proposition 1.1
Pll Certain values (i.e., service, prestige, challenge,
autonomy, etc.) were critical determinants in the
physician's choice of medicine as a profession.
Overview. While professional autonomy is the
primary focus of this research, it is but one of several
factors of importance to physicians. This sub-
proposition will briefly discuss nine factors which may
have been of importance to physicians in their selection
of medicine as a profession. The relative importance of
these factors may lie at the heart how practicing
physicians react to certain factors, such as regulatory
policy, in the context of health care. The research will
attempt to establish a rank-ordering of these factors in
terms of how they affected entry into medicine by the
members of the study group.
63


Factors Affecting the Selection of Medicine as a
career. Earlier, unpublished, analyses by Karshmer
(1988, 1990) suggests that several professional and
personal factors are important determinants in the
selection of medicine as a profession.
1. Service to society; Service to society, in this
context, refers to the contribution that an individual
physician can make to the general, collective well-being
of a community through his/her efforts professionally and
personally.
2. Service to individuals: Medicine, in almost a
unique way in the world of the professions, allows its
practitioners to see very direct effects on the lives and
well-being of the individuals and families that they
serve.
3. Prestiqe/respect; An August 1991 Cable News
Network poll indicated that physicians continued to enjoy
the highest level of respect and prestige of all
professions surveyed. While many physicians believe that
medicine is no longer as highly respected as it once was,
it is still held in high regard by the general public and
other professions.
4. Science; Many members of the medical profession
are attracted to the discipline by the scope and
complexity of the "science" they are called-upon to
manage on behalf of their patients. While in strictly
64


technical terms physicians are technologists rather than
scientists, for many this affinity for science is a
leading attribute of the practice of medicine.
5. Intellectual challenge: The practice of
medicine is a highly individualized discipline. The
complexity of patients' individual needs, coupled with
the breadth, depth and ambiguity of much of the science,
makes the practice of medicine intellectually
challenging.
6. Problem solving: The very essence of the
practice of medicine is problem definition and problem
solving. Both aspects are complicated by the
idiosyncrasies of the patients and the needs of both the
community at large and the health care community in
particular.
7. Autonomy/independence: The history of the
practice of medicine in the United States is rife with
efforts on the part of the medical community to remain
autonomous and independent. In a significant way,
professional autonomy and independence are seen by many
practitioners as the attributes that allow them to
function effectively on behalf of their patients and the
community.
8. Lifestvle/income: Medicine historically has
been, and will probably continue to be, one of the more
65


attractively compensated professions. Increasingly,
however, the lifestyle and income of physicians are being
scrutinized (and criticized). Thus, some practitioners
have become concerned about the continuance of rewards
that physicians deem not only appropriate, but necessary
if the "best" are to be attracted to the field of
medicine.
9. Lifetime of learning: The "tradition" of medi-
cine is one of intense undergraduate and graduate educa-
tion coupled with ongoing continuing medical
education.
Proposition 2.0
P2.o Several factors in the context of health care
(hygiene factors, according to Herzberg) have a
disquieting effect on physicians1 attitudes about
their profession.
Overview. Proposition 1 and Proposition -j_ ^
addressed the issues of the importance of autonomy to
medical professionals, and the importance of autonomy,
among other factors, to the selection of medicine as a
profession. This proposition suggests that a number of
factors in the context of health care may negatively
impact the way in which physicians perceive their
profession and introduces those factors in light of the
Motivation-Hygiene Theory construct.
The Importance of Context Factors. Herzberg,
Mausner, and Snyderman, in their seminal work dealing
66


with the Motivation-Hygiene Theory, suggested that job
factors were not necessarily bi-polar that is that the
opposite of a dissatisfier is not necessarily a
motivator. While low pay may, for example, cause
dissatisfaction, higher pay will not automatically result
in satisfaction. Additionally, job content factors
(motivators) appear to contribute to satisfaction while
elements in the context (hygiene factors) such as
external regulations, seem to contribute to
dissatisfaction (Herzberg, Mausner, and Snyderman 1959).
Medicine is generally seen to be rich in job content
factors (internal factors, i.e., achievement, recognition
and responsibility), which partially explains the appeal
of medicine to highly motivated, self-directed, focused
people. This property of medicine may also explain the
apparent paradox in which physicians can find themselves
highly satisfied with their profession in general and
highly critical of many of its elements.
Context, or hygiene factors (factors external to the
practice of medicine) tend to constrict autonomya
feature that is frequently seen as an intrinsic, if not
essential, value of the profession. It is assumed that
negative attitudes about context factors are "incremen-
tal" dissatisfiers.
Twelve (12) context factors were identified through
earlier research by Karshmer (1990).
67


A. Malpractice
B. Public view of the profession
C. Patient expectations
D. Role of non-physician health care providers
E. Requirements of insurance companies and other
third-party payers
F. Government involvement in the delivery and
financing of health care
G. Competition in medicine
H. Quality assurance mechanisms
I. Alternative delivery systems
J. Aging of the population
K. Specialization in medicine
L. Pace of developments in medicine
Proposition 2.1
£>2.1 Some factors in the content of health care are
perceived by physicians to be autonomy
constraining11 in nature. Forces that appear to
constrain the professional autonomy of physicians
have a negative effect on the vay in vhich
physicians perceive various dimensions of their
professional lives.
Overview. Proposition#0 introduces the notion of
"hygiene" or context factors and presents twelve such
factors. This proposition suggests that some or all of
these factors may be seen by physicians as being autonomy
constraining and thus may have a negative affect on
various dimensions physicians professional lives. Five
such dimensions are discussed.
Discussion. For the purposes of this analysis,
"autonomy constraining" refers to policies and/or actions
that negatively affect, or appear to negatively affect,
the range of diagnostic, therapeutic, organizational or
personal options available to physicians. Grumbach and
Bodenheimer relate that:
68


... to the extent that cost-containment mechanisms
wrest away from physicians the ability to determine
the type and quality of services, physician autonomy
is reduced (Grumbach and Bodenheimer 1990, 122).
As the role of the physician in the health care
system is dominant, the response patterns of this group
to policy initiatives is critical. Ku and Fisher (1990),
in a study of physician attitudes toward health care cost
containment policies, reported that physicians generally
supported "policies that increased responsibilities or
costs for patients and disfavored policies that decreased
physicians' autonomy of practice" (25). This study
surveyed 500 physicians nationally in 1984 and
specifically looked at 23 health care cost-containment
policies. The results were reported in 1990.
Because much Federal regulatory policy (i.e., DRGs,
utilization review, concurrent review, second opinions,
etc.) and malpractice issues are perceived by physicians
to limit the range of their professional options, such
mechanisms can be viewed as autonomy constraining
(Grumbach and Bodenheimer 1990). The sequelae of such
restrictions were described by Grumbach and
Bodenheimer:
The harness and prod of utilization review have
turned American physicians into the most second-
guessed and paper work-laden physicians in western
industrialized democracies (Grumbach and Bodenheimer
1990, 123).
When professional autonomy appears to be
69


constrained, physicians may sense a diminution in the
quality of their professional experiences. Several
dimensions of professional experience can be identified.
Five dimensions of professional experience w6re tested in
this study in an effort to evaluate the effects of
changes in professional autonomy. These are:
1. Effect on income
2. Effect on self-esteem
3. Effect on happiness with work
4. Effect on peer relations
5. Effect on patient relations
Proposition 2.2
p2.2 Tbe 9estalt of medicine is, in and of itself, a
powerful "satisfier." Physicians can, on one hand,
see themselves as "satisfied" professionals, while
on the other hand they can also find aspects of the
profession less than satisfactory.
Overview. The previous proposition and sub-
proposition dealt with the notion of context factors, the
factors themselves, and the dimensions of professional
life that might be affected by them. This proposition
addresses the idea of physicians being able to maintain a
positive attitude about their practices in the face of
varying levels of negative response to context factors.
Discussion. According to Shafritz (1985), "gestalt"
is a "configuration, pattern, or otherwise organized
whole whose parts have different qualities than the
70


whole" (229). Rehm relates that the Gestalt
psychologists believed that:
. . the relationship among components, rather than
their fixed characteristics, determines what is
perceived. Thus, a person perceives a whole pattern
that may be much more than the collection of
individual elements (Rehm 1991, CD/ROM entry).
Shaver adds:
. . the process of perception involves not simply
the faithful encoding of external stimulation, but
rather requires the active organization of that
stimulation into a more meaningful whole unit or
gestalt (Shaver 1977, 96).
Thus, while, in the aggregate, physicians may be
able to find that their "overall experience as a
practicing physician" is quite good (the gestalt), they
can also find dimensions of their professional lives
(fixed characteristics) to be dissatisfying.
The suggestion implicit in this sub-proposition is
that physicians still find their professional autonomy
within the "Zone of Acceptance" and, thus, will continue
to function in a positive manner in the context of health
care. The other side of this logic is that at some
point, the impingement of context factors on physicians'
attitudes may cause physicians to leave the "Zone of
Acceptance" or the "Zone Indifference," and, in so doing,
perhaps, refuse to cooperate with regulatory initiatives
to restructure health care financing and/or organization.
71


Variables
The propositions summarize a number of observations
made by the researcher during the course of delineation
of issues associated with the effect of regulatory
policy, and other issues in the context, on physicians'
perceptions of autonomy. These propositions form the
basis for establishing association between the
independent variable and the dependent variable, as
modified by the intervening variables.
In this thesis, the variables may be classified as
follows:
Independent variable. Regulatory, societal, and
law-related (torts and malpractice) forces in the
environment of health care that may have a constraining
effect on physicians.
Dependent variable. Physicians' perspectives on
professional autonomy.
Intervening variables. In ex post facto research,
certain constructs, or intervening variables, may also
affect the apparent associations between the independent
and dependent variables. Intervening variables,
according to Isaac and Michael, are "conceptual states"
and generally
cannot be directly observed or measured and are
hypothetical conceptions intended to explain
processes between the stimulus [independent
variable] and the response [dependent variable]
(Isaac and Michael 1971, 16 ).
72


Physicians' expectations regarding "proper" roles,
responsibilities, and images associated with the practice
of medicine. These expectations are made up of a
constellation of images derived from the socialization
process that occurs throughout several phases of a physi-
cian's development, as follows.
1. Personal (physical, intellectual and societal)
growth: From earliest cognizance, people who use health
care services are acculturated with the image of the
professional preeminence of the physician. The numerous
rules and laws pertaining to who can practice medicine,
coupled with the complexity and obscurity of the body of
knowledge of the profession, have led to the acceptance
by society of a professionally autonomous discipline
that, until recently, has been all but immune from
external examination and/or control. Thus, the person
entering the field often does so with the societally
acceptable expectation of autonomous, independent,
professional action.
2. Medical education: While the physician's
educational process conveys mixed messages of
powerlessness and authority, the gestalt of autonomy is
passed on through the educational process itself. On one
hand, the student is systematically overwhelmed by the
sheer mass of information to be internalized and the
73


keenly competitive nature of the other students. On the
other hand, these features lead the aspiring professional
to the conclusion that this rather large body of
knowledge can only be applied through the filter of
independent, autonomous judgment. This judgment can be
tested against that of co-professionals, but not by
outsiders. The process of medical education, while not
necessarily the most intellectually expansive, requires a
high degree of perseverance and goal orientation on the
part of the studentfactors that contribute to the image
of the autonomous professional.
3. Post-graduate training (internship and residen-
cy) : Training at the internship and residency levels
gives the new physician an in-depth opportunity to be co-
opted by the prevailing medical culture and ideology.
The intensity of the training milieu, and the mentor
relationship with senior physicians, act to reinforce and
perpetuate certain values and mystiques of the
professionautonomy being one of the most critical.
4. Medical practice subculture reinforcement: By
the time physicians complete their formal training, the
autonomy value has become an ingrained "given" of the
profession. While external pressures threaten this
highly valued attribute of medicine, physicians find
strong and continuing reinforcement from their peers to
protect autonomy as an inviolable underpinning of the
74


I
I
I
provision of quality medicine. While peer review by
co-professionals is tolerated, external controls, exer-
cised by non-physicians, are generally rejected as they
threaten the autonomous nature of the profession.
Tvoe of Research Conducted
This study can be characterized as directed (ground-
ed) qualitative research. The nature of the sample used
for analysis was such that the results are descriptive
rather than predictive. In this context, "directed"
refers to empirical inquiry, driven by observation.
Merton (1957) describes this approach as "the empirical
generalization: an isolated proposition summarizing
observed uniformities of relationships between two or
more variables" (95).
The notion of directed research implies, in part,
empirical inquiry is so organized that if and when
empirical uniformities are discovered, they have
direct consequences for a theoretical system. .
(Merton 1957, 95).
The words "in part" reflect Merton's (1957) concern
that strict adherence to, and testing of "predetermined
hypotheses" "... stultifies possibilities of obtaining
fertile new findings" (98). Merton, however, argued
that:
75


Fruitful empirical research not only tests
theoretically derived hypotheses, it also originates
new hypotheses. This might be termed the
"serendipity" component of research, i.e., the
discovery, by chance or sagacity, of valid results
that were not sought for (Merton 1957, 98).
This type of analysis, according to Glaser and
Strauss (1967), misses a vital dimension of grounded
theory which posits the "discovery of theory from
data . theory [which] fits empirical situations"
(25) .
This concept [Merton's notion of serendipity] does
not catch the idea of purposefully discovering
theory through social research. It puts the
discovery of a single hypothesis on a surprise
basis. Merton was preoccupied with how
verifications through research feed back into and
modify theory. Thus, he was concerned with grounded
modifying of theory, not grounded generating of
theory (Glaser and Strauss 1967, 2n).
Glaser and Strauss apply four tests to "grounded,
substantive theory."
The practical application of grounded sociological
theory, whether substantive or formal, requires
developing a theory with (at least) four highly
interrelated properties. The first requisite
property is that the theory must closely fit the
substantive area in which it will be used. Second,
it must be readily understandable by laymen
concerned with this area. Third, it must be
sufficiently general to be applicable to a multitude
of diverse daily situations within the substantive
area, not to just a specific type of situation.
Fourth, it must allow the user partial control over
the structure and process of daily situations as
they change through time (italics mine) (Glaser and
Strauss 1967, 237) .
Further, descriptive research, as explained by Isaac
and Michael, characterizes research that:
76


is used in the literal sense of describing
situations or events. It is the accumulation of a
data base that is solely descriptiveit does not
necessarily seek or explain relationships, test
hypotheses, make predications, or get at meanings
and implications (Isaac and Michael 1971, 18).
Qualitative research, according to Kirk and Miller,
"is an empirical, socially located phenomenon, defined by
its own history, not simply a residual grab-bag
comprising all things that are 'not quantitative'" (Kirk
and Miller 1986, 10). Additionally, qualitative research
is "socially concerned, cosmopolitan, and, above all,
objective" (10).
"Objectivity," too, is an ambiguous concept. In one
sense, it refers to the heuristic assumption, common
in the natural sciences, that everything in the
universe can, in principle, be explained in terms of
causality. In the social sciences, this assumption
often misses the point, for much of what social
scientists try to explain is the consequences of
inner existential choices made by people. In
ordinary language, when we ask "why" a person acts
as he or she does, we are generally inquiring
teleologically about his or her purposes. Indeed,
if knowledge itself is taken to be merely the
inevitable consequence of some cause and effect, its
logical status would seem to be compromised (Kirk
and Miller 1986, 10).
Thus, the purpose of the study adheres to the
following ideas of Van Dalen and Meyer as reported in
Isaac and Michael (1971):
a. To collect detailed factual information that
describes existing phenomena.
b. To identify problems or justify current
conditions and practices.
c. To make comparisons and evaluations.
d. To determine what others are doing with similar
77


problems or situations and benefit from their
experience in making future plans and decisions
(Isaac and Michael 1971, 18 ).
78


CHAPTER 4
METHODOLOGY
Overview
The purpose of this research, as stated in Chapter
1, is to explore the effects of context factors,
including regulatory policy, on physicians' perceptions
of autonomy. The research accomplishes this through the
collection and analysis of data from a large,
geographically dispersed group of physicians. This group
was made up of physicians practicing in a wide variety of
medical specialties.
The study group was made up of physicians attending
continuing medical education (CME) activities in Vail,
Colorado, during the months of February and March 1991.
This type of sample is known of as an "availability
sample" or a sample made up of "those who volunteer or
are otherwise available" (Garson 1971, 91). Naturally,
an availability sample is not necessarily representative.
A complete list of the CME activities included in this
sample is attached (see Appendix B).
The selection of this group was predicated on the
researcher's access to these physicians and the
multispecialty, geographically dispersed nature of the


group. As the researcher was the director of the CME
program sponsoring the programs, he had free access to
the attendees. While the original study design called
for a survey of physicians licensed to practice medicine
in Colorado, the CME attendees were substituted in order
to attain a higher response rate more affordably. The
response rate of 49 percent signifies only that
approximately one-half of the physicians asked to
complete the questionnaire did so. Thus, the data
derived from this group allow for description of how this
multi-specialty group of geographically dispersed
physicians responds to the factors under analysis.
While analysis of these data will not permit
prediction, they are suggestive for purposes of theory
generation. Additionally, comparison of demographics of
the study group with physicians nationally indicated that
the study group had many characteristics similar to a
representative American Medical Association population.
The national data used for this purpose were collected by
the American Medical Association and are reflective of
the characteristics of more than 600,000 physicians.
This comparison is presented as background for
understanding of the substantive data derived from the
study group.
80


In addition, an effort has been made to compare
study group findings with other studies. Again, such
comparisons have utility in extending the
generalizability of findings.
Given these caveats, descriptive data about this
group are potentially useful as there is a virtual vacuum
of such analysis locally, regionally, or nationally.
While, from a statistical point of view, one may not be
able to say that the results are useful from a predictive
point-of-view, one is also unable to dismiss the
findings. Isaac and Michael explain that descriptive
research:
. .is used in the literal sense of describing
situations or events. It is the accumulation of a
data base that is solely descriptive it does not
necessarily seek or explain relationships, test
hypotheses, make predictions, or get at meanings and
implications. . (Isaac and Michael 1971, 18).
Thus, the data collected for this study are
descriptive of the attitudes of a group of 335 physicians
whose demographic characteristics are known in relation
to a national American Medical Association population.
Additionally, description may eventuate in
prediction.
Data Collection
The description of physicians' attitudes towards
factors that may threaten their autonomy is the goal of
this research. Why attempt to describe physicians'
81


perceptions of autonomy? Robinson, Athanasiou, and Head
(1970) relate several answers to this question.
One answer is ". . because they are there."
Another answer involves the belief that what people
"say" has (or should have) some relation to what
they do. . Actually, we attempt to study atti-
tudes in order to verify the hypotheses that 1) they
are really there. 2) that they have some relation to
what people do. and 3) that knowledge of attitudes
will help us to understand behavior (italics mine)
(Robinson, Athanasiou, and Head 1970, 95).
In an effort to better understand physician
responses to the affects of context factors on their
perceived autonomy, a questionnaire was formulated to
derive information about their attitudes. The attitudes
of interest fell into the following categories:
1. Attitudes that led physicians into medicine in
the first place.
2. Attitudes about the importance of autonomy and
independence, especially as they affect professional and
personal needs of physicians.
3. Attitudes about expected versus actual levels of
autonomous behavior.
4. Attitudes about the impact of certain factors in
the context of medicine on the physician's ability to
practice medicine in an acceptable manner.
A search was conducted to find an instrument that
could be used to examine these issues in such a manner as
to be able to draw some conclusions about each "attitude
area" individually as well as to be able to relate
82


responses from two or more of the areas examined. While
a few instruments of limited utility were found, none was
expansive enough to gather data responsive to the four
areas discussed above. Thus, a questionnaire was
developed specifically for the study. The instrument
incorporated the following types of questions.
1. Questions that derived basic data about the study
group:
a. Demographic data
b. Practice-specific data
c. Other descriptive data
2. Questions relating to professional autonomy;
a. Physicians' expectations
b. "Realities" experienced by physicians
c. Causal relationships
3. Questions relating to physician satisfaction
4. Questions that linked and attempted to crossvali-
date 2 and 3
A first draft of the questionnaire was circulated to
six local private practice physicians. These physicians
also received an overview of the research project as
background for their analysis of the draft questionnaire.
The transmittal letter asked:
1. Do the questions measure what is intended to be
measured?
83


2. Are all of the terms clear or do some need to be
defined?
3. Do you think that the respondents will interpret
the questions in a reasonably similar manner?
4. Do the responses provided cover all potential
answers to the questions?
5. Do the questionnaire and the cover letter create
a positive impression, one that motivates physicians to
respond?
6. Does any aspect of the questionnaire suggest
bias on the part of the researcher?
7. Do the transition statements convey the proper
information?
8. Are there questions you would either delete,
add, or alter?
The responses were reviewed and suggestions were
incorporated into the second version of the
questionnaire.
A second iteration of the questionnaire was field
tested in Vail, Colorado, on February 11, 1991. Twenty
physicians attending the Vail Psychiatry Conference and
the Vail Emergency Medicine Conference were asked to
complete the revised questionnaire. In addition to the
questions asked of the first group of six physicians, the
field test group was asked to record the time required to
complete the survey instrument. Field test respondents
84


indicated that the survey instrument took about fifteen
minutes to complete. The field test questionnaires, and
accompanying comments, were carefully analyzed and the
observations and suggestions were incorporated into the
final version of the survey form.
Consistent with the original study design, the
second iteration of the questionnaire, with accompanying
supporting materials, was submitted to the Colorado
Medical Association Board of Directors. The purpose of
this submission was to gain the approval and assistance
of the Medical Association in circulation of the
questionnaire to member physicians in Colorado. The
Association Board of Directors approved the questionnaire
on February 22, 1991. While the questionnaire was not
circulated to all physicians in Colorado, the review and
approval by the Colorado Medical Association's Board of
Directors constituted an important endorsement of the
survey.
The original study design called for surveying all
Colorado physicians. The cost of this strategy, coupled
with a low anticipated response rate, were the principal
reasons for its abandonment. Instead, a sample that drew
upon physicians attending continuing medical education
conferences sponsored by Denver's Rose Medical Center and
conducted in Vail, was substituted.
85


The final version of the survey instrument was
circulated to all physicians attending ten medical
education conferences in Vail, Colorado, during February
and March 1991. Each questionnaire was accompanied by a
cover letter (Appendix C) and the survey was announced
and endorsed, without comment, by the physician chair of
each conference. The questionnaire was neither discussed
nor reviewed at this time. The actual number of
physicians from the United States attending these
conferences was 682. Of this group, 335 (49.12 percent)
submitted usable responses.
Review of Statistical Methodology
Overview
The 335 questionnaires were completed by physicians
attending the medical conferences listed in Appendix B.
Each completed questionnaire contained up to 137
responses to the 38 questions and sub-questions. The
questionnaires were individually coded and prepared for
analysis using the statistical package SPSS/PC+, version
4.0. The following sections address the types of data
derived, the nature of the sample, and the types of
statistical tests applied in the analysis of these
data.
Tvoes of Data and the Nature of the Sample
The selection of statistical procedures to apply was
conditioned on two questions; the first about the data
86