Citation
Barriers to clinicians' implementation of behavioral interventions related to sexual practices and drug use of patients living with HIV

Material Information

Title:
Barriers to clinicians' implementation of behavioral interventions related to sexual practices and drug use of patients living with HIV
Creator:
Shure, Christine Lemore
Place of Publication:
Denver, CO
Publisher:
University of Colorado Denver
Publication Date:
Language:
English
Physical Description:
xi, 143 leaves : ; 28 cm

Subjects

Subjects / Keywords:
Medicine and psychology ( lcsh )
HIV-positive persons -- Sexual behavior ( lcsh )
Physician and patient ( lcsh )
Health counseling ( lcsh )
Health counseling ( fast )
HIV-positive persons -- Sexual behavior ( fast )
Medicine and psychology ( fast )
Physician and patient ( fast )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Bibliography:
Includes bibliographical references (leaves 136-143).
Thesis:
Anthropology
General Note:
Department of Anthropology
Statement of Responsibility:
by Christine Lenore Shure.

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:
122890915 ( OCLC )
ocn122890915
Classification:
LD1193.L43 2006m S48 ( lcc )

Downloads

This item has the following downloads:


Full Text
BARRIERS TO CLINICIANS IMPLEMENTATION OF BEHAVIORAL
INTERVENTIONS RELATED TO SEXUAL PRACTICES AND DRUG USE
OF PATIENTS LIVING WITH HIV
by
Christine Lenore Shure
B.S., University of Missouri-Columbia, 1997
A thesis submitted to the
University of Colorado at Denver and Health Sciences Center
in partial fulfillment
of the requirements for the degree of
Master of Arts
Anthropology
2006
,p.. w i,?..
ft, s \
/-u. f
...


This thesis for the Master of Arts
degree by
Christine Lenore Shure
has been approved
by
Susan Dreisbach
Tiay, io,
Date


Shure, Christine Lenore (M.A., Anthropology)
Barriers to clinicians implementation of behavioral interventions related to
sexual practices and drug use of patients living with HTV
Thesis directed by Professor Kitty K. Corbett
ABSTRACT
Clinicians medical encounters with HTV positive patients represent a key
but underutilized moment for discussing ongoing HIV and STD transmission
risk. This thesis explores the main barriers that health care providers face in
implementing behavioral interventions related to sexual practices and drug use
with their HTV positive patients. These barriers to thorough and appropriate
health communication are assessed in terms of being primarily social or
structural in nature. In a project with CDCs National Network of STD/HIV
Prevention Training Centers, semi-structured interviews with key informants
(N=6) and HIV clinician providers (N=l 1) provided data for the study.
Among the social barriers, various forms of clinician discomfort with
discussions surrounding sexual practices and drug use was the most frequent
barrier. This discomfort stemmed from a lack of knowledge on the subject
matter, the stigma of HTV, the possibility of a negative reaction from a patient,
difficulty dealing with the partners of HTV positive patients, and a reticence to
cause the patient discomfort with these discussions. Conflicting beliefs about
the role of the clinician as a behavioral interventionist was also frequently
mentioned as a social barrier. Many clinicians felt that a more comprehensive
referral network was needed to help them with these behavioral interventions.
A number of clinicians also felt it was not the role of the biomedical clinician to
perform behavioral interventions during the medical encounter.
A lack of time was mentioned as a structural barrier by most of the
participants interviewed, and looms as one of the largest barriers to
implementation of the guidelines. Other structural barriers included


administrative barriers to change, monetary reimbursement concerns, a need for
more training, and the need for more comprehensive referral networks.
The five perceived attributes of innovations within Diffusion of
Innovations theory was used in order to analyze the barriers found in this study.
Of the five attributes a lack of relative advantage looms large for this
innovation, with some barriers also existing in compatibility with existing
beliefs, observability, trialability, and complexity. Quality improvement
studies, health communication issues, and the interrelation of social and
structural barriers are also discussed.
This abstract accurately represents the content of the candidates thesis. I
recommend its publication.
Signed
Kitty K. Corbett


DEDICATION
This thesis is dedicated to all first generation college students and their
families.


ACKNOWLEDGEMENT
I would like to thank the Departments of Anthropology and Health and
Behavioral Sciences for their support with my degree and this thesis. I would
especially like to thank Kitty Corbett for her many words of wisdom that
contributed to my thesis and to life in general. Finally, I would like to thank my
husband, Joel Shure, for his undying support in my studies.


TABLE OF CONTENTS
Tables............................................................xi
CHAPTER
1. INTRODUCTION..............................................1
Objectives...................:.........................1
Aim 1.............................................1
Aim 2.............................................2
Background..............................................2
Significance............................................4
2. REVIEW OF THE LITERATURE...................................8
HTV and the Biomedical System...........................8
Prevention with Persons Living with HTV: Efficacy
Studies...........................................9
Training to Overcome Missed Opportunities for
HIV Prevention...................................11
Barriers to Use of Behavioral Interventions in
Practice.........................................15
Quality Improvement Theory and Research................19
Diffusion of Innovations.........................20
Quality Improvement Studies in Health Care.......26
vu


The Anthropology of the Medical Encounter..............30
Biomedicine as Ethnomedicine....................32
Health Communication in Biomedicine....................34
Highlights and Key Gaps in the Literature..............39
3. PRELIMINARY QUANTITATIVE DATA........................... 40
Methods................................................40
Participant Profiles............................41
Measurement of Variables in the Evaluation
Instruments.....................................43
Findings from the Post-Course Survey...................44
Findings from the Long-Term Survey.....................47
Discussion of Survey Results...........................49
4. METHODS...................................................51
Data Collection........................................51
Course Details..................................52
Human Subjects..................................53
Sampling........................................53
Data Management and Analysis...........................56
Qualitative Data................................56
vm


5. FINDINGS
61
Social Constraints........................................61
Discomfort as a Social Barrier......................62
Role as a Social Barrier............................81
Structural Constraints....................................88
Time as a Structural Barrier........................89
Other Structural Barriers: Administrative, Monetary
(ICD-9 codes), Training, and Referral Networks......94
6. DISCUSSION..................................................109
Introduction........................................... 109
The Interrelation of Structural and Social Barriers......110
The Role of Health Communication.........................113
Quality Improvement Studies Revisited....................114
An Analysis of Results through Diffusion of
Innovations Theory.................................116
7. CONCLUSIONS.................................................125
Implications of the Project..............................125
Recommendations....................................126
Limitations of the Data..................................127
Future Directions and Research Needs.....................129
IX


APPENDIX
A. INTERVIEW QUESTION
GUIDE.......................................131
B. PRELIMINARY QUANTITATIVE DATA: PARTICIPANT
PROFILE TABLES..............................132
BIBLIOGRAPHY.................................... 136
x


LIST OF TABLES
Table
3.1. Evaluation forms used and time of completion...........................41
3.2. Perceived barriers reported in the post-course evaluation..............45
3.3. Perceived barriers reported in the long-term evaluation................49
4.1. Major themes and codes.................................................60
B. 1. Gender of course participants........................................132
B.2. Profession of participants............................................132
B.3. Primary professional or functional role of participants...............133
B.4. Number of years practicing in their reported primary profession.......133
B.5. Facility/setting......................................................134
B.6. HIV/AIDS services.....................................................135
xi


CHAPTER 1
INTRODUCTION
Objectives
HIV clinicians face many barriers to thorough health communication with
their patients. This is especially true when the health communication involves
discussions of behavioral risk factors related to sexual practices and drug use.
Despite the barriers faced by HIV clinicians, these conversations, should take
place during the medical encounter in order to help control the spread of HIV.
These conversations are especially important since as many as one in three
HIV-positive people continue unprotected sexual practices after learning they
are HIV infected (Kalichman et al. 2001).
This project has two aims:
Aim 1
First, this thesis is designed to gain insight into whether the key barriers
that clinicians face to implementation of behavioral interventions with their HIV
positive patients are primarily social (individual level or interactional) or
structural (system level or institutional) in nature. Identification of structural
and social barriers in prevention communication is necessary to design effective
interventions that are responsive to these barriers. It will also shed light on the
bigger issue of patient-physician communication within the biomedical system.
Identifying barriers and challenges to implementation of the national guidelines
will also help the National Network of STD/HIV Prevention Training Centers
(NNPTC) better address the training needs of their course participants.
1


Aim 2
Second, this thesis is designed to gain a more detailed understanding of
the identified barriers and challenges that medical providers who serve persons
living with HIV face in following national guidelines for incorporating behavior
change interventions for STD and HIV prevention into medical visits with their
HIV positive patients. This increased understanding is accomplished through
the use of elements of Diffusion of Innovations theory as a framework for the
discussion of the thesis results. The detailed understanding of these social and
structural barriers will assist in implementing the changes that need to take
place in the biomedical encounter in order to better facilitate these discussions.
Background
In 2003, the Centers for Disease Control and Prevention (CDC), the
Health Resources and Service Administration (HRSA), the National Institutes
of Health (NIH), and the HIV Medicine Association of the Infectious Disease
Society of America (IDSA) put forth a list of recommendations for medical care
providers in order to better incorporate HIV prevention into the medical care of
persons living with HIV:
Screen for transmission risk behaviors and sexually transmitted diseases,
provide brief behavioral risk-reduction interventions in the office setting,
refer selected patients for additional prevention interventions and other
related services, and facilitate notification and counseling of sex and
needle-sharing partners of infected persons (CDC 2003a).
These recommendations stimulated the creation of a curriculum to train
clinicians in dissemination of the guidelines during the medical encounter with
2


persons infected with HIV. Following the CDC/HRSA/NIH/IDSA guidelines,
in 2004 and 2005, the National Network of STD/HIV Prevention Training
Centers (NNPTC) participated in the development of a Prevention in Care
curriculum to train medical care providers of HIV positive patients. The course
aims to train HIV care providers to incorporate science-based intervention
methods in order to help their patients reduce risk behaviors, even in highly
constrained settings for care.
The NNPTC is a network of ten regional centers funded by the CDC to
train clinicians to diagnose, treat, manage, and prevent transmission of STDs in
the United States. As mandated by the CDC, NNPTC courses are regularly
evaluated by an independent organization. The current pilot training
curriculum, Prevention with Persons Living with HIV, has been evaluated by
the Center for Research in Health and Behavioral Sciences at the University of
Colorado at Denver and Health Sciences Center. This evaluation consisted of
pre-course, post-course, and long-term survey data that allowed for an
assessment of changes in knowledge, practice patterns, and attitudes of course
participants. Some of this survey data were presented as a preliminary study in
the review of literature section of this report. This preliminary data helped to
guide the development of this current qualitative study, and has allowed for
some comparisons between results.
This thesis represents an addition to the existing Prevention with Persons
Living with HIV pilot course evaluations. The NNPTC Evaluation Committee
has endorsed my project; it will enhance their evaluation capacity beyond what
their funding allows, and will address questions that they would like answered
about whether their trainings make a difference to real world practice settings.
3


The Prevention in Care training course has been piloted and evaluated at six
U.S. sites (Denver, St. Louis, Tampa, Baltimore, New York, and Oakland)
during the spring and summer of 2005. Participants who agreed to be part of
long-term follow-up evaluation at a post-course time point were later contacted.
When the curriculum is finalized, a CDC-approved version of the course and its
evaluation materials will be utilized in trainings of several thousand clinicians at
over 10 sites around the country. I was able to contact pilot course participants
who agreed to provide long-term follow-up evaluation information for semi-
structured interviews.
The researcher conducted in-depth, semi-structured interviews with some
of the clinicians who participated in the pilot of the curriculum. Topics in the
interviews included barriers to implementing behavioral interventions during
the medical encounter with HIV positive patients, the self-efficacy of the
clinician in clinical encounters with HIV infected persons, clinicians beliefs
about patients attitudes and willingness to change, and brief background
information on the clinical setting and the clinician (See Appendix A).
Significance
The HIV virus was first recognized in 1981, and since then the virus has
become responsible for a global AIDS pandemic. The HIV virus and AIDS are
an unprecedented threat to global health. With the advent of antiretroviral drugs
in the 1990s, and the subsequent improvement of these treatments since then,
people can live indefinitely with HIV. The antiretroviral drugs are, however,
not the answer to the AIDS epidemic. Many citizens of underdeveloped
4


countries do not have access to antiretroviral drugs, and suffer terribly from the
burden of AIDS on their population.
In the United States, while most individuals have access to antiretroviral
therapies, not all take the drugs properly or agree to take the drugs at all. A
staggeringly high number of HIV infected individuals are unaware of their viral
status. At the end of 2003, an estimated 1,039,000 to 1,185,000 persons in the
United States were living with HIV/AIDS, with 24-27% undiagnosed and
unaware of their HIV infection (Glynn 2005). The vast majority, of these
infections were caused by sex and drug use practices, with male-to-male sexual
contact, injection drug use, and heterosexual contact being the largest risk
factors. Through 2003, only a little over 20,000 cases of HIV in the US were
caused by other risk factors, which include hemophilia, blood transfusion,
perinatal, and risk not reported or identified (CDC 2003b). Most of these
alternate risk factors have been greatly reduced in recent years with the
implementation of better HIV testing methods for donated blood and better
techniques for reducing perinatal infection. Currently HIV infection rates
reflect primarily sex and drug; use behavioral risk factors. For this reason, the
identification, control and modification of these risk factors is the key to control
of the HIV/AIDS epidemic in the United States.
As with any other epidemic, in order to control transmission of the
disease, social or behavioral factors relating to its transmission must be
addressed. In the U.S., the majority of persons who are aware of their HIV
status are receiving medical care (NNPTC/AETC 2005:6). Because of this, the
medical encounter with persons living with HIV represents the most important
and in many cases the only moment in the biomedical patient/provider
relationship in which behavioral information can be gathered and appropriate
5


behavioral interventions can be disseminated. Various studies have shown that
brief counseling sessions and short prevention messages may change patient
behavior and reduce risky behavior (Thrun et al. 2003, Richardson et al. 2004,
Fisher et al. 2002, NNPTC/AETC 2005). Studies have also shown that most
HIV care providers do not ask their patients about ongoing HIV transmission
risk (Elford et al. 2000, Bull et al. 1999, Gardner et al. 2002, Duffus et al. 2003,
NNPTC/AETC 2005).
Biomedical, clinicians face many challenges in communication with their
HIV positive patients. Their communication is affected by contextual factors
and by personal and cultural meanings, as well as both parties knowledge and
attitudes that can hamper effective communication between the patient and
physician. The fact that HIV is highly stigmatized in society can also affect
communication during the medical encounter. The confidence or self-efficacy
of the physician in dealing with topics such as sex and drug use can affect what
questions are asked and whether behavioral interventions are discussed during
the medical encounter. Structural barriers present in the clinical setting can also
affect communication between patient and clinician. Time pressures,
administrative and political policies, and resources available to the clinician can
all affect health communication. This is especially true when the clinician is
under rigid time constraints or lacks adequate support from the administration to
implement behavioral interventions.
In order to effectively address the social modes of transmission of the HIV
virus, researchers must identify the best ways to communicate HIV risk
behavior information to at risk populations. While considerable effort has been
spent on aiming prevention messages toward persons not yet infected with the
virus, the Centers for Disease Control and Prevention have recently identified
6


HIV positive persons as an additional important target audience for prevention
messages (CDC 2003a). In response to this, training programs have been
developed by prevention training centers to facilitate in the dissemination of this
important intervention. Studies have also been done to test the efficacy of
dissemination of this intervention.
Clinicians and researchers can have different goals for research of the
clinical encounter (Stoeckle 1982). The researcher may be looking for
regularities in clinician communication and the correlates that explain them.
The clinician is more concerned with what works in care, and may be more
interested in the language and questions that get the most information about
disease. While these goals are not necessarily incongruent, they certainly can
be. The insights from researchers should be useful so that practitioners can
apply them in health care (Pendleton 1982). Keeping this in mind, this project
was designed to produce outcomes that can be immediately useful to both HIV
practitioners and curriculum developers.
7


CHAPTER 2
REVIEW OF THE LITERATURE
This review of the literature covers important aspects of curriculum
development and dissemination as they relate to HIV prevention interventions.
The first section covers background information on HIV prevention programs
and intervention dissemination leading to the goals of this study: identification
of social and structural barriers to patient-clinician communication and an
analysis of these barriers. Following this review of background literature are
three sections that relate to important goals of this study. The first of these
sections is a review of quality improvement studies that help with the
understanding of why clinicians may or may not implement an important
intervention. Diffusion of Innovations theory is reviewed in order to build a
theoretical framework for detailed analysis of the social and structural barriers
identified in this thesis. Next is a section titled The Anthropology of
Biomedicine, which is an analysis of the biomedical encounter through an
anthropological perspective. This section provides an insightful anthropological
perspective of the medical encounter. Finally, health communication in
biomedicine is reviewed. This section looks at a model developed by a
physician to frame the important aspects of health communication during the
biomedical encounter. This section helps to shed light on the dynamics of
patient-clinician interaction during the medical encounter.
HIV and the Biomedical System
In order to understand the barriers that clinicians face to implementation
of the behavioral interventions that they learn in the current curriculum, a
8


review of past efficacy studies, studies on training, and studies on the barriers
that clinicians face is reported below. Past efficacy studies of HIV intervention
counseling interventions have supported the development of the current
curriculum based on findings that a series of brief counseling sessions can be an
effective tool for disseminating HIV prevention messages (Kamb et al 1998,
Thrun et al. 2003, Richardson et al. 2004, Fisher WA 2002). A discussion of
these efficacy studies is followed by a brief review of recent training studies
related to STD/HIV prevention. These studies provide a useful comparison to
the present curriculum being disseminated, and show the past effectiveness of
training as a means of disseminating behavioral interventions. A discussion
then follows of the barriers that clinicians face to implementation of behavioral
interventions. This discussion provides background for why clinicians may be
experiencing the barriers that are outlined later in the findings of this thesis.
Finally, a review of literature on HIV-related stigma provides background on
this important barrier to communication.
Prevention with Persons Living with HIV: Efficacy Studies
Efficacy is an important measure because it shows the actual impact of an
intervention on the individuals that received it. Efficacy studies are many times
the first step in the creation of public health programs, followed by research into
the potential dissemination of the innovation or its effectiveness.
9


Project RESPECT
In order to test the efficacy of brief counseling sessions to prevent
infection with the HIV virus, a study was done by Kamb et al. (1998) to
compare two interactive HIV/STD counseling interventions against didactic
prevention messages typical of current clinical practice. One group of
participants received enhanced counseling, one received brief counseling
sessions, and two other groups received didactic messages similar to those
currently found in clinical settings. The study was conducted at five public
STD clinics around the US. Participants consisted of 5758 heterosexual, HIV-
negative patients aged 14 years or older who came to the clinics for STD
examinations. Reports showed that self-reported condom use was higher in
both the enhanced counseling group and the brief counseling group than it was
in either didactic message groups. At each of the five study sites, STD
incidence was lower in the counseling intervention groups than in the didactic
messages intervention groups.
The researchers, Kamb et al. (1998), reached a few important conclusions.
First, the study showed that brief counseling sessions during the clinical
encounter could be as effective as enhanced counseling sessions that involve
more time and multiple visits. The study also showed that because these
effective counseling sessions were brief, they could easily be incorporated into
the medical encounter with patients and implemented by clinicians, even in a
busy public clinic environment. Finally, the study showed that the
incorporation of a client-centered approach to the brief counseling session, in
lieu of more canned didactic messages, was a more effective method of
reducing behavioral risk factors.
10


Recent HIV/STD Efficacy Studies
More recently, studies have been conducted to assess the effectiveness of
clinician-initiated risk reduction counseling with HIV positive patients (Thrun et
al. 2003, Richardson et al. 2004, Fisher WA 2002). Clinicians have only
enough time during the clinical encounter to provide brief behavioral
interventions. Interventions presented to HIV clinicians should be time-
sensitive as well as effective. Richardson et al. (2004) did a study on the effect
of brief safer-sex counseling by HIV medical providers, which tested the
behavioral outcome of 585 HIV-positive persons. Two intervention arms were
compared to control groups. One intervention arm received brief counseling
emphasizing the negative consequences of unsafe sex, while the other
intervention arm received brief counseling emphasizing the positive
consequences of safer-sex. Unprotected anal and/or vaginal intercourse was
reduced by 38% (PO.OOl) among participants receiving the negative
consequences intervention, while no significant changes were seen in the
positive consequences behavioral intervention. This study not only provides
evidence of the efficacy of brief behavioral interventions performed by HIV
clinicians; it also sheds light on the type of intervention message that should be
emphasized.
Training to Overcome Missed Opportunities for HIV Prevention
While evidence has shown that brief clinician-initiated behavioral
interventions with HIV positive patients can reduce HIV transmission, studies
have also shown that there are currently many missed opportunities taking place
in the clinic setting (Elford et al. 2000, Marks et al. 2002, Bull et al. 1999,
11


Morin 2002). Although clinical settings may provide opportunities for risk
discussions, this potential is not always being fulfilled (Elford et al. 2000).
Elford et al. (2000) did a study with gay men in central London to
determine if these men were discussing HIV risk reduction with their general
practitioners. One thousand and four gay men who utilized one of five gyms in
central London completed anonymous surveys for the study. The results
showed that fewer than one-third of the men (30%) had discussed safer sex with
a general practitioner. Of those who had discussed safer sex with their general
practitioner, one in six (17%) had found it difficult to do so. These results are
worrisome since about one-quarter of respondents reported unprotected anal
intercourse in the three months prior to the survey time.
In a study of six public HIV clinics in California conducted by Marks et
al. (2002), HIV-positive patients were asked if clinic providers ever talked with
them about safer sex or disclosure of their HIV status. Overall, only 71% of
patients surveyed reported that a clinical provider had ever talked with them at
least once about safer sex. Interestingly, men who have sex with men (MSM)
and whites were less likely than other racially identified groups to receive HIV
prevention messages. In light of the fact that MSM represent an important
transmission risk group, these results are disturbing.
Training in the skills and knowledge necessary to carry out behavioral
interventions can help providers to eliminate many of these missed
opportunities. Thrun et al. (2003) showed that both providers and patients
respond favorably to curricula encouraging providers to incorporate prevention
counseling into HIV care. In this study, a curriculum developed for
incorporating HIV prevention into HIV care was taught in a workshop attended
by HIV care providers who worked in a large urban HIV care clinic. At the end
12


of two years most providers believed that the training helped them feel
comfortable initiating risk-reduction discussions. Most clinicians reported that
their patients responded positively to the risk-reduction discussions.
Henderson et al. (1999) evaluated nine HIV/AIDS training programs to
assess the degree to which the programs produced positive healthcare systems
changes to the delivery of HIV/AIDS care. This study is important because
systems changes, or structural changes to the biomedical system, are necessary
in order to implement new policies regarding HIV/AIDS healthcare delivery.
About half of the training participants (55.5%) were able to provide at least one
example of a systems change that had resulted from the training, with trainees
having an average of 1.03 system change examples (Henderson et al. 1999).
While changes in knowledge and skills of the individual healthcare practitioners
can be readily achieved and measured, systems changes resulting from
HIV/AIDS training can be more difficult to achieve and to assess. This
important study shows significant system changes resulting from training and
thus makes training an effective tool for this difficult purpose.
Huba et al. (2000) and then Lalonde et al. (2002) assessed the long-term
impact of HIV/AIDS education and training on clinician practice. These studies
looked at the long-term training effects on general HIV/AIDS perspectives that
clinicians have, health care provider service provision, and changes in
procedures and operations at the health care system level. In this way, they
assessed the individual changes as well as the overarching structural changes
within the biomedical system as it pertains to HIV/AIDS care. Telephone
interviews were conducted with 218 trainees 8 months after training. Generally,
respondents rated the training experience in all three areas assessed as positive.
Sustained changes resulting from the training were seen. There was increased
13


knowledge on the part of the clinicians at the patient and system levels, and
positive changes in provider-patient interactions. Some felt increased comfort,
which is instrumental in fostering better health communication. Interesting is
the fact that the system functioning domains were most positively associated
with training, even more so than the patient-care domains. This study, like the
Henderson et al. (1999) study above shows that, while system changes resulting
from HIV/AIDS training can be more difficult to achieve and to assess, these
changes are in fact happening as a result of effective HIV/AIDS training.
Closer to home, the Center for Research in Health and Behavioral
Sciences (CRHBS) conducted a long-term course evaluation of the Denver
STD/HIV Prevention and Training Centers 3-day STD/HIV Intensive Clinical
Course (Devine et al. N.d). Clinician-participants long-term retention of
knowledge, skills, and practice patterns from the training were assessed. The
study assessed participants who took the course between May 2001 and
November 2004, thus spanning 3.5 years of courses, and data on about 100
participants for the pre-course assessment and around 60 participants for the
post-course assessment. In the knowledge section, clinicians were given five
case scenarios and posed questions on them. At six months post-course there
was a significant and sustained 12.5% improvement in knowledge. Self-
reported skills improved and were sustained at six months in three categories:
communication, technical, and diagnostic. At six months post-course,
communication skills showed a 22.3% sustained improvement, technical skills
showed a 54.0% sustained improvement, and diagnostic skills showed a 67.6%
sustained improvement. Sustained improvements in knowledge and skills are
easier to attain than sustained improvements in practice patterns. This is
because, in general, behavior is difficult to change. In this study, practice
14


patterns showed a slight improvement in most categories assessed, although
most of these improvements were not significant. The researchers concluded
that sustained improvements in practice patterns proved difficult to achieve.
Despite this setback, this study has shown that training is effective at increasing
and sustaining knowledge and skills among clinicians, which is the first step to
changing provider behavior and practice patterns.
In summary, while research has shown that prevention discussions are not
taking place as often as they should, it has also shown that education and
training can positively affect many of these processes. The processes that can be
positively affected by training include those on the individual provider level as
well as systems changes in the broader biomedical context.
Barriers to Use of Behavioral Interventions in Practice
Training has been shown to be a valuable tool for changing provider
practice patterns and attitudes toward behavioral risk reduction discussions with
HIV-positive patients. While training has proven to be valuable, barriers to use
of the learned knowledge in practice continue to exist. Knowledge is easier to
influence than performance, which is easier to influence than patient outcomes:
Programs influencing only predisposing factors, such as knowledge,
attitudes, and skill, are less effective in improving performance than those
influencing predisposing, enabling, and/or reinforcing factors (Umble and
Cervo 1996:167).
While increased knowledge and skills are necessary to create practice changes,
these changes will not take place if self-efficacy barriers or structural barriers in
the clinical environment persist.
15


Sometimes the most significant barrier to providers use of skills in
practice is their never having learned the skills to begin with. The AIDS
Education and Training Centers (AETC) did a study to determine the
characteristics of hard-to-reach providers. Three key themes emerged which
identified who was hard to reach: professional group designation, providers with
low volumes of HIV-infected clients, and HIV-expert clinicians who serve high
volumes of HIV-infected clients (Bradley-Springer 2003:28). Bradley-Springer
et al. (2003) found that there were many reasons cited by these groups as to why
they were unable or unwilling to receive training. Some of these issues were
practice related, which included the stresses and limitations of a busy practice,
not enough time to attend, the expense of education, and lack of administrative
support. Isolation was also a barrier to training for these providers. Some were
located in a rural location, were culturally isolated possibly because of a special
minority group that they serve, or were even isolated in an urban area with lack
of transportation or in a high-crime neighborhood.
HIV and Stigma
Illness can become a way that society labels a person; this is especially
true for illnesses that are stigmatized. Stigma allows a disease to stamp
powerfully disconfirming cultural meanings onto the sick person (Kleinman
1988). The sick person internalizes the stigma, further affecting them by
mentally altering their illness experience. Stigma can also affect the human
rights of the individual, resulting in property loss, bodily injury or murder, loss
of marriage rights and loss of access to healthcare as well as possibly leading to
incarceration. Because of the powerful potential of the stigma of disease, illness
16


takes on a vast array of psychological and societal meanings that are reflected in
the illness narratives of the affected individual.
HIV/AIDS since its very beginnings in the 1980s has been an extremely
stigmatized disorder (RCAP 2005). HIV-related stigmatization is a process
that also reinforces existing social inequalities based on race, gender, ethnicity,
and sexual orientation (Aggleton in RCAP 2005:1). Individuals infected with
HIV have been denied employment, housing, medical care as well as having
numerous violent attacks against them. Because of these social responses,
individuals can face discrimination because of the violation of social norms.
For instance, fear of being blamed for violating sexual norms (i.e. heterosexual
monogamy) can lead to non-disclosure to sexual partners if someone is
infected (RCAP 2005:1). Some countries such as Cuba and Vietnam have
actually rounded up HIV positive individuals into camps to try to stop the
spread of infection. In this way, HIV infection has become a reason for
incarceration, thus creating extreme stigma against the illness.
In many instances, HIV as no exception, illness becomes viewed as a type
of cultural punishment for a sin or evil acts that the affected individual must
have performed. Religions and governmental systems can perpetrate this idea,
and thus further ostracize the affected individuals. The cultural idea of illness
as punishment further limits the credibility of the affected individual and
allows for justification of human rights abuses.
As an example, HIV is an extremely stigmatized disorder in Vietnam.
HIV infection is considered a social evil that is punishable and worthy of
discrimination. The government as well as the general public hold this belief.
Because of these cultural beliefs, HIV prevention campaigns are geared around
incarcerating the HIV infected as well as ending the evil behaviors, such as
17


drug use and prostitution that lead to HIV infection. Unfortunately, these
government tactics have proven to be ineffective at stopping the spread of HIV,
and the virus continues to spread at an alarming rate every year in Vietnam.
The stigma of this disease keeps individuals from seeking the proper medical
care to treat their infection, thus making their illness worse. In the case of HIV
in Vietnam, stigmatization has created a situation in which individual human
rights have been violated, thus creating inequity in care of sickness. In this type
of inequity, certain diseases are acceptable and treated likewise, while other
diseases are stigmatized and individuals are denied treatment. While the United
States has not implemented a formal policy against HIV positive individuals,
civil rights violations are common and the disease remains a social stigma.
Because of the negative social implications of HIV/AIDS, the medical
encounter with persons living with HIV is a unique and especially difficult
encounter within biomedicine. While any clinical encounter is full of personal
bias, beliefs, and emotions possessed by the patient as well as the clinician, the
medical encounter with HIV positive individuals is made more difficult with the
addition of stigma associated with the condition.
In a study done by Deborah Preston et al. (2004) on risk behaviors of
rurally based men who have sex with men, stigma was found to be an associated
risk. Stigma was predictive of higher risk behavior, and is thus a major factor
with implications for lowered health status (Preston et al. 2004). Stigma can
negatively affect an individuals self esteem and in many situations reduces
social status (Goffman 1963). The beliefs and attitudes that clinicians bring into
the medical encounter can have a stigmatizing effect on the patient. This can be
a formidable barrier to patient-clinician communication. Alternately, a
hyperawareness of potential stigma on the part of the clinician can create a
18


situation during the medical encounter in which difficult topics are not breached
lest stigmatization of the patient might happen. Both of these situations can be
avoided if the clinician uses proper communication skills. An intervention that
does not effectively address the stigma of HIV infection will be less useful in
the treatment and prevention of individuals in need.
Regardless of the type of barriers that providers face to using new skills in
their practice, work can be done in order to overcome these barriers and to
realize the full potential of trainings. This can be done by careful analysis of the
barriers that clinicians face, and the revising of curricula to fully address these
barriers.
Quality Improvement Theory and Research
The current study represents a quality improvement study as it is intended
to improve health communication between providers and HIV positive patients,
as well as help to identify and alleviate the current barriers to implementation of
behavioral interventions with these patients. It is therefore important to review
the relevant literature on quality improvement research and theory.
Diffusion of Innovations theory developed by Everett Rogers (2003)
provides a useful theoretical framework for quality improvement research by
analyzing why clinicians may or may not adopt the behavioral interventions
proposed in this study. The five attributes of innovations outlined by Diffusion
of Innovations theory are discussed as a useful tool for analyzing the results of
this study. These sources of literature provide a useful background for
understanding how the current study can contribute to the improvement of the
curriculum it evaluates.
19


All too often, clinical research that demonstrates the efficacy of a novel
clinical process or innovation is not widely adopted and practiced by the target
clinical community. There is a chasm between the knowledge of efficacious
clinical guidelines and processes and the actual clinical practice of these
healthcare improvements. Because of this gap, the successful implementation
of new clinical guidelines for practice into the existing clinical environment
should involve careful attention to the translation of research into practice.
Without this last step, the value of the research is lost. Through quality
1
improvement analysis, my research attempts to understand why HIV providers
who have learned valuable and efficacious behavioral interventions for their
HIV-infected patients may not integrate these interventions into their practice.
Through this research, I hope to strengthen an already existing curriculum by
analyzing data collected from the target community after they have received the
intervention. In this way I hope to help close the gap between the clinical
research that went into developing this intervention and the clinical practice of
these guidelines.
Diffusion of Innovations
Robert Sanson-Fisher et al. (2004) assert that the behavioral sciences can
make a substantial contribution to the process of changing providers behavior.
Behavioral change theories should be used to develop theoretical frameworks
that explain how behavioral change in the clinical setting is achieved. Accurate
measurement of relevant clinical behavior should also take place in order to
properly evaluate the effectiveness of an intervention. This can be
accomplished through direct observation of clinic activities, as well as
20


examination of the accuracy of self-report by the healthcare providers. Finally,
Sanson-Fisher et al. (2004) argue that there should be an in-depth examination
of the incentives and barriers that clinicians face to implementation of
guidelines, and thorough evaluation of the intervention.
One behavioral theory that has proven useful for an in-depth examination
of the incentives and barriers that clinicians face, and is utilized in this current
study, is Diffusion of Innovations Theory developed by Everett Rogers (2003).
Diffusion of Innovations Theory is a useful theoretical base in quality
improvement research because it is instrumental in identifying and explaining
why a certain intervention is or is not being adopted, who is adopting the
intervention, and how it is changed during the process of being adopted by the
community (Schoenwald and Hoagwood 2001). Everett Rogers notes that an
assumption is often made that after an innovation is developed, and its efficacy
and effectiveness demonstrated, widespread adoption and uptake will occur
automatically (Rogers 2003). This is an assumption that can lead to a failure of
adoption and failure to sustain use of proven efficacious and effective
interventions. Much time and money can be wasted on the development of
innovations that never get fully realized in their target community. Diffusion of
Innovations Theory helps to identify the predictors of adoption as well as the
barriers to adoption of an innovation (Rogers 2003).
Diffusion of Innovations Theory proposes that the attributes or
characteristics of an innovation most likely to affect the speed and extent of the
adoption and diffusion process can be identified (Rogers 1983, Oldenberg and
Parcel 2002). It provides a useful paradigm for understanding how new
interventions are introduced and adopted within the medical system. The theory
consists of various constructs, which combine to form the overarching doctrine.
21


These elements include: the definition of diffusion of innovations, the
innovation-decision process, attributes of innovations and their rate of adoption,
adopter categories, diffusion networks, and the change agent. While these
elements combine to provide a clear and thorough picture of all aspects of the
innovation process, the constructs can be used separately to understand specific
aspects of an innovation. The researcher has utilized the basic concepts of
diffusion as well as Rogers Attributes of Innovations in order to analyze the
results of this study.
Rogers defines diffusion as:
the process by which an innovation is communicated through certain
channels over time among members of a social system, with the aim being
to maximize the exposure and reach of innovations, strategies, or
programs (Rogers 2003).
Rogers' above definition of diffusion contains crucial elements in the diffusion
of new ideas. There are four elements of this definition as is defined and
elaborated upon below. (1) Innovation an idea, practice, or object that is
perceived as new by an individual (Rogers 2003:12). The innovation need not
be a new idea to the individual, but the attitude, knowledge about, and/or
decision about action regarding the idea must be new. In the present study, the
innovation is the guidelines for conducting behavioral interventions with HIV
positive patients. The concept is not new, although the recent emphasis paid to
this form of patient-provider communication is new. (2) Communication
Channels these are the means by which messages get from one individual to
another (Rogers 2003:18). In the case of the intervention in this study, the
communication channels would be the training course as well as the clinicians
22


that have undergone the training. The patient-clinician medical encounter is
also a communication channel. The concepts of health communication
discussed later in this review of literature are pertinent to an understanding of
the clinician as a communication channel. (3) Time this involves the rate of
adoption for the individual as well as the larger social system. Rogers has
developed mathematical formulas that can help to predict this rate of adoption.
While the rate of adoption is an important concept, the present study was not
designed to explore this area of diffusion theory. (4.) Social System this is a
set of interrelated units that are engaged in joint problem solving to accomplish
a common goal (Rogers 2003:23). In the case of the present study, the units are
individuals; and in some sense the units can also be the clinics or even health
organizations.
Diffusion research is probably best known for facilitating an
understanding of the individual by assigning them to different adopter
categories. It is also a valuable tool for analyzing characteristics of innovations.
In this study the researcher has used the latter construct in the analysis of the
study results.
The Five Attributes of Innovations
Everett Rogers has outlined five attributes of innovations that can help to
explain the rate of adoption. The attributes are: relative advantage,
compatibility, complexity, trialability, and observability. The researcher has
explored how these attributes help to explain the barriers that clinicians report
they encounter during implementation of the guidelines. The relative advantage
of an intervention and its compatibility with existing ideas tend to be the most
23


important attributes of innovations when considering their ability and rate of
adoption. Complexity, trialability, and observability do also play a role in the
adoption of many innovations. While barriers related to these attributes may
not be present in an innovation, if the potential adopters perceive them as being
present it is detrimental to the adoption process.
Relative advantage. Relative advantage is defined as, the degree to
which an innovation is perceived as better than the idea it supersedes(Rogers
2003:15). There are subdimensions of relative advantage which include
economic profitability, low initial cost, a decrease in discomfort, social prestige,
a saving of time and effort, and immediacy of reward (Rogers 2003: 223). An
innovation will be adopted more quickly if all or most of these subdimensions
are advantageous. An innovation may not be adopted at all if there are large
barriers of this nature.
Compatibility. Compatibility is defined as, the degree to which an
innovation is perceived as consistent with the existing values, past experiences,
and needs of potential adopters (Rogers 2003: 240). Compatibility of the
innovation reduces uncertainty with the advantages of adoption and fits well
with the individuals current situation. Rogers asserts that an innovation can be
compatible or incompatible with sociocultural values and beliefs, previously
introduced ideas, and/or client needs for the innovation (Rogers 2003: 240).
When introducing an innovation to a new culture, one can experience reduced
adoption because of conflict with deeply held religious or cultural beliefs.

24


Innovations are also more easily adopted when they are familiar or
compatible with previously held ideas. Rogers asserts that, old ideas are the
main mental tools that individuals utilize to assess new ideas and give them
meaning (Rogers 2003: 243). When an innovation is very compatible, there is
a reduced need for behavior change. Furthermore, when an innovation
supersedes the previous idea or institution, there is more incentive to adopt.
Finally, successful adoption of an innovation requires that the innovation
meets a felt need of the individual. In the case of prevention innovations, the
individuals may not feel a need for the innovation, thus making it more difficult
for the change agents to implement.
Complexity. An innovation should be easy to understand and use.
Training courses should consist of short, interactive and experiential sessions
with straightforward concepts and simple intervention messages. The potential
adopter can feel frustrated if the innovation is too complex, or they may not
remember the important elements of the innovation.
Trialability. An innovation tends to be more successful if it can be
experimented with on a limited basis (Rogers 2003). Moreover, if an
innovation can be re-designed through trial by the adopter in order to fit
individual needs or comfort, than it has a better chance of being adopted.
Observability. This relates to the degree to which the results of an
innovation are visible to others (Rogers 2003). In the case of prevention
innovations, the observability of the outcome is not always visible. HIV/AIDS
25


prevention is one example of this, especially prevention with positives in which
the outcome is the prevention of HIV transmission to partners of HIV positive
individuals.
The five perceived attributes of innovations are revisited in the discussion
section of this study in light-of the barriers that clinician participants report they
have when attempting to implement the guidelines for incorporating behavioral
interventions into the medical encounter with patients living with HIV.
Quality Improvement Studies in Health Care
The ultimate goal of quality improvement studies in health care is to
improve health outcomes. One major aspect of this literature focuses on the
improvement of health care provider trainings in order to optimize the chances
of desired practice changes taking place. Although there is a wealth of research
on the efficacy of clinical guidelines, there is continuing evidence of the failure
to translate clinical research findings into existing practice. As a result of this, it
is thought that 30% to 40% of patients do not receive treatments of proven
effectiveness (Sanson-Fisher, Grimshaw, and Eccles 2004). Many times the
wrong choice of intervention is made. For example, Lisa Bero (1998) found
that the most common method of intervention is passive dissemination of
information, such as a conference publication or mailing of educational
material, despite the fact that this method has been shown to be generally
ineffective and only results in small practice changes (Bero et al.1998).
26


She asserts that:
the choice of intervention should be guided by the evidence on the
effectiveness of dissemination and implementation strategies, the
characteristic of the message, the recognition of external barriers to
change, and the preparedness of the clinicians to change (Bero et
al. 1998:5).
In order to tackle the difficult topic of changing provider practice patterns
and behavior, Wyszewianski and Green (2000) have developed a framework for
improvement of training focused on changing clinicians practice patterns.
Although most traditional clinician training focuses on knowledge,
Wyszewianski and Green assert that there is a need to also have behavior-
oriented training, especially for hard-to-reach providers. While the knowledge-
oriented training involves an evidence-based approach, namely by
disseminating information through continuing education and guidelines, the
behavior-oriented approach adds a focus on general behavior change strategies.
These strategies include rewards, strong incentives for positive behavior change,
and the removal of obstacles to change. The researchers conclude that while
this framework still needs to be empirically tested, the broader strategy that it
proposes will most likely prove useful.
In a quality improvement study of methodology, a systematic review of 14
continuing medical education activities (CME), Davis et al. (1999) found that
the CME courses that relied on didactic sessions, in contrast to CME sessions
that involved participant activity and practice of skills, were not as effective at
changing physicians performance. From these results the researchers contend
that methodologies involving only didactic sessions should receive less credit
and funding than fully integrative and interactive sessions. This study shows
27


that the methodology of delivery is as important to successful dissemination of
the intervention as the information being covered in the intervention.
In a quality improvement analysis, Bero et al. (1998) did an assessment of
systematic reviews of interventions and found that many of these reviews were
not assessing important factors such as cost effectiveness of the intervention and
the successful reaching of target audiences. In this sense, not only are many
interventions missing their target audience, the reviews of the effectiveness of
these interventions are also missing valuable opportunities for critique and
improvement.
In a review of the implementation of clinical practice guidelines, Davis et
al. (1997) looked at various factors that effect the adoption of guidelines. They
found that the adoption of guidelines was affected by certain qualities of the
guidelines, characteristics of the health care professional, characteristics of the
practice setting, regulations, and patient factors. Davis et al. used Everett
Rogers Diffusion of Innovations theory to explain why these types of factors
would affect adoption. If the clinical guidelines coincided with and represented
only a small change to the existing beliefs and values of the clinician and
patient, and represented only a small change to the existing structural aspects of
the clinical environment, they were much more likely to be adopted. The
researchers contended, the adoption of any innovation or the dissemination of
new medical knowledge should be considered in a holistic, contextual manner
(Davis et al. 1997:6). Interactive guideline dissemination such as community
based interventions and practice-based interventions were found in this study to
be more effective than didactic passive dissemination of guidelines. Traditional
didactic lecture-based CME were least effective and reminder systems,
academic detailing, and multiple interventions were most effective, with audit
28


and feedback also being a useful tool. These results mirror later results found
by Davis et al. (1999) and Wyszewianski and Green (2000) described earlier in
this section in relation to CME and training activities.
Cabana et al. (1999) did a study analyzing why clinicians fail to follow
clinical practice guidelines. They assert that little is known about how
physicians change their practice methods in response to guidelines, and that in
many cases guidelines have little effect on changing physician behavior. In an
analysis of the guideline-related barriers reported in numerous studies from the
mid sixties through 1999, Cabana et al. (1999) found two important trends.
First, physicians were more likely to report that guidelines were difficult to
follow when asked about them in theory than they were when asked about
specific guidelines. What this tells us is that one of the crucial steps for
guideline adoption for clinicians is getting the clinicians to at least try the
guideline. Secondly, the researchers found that it was easier for clinicians to
follow guidelines that recommended adding a new behavior than it was for them
to follow a guideline recommending the elimination of an established behavior.
This basic trend holds true for human behavior change in most settings, and
therefore it is not surprising that it was also found to be true with clinician
practice change. The principles of Diffusion of Innovations theory once again
come into context, with guidelines being more successful that require less
drastic changes.
Davis et al. contend that there are complex intrapersonal, interpersonal,
and professional educational variables that affect the physician-learner (Davis
et al. 1999:873). The quality improvement studies described above reflect this
assertion. What these studies above show is that in order to be fully effective,
quality improvement research should cover all aspects of the intervention
29


delivery process, including the method of delivery, barriers faced by the
clinician-participants in implementation, and the reviews of the processes
themselves. Covering one aspect of possible quality improvement research, my
research focuses on the barriers clinicians face to implementation of behavioral
interventions. Since these behavioral interventions are based on guidelines, a
focus on quality improvement research related to guideline dissemination is in
order. This study does just this, utilizing Diffusion of Innovations theory as a
framework for analyzing results in this study.
The Anthropology of the Medical Encounter
A review of anthropological approaches to the medical encounter is
provided below. Medical anthropology has helped the researcher to understand
the social and structural intricacies of patient-clinician interaction during the
biomedical encounter. The review of medical anthropology in this section also
provides some background for understanding the origins of Diffusion of
Innovations theory. Compatibility with the cultural beliefs of the social system
is important for adoption of innovations (Rogers 2003). The earliest roots of
diffusion research are in anthropological studies, and current anthropological
research commonly contains elements of diffusion theory. The communication
approach [found in diffusion research] is utilized by a number of disciplines,
including anthropology, the products of which are seen in the government
communication institutes, such as the national institutes of health and the
centers for disease control (Rogers 2003).
Anthropological theory allows for an in depth analysis of the material or
physical aspects of humans and human systems as well as the nonmaterial
30


aspects of culture, behavior and consciousness. Within anthropological theory,
materiality is assumed to be only one aspect of a larger system (Hahn and
Kleinman 1983:308), and social relations are simultaneously a material and an
ideational reality (Godelier in Hahn and Kleinman 1983). Medical
anthropology, in particular, studies human health systems in their broad social
and cultural contexts (Barfield 1997).
There are three fundamental premises of medical anthropology that are
outlined by Thomas Barfield in his Dictionary of Anthropology (1997), which
make this discipline useful for gaining a comprehensive understanding of
current health communication issues within the biomedical encounter. First,
illness and healing are best understood holistically, in the contexts of human
biology and cultural diversity. Illness and disease cannot be fully understood in
the context of biomedicine alone, because this context is focused primarily on
the physiological aspects of illness. Second, disease is both influenced by
human behavior and requires biocultural adaptations. Across the board, through
numerous studies spanning several decades, the social sciences have shown this
premise to be true. Finally, and perhaps the most important premise for my
present study, is the assertion that the cultural aspects of health systems have
important pragmatic consequences for the acceptability, effectiveness, and
improvement of health care (Barfield 1997:316). The successful diffusion of
innovations can only take place once the cultural aspects of health systems are
understood. Health communication between patient and clinician depends upon
this understanding.
31


Biomedicine as Ethnomedicine
In a medical anthropology text from the 1970s, ethnomedicine is defined
as the beliefs and practices relating to disease which are the products of
indigenous cultural development and are not explicitly derived from the
conceptual framework of modem medicine (Bauwens 1978). This assumption
that biomedicine is somehow separate from its surrounding culture is untrue.
As with any other system of medicine, biomedicine is a product of the culture
that created it, and is thus a sociocultural system. In current definitions of
ethnomedicine, biomedicine is always included.
Biomedical theory emerged out of positivist notions that the world, or
human conditions, can be explained fully through materialist or concrete theory
(Waitzkin 1991). Within this framework, nonmaterial events such as
consciousness and culture hold little or no weight (Hahn and Kleinman 1983).
Health care, however, is a context or culture within which all health
communication takes place (Pettegrew 1982). In the U.S., this context is
primarily biomedical, and thus to understand healthcare in the U.S. one should
have a firm understanding of the cultural foundations of the biomedical system.
Although biomedicine sees itself as a fully scientific discipline that is separate
from morals, ethics and the social system, it is in fact just the opposite.
Biomedicine is an ethnomedical system that is created out of social values
and norms and operates within these limitations. It is subject to the same biases
as any other sociocultural system, and thus these biases, beliefs, and actions
should be analyzed. The medical encounter, being a social relation, should be
assessed not only as a biomedical diagnostic tool, but also as a human social
interaction. A wealth of studies spanning all of the social science disciplines
32


and decades of research have shown that human behavior is one of the main
determinants of health outcomes. Health simply cannot be achieved when the
consciousness, actions, and social setting of the individual is not taken into
account.
The medical encounter inevitably includes a world of sociocultural issues,
from the patient as well as the clinician, that need to be properly acknowledged
and managed in order for healthcare to be successfully administered:
The [medical] interview is not simply an interpersonal communication
skill or technique, but a meeting of persons full of emotions, attachments,
ideologies, values, rights, duties, politics, and conflicts (Billings and
Stoeckle 1999:ix).
In biomedicine, communication is limited a great extent to the medical
encounter, which is rich in context and meaning that can hamper effective
communication between the patient and physician. The medical encounter
represents the most important and in many cases the only moment in the
biomedical patient/provider relationship that behavioral information can be
gathered and appropriate behavioral messages can be disseminated. The
clinician brings into the clinical encounter as much social and personal
baggage as the patient. For this reason, it is important to conduct in depth
studies of the clinicians beliefs about and barriers to proper health
communication during the medical encounter. My study does just this by
examining the individual as well as structural barriers clinicians face to
implementing behavioral interventions with their HIV-infected patients.
33


Health Communication in Biomedicine
One of the main aims of this study is to identify and understand the social
and structural barriers to patient-clinician communication and the interrelation
of these barriers in an attempt to improve this interaction. The researcher has
found it useful to look at a health communication model that a clinician created
to describe the different aspects of this communication. This model helps the
current study by illustrating what the different factors are for health
communication to take place between patient and clinician, and where barriers
to this communication are likely to occur.
Diffusion of Innovations is essentially about communication: Diffusion is
a special type of communication that is concerned with messages that are new
ideas (Rogers 2003:5). Trying to change the biomedical model, or how
clinicians communicate with their patients, is essentially a type of social change.
Communication is necessary for social change (Rogers 2003). There are two
points of communication that this thesis is concerned with. The first
communication is between the trainer and the clinician during the introduction
of the innovation. The second point of communication is between the clinician
and the patient during the dissemination of the innovation. Both of these
communication points are important to the success of the public health
intervention.
Health communication is not confined to the patient-physician interaction,
but covers all aspects of the healthcare context in which communication takes
place (Pettegrew 1982). Some of these interactions include the human-machine
interface of health information and the dissemination of community-wide health
messages, as well as the medical encounter with patients. Health
34


communication as a social science field is an important component of the
broader field of health education and training, one aspect of which is the
teaching of communication skills to healthcare practitioners and administrators.
Effective patient-physician communication in the health care context is
essential for the collection of vital patient information and the dissemination of
important behavioral interventions. Effective communication does not,
however, always take place during the medical encounter. This lack of effective
communication is compounded by a lack of recognition of the problem on the
part of clinicians.
When problems arise with others in the health care context, many
practitioners tend to blame personality differences, work pressures, or the
organizational bureaucracy-anything but communication and their role in
making it succeed or fail (Pettegrew 1982:4).
Clinicians hold the key to effective communication with their patients;
however, this incorrect placement of blame on outside factors out of the control
of the clinicians can have an effect on the feelings of self-efficacy that the
clinicians possess about their ability to effectively communicate. Many
clinicians believe that their main role as a medical provider is to provide
biomedical diagnostics and treatment of disease, placing behavioral
interventions low in priority within their practice. Despite this belief, the role of
the clinician involves both the medical diagnostics and treatment of disease as
well as communication of information about the patients condition, personal
support of the patient, and prevention education interventions.
Awareness of interpersonal processes that take place during the clinical
encounter can be useful for overcoming barriers to communication (Stoeckle
1982). To facilitate discussion and understanding of the importance of the
35


interpersonal process in the medical encounter, John Stoeckle (1982) has
outlined a list of medical tasks that clinicians should address during clinical
encounters. This list of medical tasks illustrates the inherent role of the
interpersonal process within the medical encounter:
1. Medical and psychological diagnosis and treatment.
2. Communication of information about illness, diagnosis, and treatment.
3. Personal support of patients.
4. Optimal maintenance of the chronically ill, in which adherence to
treatment is important.
5. Prevention of disease and disability through education, persuasion, and
preventative treatment-health education.
Medical schools tend to focus on the learning of specific skills and
knowledge, much of this knowledge being biological or chemical in nature
(Stoeckle 1982). The main tasks of the clinical encounter involve
communication skills as much, if not in some cases more, than they involve
technical expertise. A strong awareness of the interpersonal process allows the
clinician to thoroughly and effectively accomplish all of the tasks that need to
take place during the clinical encounter. The interpersonal process includes the
clinicians awareness of what they are bringing into the clinical encounter and
communicating to the patient, as well as the patients perspective, requests, and
expectations (Stoeckle 1982). In order to fully communicate with the patient:
The clinician must understand the patients expectations of medical aid;
their attributions, attitudes, and self-treatment of illness; their family,
social networks, and community resources for coping; their personality
style and non-verbal behavior (Stoeckle 1982:13).
When treating HIV positive patients in particular, a firm knowledge of the
patient dynamics above and healthy communication skills are essential for
36


gathering knowledge from the patient about sexual practices and drug use, and
for developing behavioral interventions for the patient.
Every patient is different, which means that every medical encounter will
also be different. The clinician should develop the ability to communicate in a
client-centered manner, instead of relying solely on universal prevention
messages that may not cover the patients intervention needs. Tailored
behavioral interventions are an essential part of patient care, and the delivering
of these interventions demand that the clinician can thoroughly communicate
about a wide range of personal topics with the patient during the medical
encounter.
Interpersonal communication skills involve awareness of many aspects of
the patients life and condition as well as an awareness of the personal biases,
ethics, knowledge, attitude, body language, and appearance that the clinician
brings into the medical encounter. Practitioners need to understand the
interpretation of the nonverbal cues of others as well as how those others form
impressions of the practitioners (Pendleton 1982:22).
The clinician should also recognize structural aspects of communication
that can affect the clinical encounter. These include aspects of the clinical
atmosphere such as decor and dress, physical space, time, location, and
administrative policies (Stoeckle 1982). While the clinician may not have
control over some of these structural aspects of communication, they are
nonetheless responsible in the eyes of the patient for creating a comfortable
atmosphere within these structural confines. Failing to create a physical
atmosphere in which communication can take place can greatly hamper the
interpersonal process.
37


In order to effectively negotiate these personal aspects of communication,
the clinician must first accept their role in the medical encounter as that of more
than a biomedical diagnostician and treatment specialist. The medical training
clinicians receive focuses mainly on learning biomedical facts. Continuing
education and communication training programs are essential for covering
issues on interpersonal communication that are essential for an effective
medical encounter. Communication trainings can only be effective once there is
a full understanding of the barriers and challenges that clinicians.feel they face
in the interpersonal process, including their perception of their role in the
process. A discussion of health communication issues in light of the study
results is accomplished in the discussion section of this report.
38


Highlights and Key Gaps in the Literature
The above literature provides insight for understanding why clinicians
may not implement the current behavioral intervention; the researcher hopes to
close some of the gaps present in this literature with the present study. The
above resources also address the main aims of this study.
The current study differentiated between the social and structural barriers
encountered by clinicians to health communication about behavioral
interventions during the biomedical encounter. This study is filling a key gap in
the current literature regarding this important distinction. A discussion of the
interrelationship of social and structural barriers is also presented. While
barriers clinicians face are separated into social and structural categories in the
findings section, there is a discussion about the interrelationship of social and
structural barriers in the discussion section of this study.
Prevention with persons living with HIV is a relatively novel concept that
has just recently been addressed by the CDC and disseminated to local health
departments. Because of this, the current evaluation of the pilot curriculum
represents one of only a few studies that have been done on this important area
of quality improvement health research. Concepts of quality improvement
research are revisited in the discussion section of this study in light of the
current findings. In particular, constructs from Diffusion of Innovations theory
are explored in the discussion section with regards to the study results of the
barriers that clinicians face to implementation of the guidelines.
39


CHAPTER 3
PRELIMINARY QUANTITATIVE DATA
This chapter consists of quantitative survey results on barriers clinicians
face to implementation of guidelines for implementing behavioral interventions
with HIV positive patients during the clinical encounter. The survey results
provided useful preliminary data of barriers and challenges clinicians face with
implementation of guidelines for conducting behavioral interventions with HIV-
infected patients. This survey data were useful in the development of the
interview guide used in this study.
Methods
The data collected were from participants evaluation forms for a pilot of a
course on Incorporating HIV prevention into the medical care of persons living
with HIV. At each pilot course, staff explained and distributed the surveys and
collected them post-course. An on-line long-term evaluation survey was also
distributed electronically to course participants approximately 3 months after
the course took place.
40


Table 3.1. Evaluation forms used and time of completion
Evaluation Forms Time of comnletion
Participant information form Post-Course & Long-Term Eval. Prior to course
Module 1: Risk Assessment and STD Screening Modi Pre-course form Prior to course in all sites
Modi Post-course form Immediately after Module 1
Module 2: Brief Risk Reduction Interventions Mod2 Pre-course form After Module 1 in Baltimore; Prior to course in other 5 sites
Mod2Post-course form Immediately after Module 2
Module 3: Partner Services and Referrals Mod3 Pre-course form After Module 2 in Baltimore; Prior to course in other 5 sites
Mod3 Post-course form Immediately after Module 3
Long-Term Evaluation Combined survey for all three modules 3-4 months after course completion
Participant Profiles
For this evaluation, only course participants who were clinicians were
included in the data. There were a total of sixty-four clinician participants in the
post-course survey, and twenty-two clinician participants in the long-term
follow-up survey. Both sets of survey participants are comprised mainly of
physicians: 61.7% N=37 of post-course participants, and 66.7% N=14 of long-
41


term participants. Most participants also work in the functional role of care
provider/clinician: 57.1% N=36 of post-course and 66.7% N=14 of long-term.
Some participants work in a different role other than as a clinician, with 4.8%
N=3 of post-course and 9.5% N=2 of long-term participants working in
education/training, 3.2% N=2 of post-course and 4.8% N=1 of long-term
participants working as administrators, and 4.8% N=1 of long-term participants
working as a case manager. Thirty-five percent N=22 of post-course and 14.3%
N=3 of long-term participants did not specify their primary professional role.
Post-course survey participants reported working in their primary profession for
an average of 11.65 years, and long-term participants reported working in their
primary profession for 10.95 years.
The majority of participants work in a hospital or hospital-based clinic
setting, with 68.8% N=44 of post-course and 54.5% N=12 of long-term
participants working in this setting. The rest of the participants work in public
settings including: community or migrant health, community based
organizations, STD/Family planning, community mental health center, and
correctional facility. An exception to this is 6.3% N=4 of post-course and 4.5%
N=1 of long-term participants work in a solo/group private medical practice,
and 1.6% N=1 of post-course and 4.5% N=1 of long-term participants work in a
HMO/managed care setting.
Just over two-thirds of participants in both surveys reported that they
provide all HIV services throughout the course of the disease to their HIV
infected patients, with 68.8% N=44 of post-course and 68.2% N=15 of long-
term participants reporting this.
42


The participants in the interviews for the qualitative thesis represent a
subset of the individuals that participated in the training course as well as key
informants involved in the curriculum development.
Measurement of Variables in the Evaluation Instruments
All quantitative data were entered into a database, managed, and analyzed
using the Statistical Package for the Social Sciences (SPSS), Microsoft Excel,
and Microsoft Word. Descriptive statistics were generated from the evaluation
instrument data, some of which were used to understand the big picture of the
barriers and challenges faced by clinicians in implementing behavioral
interventions with persons living with HIV.
Post-course, and then again at approximately three months after having
completed the course, participants were asked to assess how strongly they
agreed or disagreed about the influence of specific barriers to their
implementation of behavioral interventions during the medical encounter with
HIV positive patients. The participants chose one of four rankings for each
variable: strongly disagree, disagree somewhat, agree somewhat, or strongly
agree. For analysis, categories were collapsed: strongly disagree and disagree
somewhat were collapsed together as were agree somewhat and strongly agree.
43


Findings from the Post-Course Survey
The pilot course was divided into three modules: Module 1 -Risk
Screening, Module 2-Universal Prevention Messages and Tailored
Behavioral Interventions, Module 3-Partner Counseling and Referral
Services. Post-course barriers data were collected at the completion of each,
module, and was collected three times during the workshop. While most
participants completed all three modules, some participants only attended one
or two of the modules. These data allow us to analyze the barriers to
implementation of specific aspects of the guidelines.
44


Table 3.2. Perceived barriers reported in the post-course evaluation.
Please indicate if you think the following will be a barrier to your ability
to use fully the skills and information presented in this workshop.___
Barriers Module 1(M1) Module 2(M2) : Module 3(M3)
Total N=46 Total N=50-52 Total N=45-47
Agree Agree Agree Agree Agree Agree
N % N % N %
The skills are not relevant to my role 8 17% 3 6% 6 13%
I need more training 17 37% 25 48% 18 39%
I need someone with experience to help me use these skills at my workplace ' 11 24% 21 40% 18 38%
Lack of support from other staff I 10 22% 7 : 14% 15 32%
Concerns about privacy and confidentiality 6 13% 7 13% 19 40%
Lack of referral resources if patient has inadequate insurance coverage ; 23 20% 24 47% 17 ; 37%
Lack of time ; 34 74% 39 75% 29 62%
Administrative barriers to change 20 44% 16 32% 14 31%
Staff are uncomfortable with these issues ) 18 39% 12 : 23% 18 ! 38%
Patient perception of stigma from a medical care provider* | N/A N/A N/A N/A N/A ; n/a
*data for this barrier were not collected in the post-course instrument
45


As shown in Table 3.2, in all three modules at the post-course data point,
the majority of respondents felt that a lack of time (Ml :74%; M2:75%;
M3:62%) was a main barrier to implementation of the guidelines. In all three
modules, lack of time is the only barrier that was felt by over 50% of
respondents.
While the rest of the barriers were reported by fewer than 50% of
participants, the frequencies are still high enough in many instances that roughly
one-third or more participants experienced them. A number of clinicians felt
that they needed more training, and this need was felt in all three modules
(Ml:37%; M2:48%; M3:39%). Administrative barriers to change were also felt
by roughly one-third or more participants (Ml :44%; M2: 32%; M3:31%).
Some barriers were felt strongly in one or two modules, but not in all three. In
these instances, one-third or more of participants in one or more modules felt
these issues were barriers to their practice. Many participants in modules 2 and
3 and a moderate number in module 1, felt that they needed someone with
experience to help them use these skills at their workplace (Ml :24%; M2:40%;
M3:38%), and they felt that they had a lack of referral resources if the patient
had inadequate insurance (Ml :20%; M2: 47%; M3:37%). A number of
participants in modules 1 and 3 felt that staff are uncomfortable with these
issues (Ml :39%; M3:38%), while only a moderate amount in module 2 (23%)
felt this way. Concerns about privacy and confidentiality were felt by 40% of
participants in module 3, but by a small amount of participants in the other
modules (Ml:13%; M2:13%). Lack of support from other staff in
implementation of the guidelines was felt most strongly in module 3 (32%), with
fewer participants in modules 1 and 2 seeing this as a strong barrier (Ml :22%;
M2:14%). The vast majority of respondents in all three modules (Ml :83%;
46


M2:94%; M3:87%) felt that the skills presented in the workshop are relevant to
their role as a clinician.
Findings from the Long-Term Survey
In the long-term follow-up evaluation, the barriers survey was presented to
participants as a comprehensive evaluation, and represents the overall feelings
the participants had about all three modules. While direct comparisons between
post-course and long-term data are impossible because of the differing data
collection techniques, general trends can be established.
Lack of time (71% N=17) was sustained as the main barrier to
implementation of the guidelines, and the vast majority of respondents (96%
N=23) still felt that the skills presented in the workshop are relevant to their
role. Only 12% (N=3) of respondents had concerns about privacy and
confidentiality, and only 17% (N=4) felt that staff is uncomfortable with these
issues, both of which show decreases from the post-course data. Fifty percent
of participants believe that they need more training, which represents an
increase from the post-course data for all three modules. There is also an
increase in individuals who feel that they need someone with experience to help,
them use these skills at their workplace (46% N=11), and a slight increase in
participants who believe they lack support from other staff (34% N=8). There is
a slight decrease in participants who feel they encounter administrative barriers
to change (29% N=7). 29% of participants still believe they lack referral
resources if the patient has inadequate insurance coverage. A new barrier was
added to this survey instrument that was not present in the post-course
47


instrument: Patient perception of stigma from a medical care provider. This
barrier was reported by 21% (N=5), or about one quarter, of respondents.
48


Table 3.3. Perceived barriers reported in the long-term evaluation
Please indicate if you think the following will be a barrier to your ability
to usefully the skills and information presented in this workshop.
Barriers Agree
N % i
The skills are not relevant to my role j 1 4% j
I need more training i 12 50% j
I need someone with experience to help I me use these skills at my workplace j 11 46%
Lack of support from other staff 8 34% ;
Concerns about privacy and confidentiality | 3 12%
Lack of referral resources if patient has j inadequate insurance coverage j 7 29% ;
Lack of time j 17 71% j
Administrative barriers to change i 7 29%
\ Staff are uncomfortable with these issues j 4 17%
Patient perception of stigma from a j medical care provider \ 5 21%
Total N=23-24
Discussion of Survey Results
Some useful comparisons between surveys have been made above in the
survey data. These preliminary survey findings affirm that barriers exist and are
highly relevant to effective dissemination of the guidelines. Additionally, these
survey data have provided a comparison to my qualitative thesis results, and has
helped to frame the questions asked in my interview guide. There are a few
49


fundamental limitations to this data. As is the nature of survey data, it is
difficult to understand how the participants are interpreting each barrier in the
survey. The question arises as to how meaningful this survey data can be, given
this limitation. This is one reason why it is important to elaborate upon this
survey work by conducting in-depth interviews with curriculum participants.
As will be discussed in more detail in the results, there are many themes that do
not loom large in the survey data that do so in the interview data. One possible
reason for this is that it is impossible to elaborate views in the survey data,
while the interviews allow a more in-depth and nuanced analytical approach.
Finally, due to data collection issues, it is difficult to make a comparison
between the above data sets. For instance, while the pilot data have a post-
course and long-term component, the post-course data were collected in three
separate modules, and the long-term data in one comprehensive survey.
Participant information is complete for the post-course data, but is spotty for the
long-term set. Thus participant profiles are impossible to compare.
Some comparisons between survey results and qualitative results have
been made in the results section to follow. The preceding limitations hinder
powerful comparisons between the surveys and between the surveys and
qualitative results. The researcher has decided to include these comparisons
keeping these limitations in mind.
50


CHAPTER 4
METHODS
Data Collection
This study examines the barriers and challenges that HIV clinician-
providers face in implementation of guidelines for conducting behavioral
interventions with their HIV-infected patients. My thesis is based on data
collected from interviews that I conducted as part of the larger NNPTC
curriculum development and evaluation process. The pilot curriculum course
took place at six sites in 2005 around the continental United States: Baltimore,
New York, St. Louis, Denver, Oakland, and Tampa. The pilot curriculum was
conducted by the National Network of Prevention Training Centers in
collaboration with the AIDS Education and Training Centers through funding
from the Centers for Disease Control and Prevention in Atlanta, Georgia. The
data collected for this study consist of semi-structured interviews with key
informants involved in the curriculum development process, and semi-
structured interviews with HIV providers who participated in the pilot of the
curriculum. The data were collected over a two-month period during the
summer of 2005.
The interview data were supplemented with data from a preliminary study
that presents survey results on barriers clinicians face to implementation of
guidelines for implementing behavioral interventions with HIV positive patients
during the clinical encounter. The literature resources enhanced the collected
data by giving a background on the theory and preliminary studies used to
initially develop the guidelines and curriculum, and gave insight into the most
recent trends in HIV infection and behavioral interventions. The survey
51


provided useful preliminary data of barriers and challenges clinicians face with
implementation of guidelines for conducting behavioral interventions with HIV-
infected patients. The survey data were useful because the participants came
from the same pilot curriculum study as the participants that completed
interviews for this study.
This study has been designed and the sampling method has been chosen in
order to fill in a key gap in the research as yet conducted on the medical
encounter with persons living with HIV. This study is also designed to
maximize rapid dissemination of the research findings. Critical findings with
important public health implications will be transmitted to the Denver STD/HIV
Prevention Training Center and the CDC for public dissemination and
discussion.
Course Details
The course consists of three one-hour modules:
Module 1: Risk Assessment and STD Screening
Module 2: Brief Risk Reduction Interventions
Module 3: Partner Services and Referrals
The modules for this course are usually done one after another and are
approximately one hour in length. Module 1 consists of information on how to
conduct a risk assessment during the medical interview with a person living
with HIV. This module also addresses the common STDs that are found in
high-risk individuals and how to identify and treat those STDs. Module 2
covers the steps in conducting a brief risk reduction intervention with a person
living with HIV. In this module clinicians are instructed on how to ask
52


questions about high-risk behavior related to sexual behavior and drug use. It is
emphasized in this module that these interventions are to be very brief, lasting
only a couple of minutes, and are often accompanied by follow-up intervention
questions at later medical visits. The clinicians also learn how to create a client-
centered, individually tailored risk-reduction plan for the patient to follow.
Module 3 consists of information on partner services and referrals. Clinicians
are instructed on the different ways that a HIV positive person can disclose their
status to a partner. This module also covers referral resources for the clinician
in case a patient needs more in-depth counseling than what can be provided
during a brief interaction between patient and clinician.
Human Subjects
All research was reviewed and approved by the Human Subjects Research
Committee at the University of Colorado at Denver and Health Sciences Center
on May 19, 2005 (Approval #2005-115).
Sampling
For this study, seventeen participants were recruited. Purposive criterion
sampling was accomplished in order to select participants. This was a
convenient way to select participants that could give a wealth of information on
their experiences in the curriculum development process or participation in the
pilot curriculum. The participants consisted of eleven clinicians involved in the
care of HIV positive patients, three key informants involved in curriculum
development, one key informant involved in clinical training, one key informant
53


who is a Partner Counseling and Referral Services specialist, and one HIV
case manager key informant who works closely with clinicians. The eleven
clinician participants were selected for interviews based on the following
criteria: (1) They currently provide clinical care to HIV-infected patients, (2)
They see five or more HIV positive patients per week, (3) They participated in
the pilot course curriculum (one exception was a clinician who was very
familiar with the course curriculum, but had not actively participated in it).
Some of the key informants were chosen for this study based on their
involvement with curriculum development and dissemination. The PCRS
specialist and the HIV case manager were chosen for this study because they
work closely with HIV providers and because of their professional background
were able to give a unique perspective to the issues.
Of the clinicians interviewed in this study, six were physicians, two were
nurse practitioners, and three were UNs. All of the clinicians interviewed
worked in a public setting, with all working in a health department or public
hospital setting. Ten clinicians worked in a HIV/ID clinic, and one clinician
worked in a STD clinic. All clinicians saw at least five HIV positive patients a
week with most seeing more, and some seeing as many as 40 to 50 patients a
week. Seven of the clinicians were female and four were male.
These clinicians represent high-volume HIV clinical care providers who
have received the pilot course training on, or in the case of one clinician are
knowledgeable about, the clinical guidelines for incorporating HIV/STD
prevention into the medical care of persons living with HIV.
The participants for this study were recruited from the pool of 110
clinicians and non-clinicians who attended one of six pilot training courses on
incorporating HIV prevention into the medical care of persons living with HIV.
54


Sixty-eight of the pilot participants were clinicians. Many of the clinician
participants in the pilot were invited to the event because of their lengthy
experience and high standing in the field of HIV care. These individuals had
the ability to be effective role models for the larger community of HIV
clinicians, and had the ability to give valuable feedback to the curriculum
developers about the training.
About three months after the courses took place, a long-term evaluation
survey was electronically disseminated to all fifty-one clinician participants that
filled out post-course evaluation instruments in the courses. Respondents to this
survey were asked if they could be contacted for an interview about the barriers
to implementation of the guidelines. If the participant responded yes, the
researcher then contacted them. It was then determined, based on the above
criteria, if they qualified for the study. Of the twenty-two clinician participants
that filled out the long-term evaluation forms, eleven of the clinician
participants agreed to a follow-up interview and were later contacted. In
addition, some non-clinician respondents were interviewed as key informants
because they worked closely with HIV clinicians, and could provide a unique
perspective on the questions at hand. Key informants involved in curriculum
development were chosen and contacted by the researcher based on their high
level of expertise and involvement. An interview was either set up in person, if
possible, or over the phone.
The interview process began with an introduction about the project by the
researcher and a discussion of informed consent. Consent was obtained either
by signature on a consent for or verbally. To open up the interview, the
participants were asked a general question about barriers that they face to
implementation of the guidelines for incorporating HIV prevention into the
55


medical care of persons living with HIV. In some of the interviews, probes
were used to gather information about specific barriers.
Data Management and Analysis
Qualitative Data
In depth, semi-structured open-ended interviews were used in order to
better understand the barriers and challenges to implementing behavioral
interventions in the clinical care of persons living with HIV through an analysis
of the subjective experiences of the clinicians. Interviews were conducted with
key informants involved in the curriculum development process, HIV provider
clinicians who participated in the pilot curriculum course, and a HIV counselor
who works closely with clinical staff.
The question guide that was used in the interviewing is found in Appendix
A. Time efficiency is a main concern for this project because clinicians tended
to have little spare time, and preferred to have interviews last only 15-20
minutes. The semi-structured interviewing method was chosen because it
minimizes the variation among interviews, creates a highly focused interview
that is time-efficient, and facilitates analysis by making responses easy to find
and compare (Patton 2002). While the question guide was used in the interview
process, much liberty was used to probe pertinent topics and issues that arose
but were not anticipated when the interview guide was written. Key informants
were routinely asked to respond to barriers topics that clinicians reported in the
preliminary survey and clinician participant interviews, while the clinician
participant interviews were kept open-ended in order to capture the individual
barriers felt by these individuals.
56


Clinicians were interviewed either by phone or, when possible, in person.
All interview participants read and either signed a consent form or gave a verbal
consent. The consent form given to all participants was approved by the human
subjects research committee, as was the option of providing only verbal consent.
The interviews were recorded either using a telephone recording device or a
digital voice recorder. The recorded interview material was then transcribed,
coded and analyzed for emerging themes using Atlas ti.
Coding
The interviews were transcribed verbatim and then loaded into AtlasTi for
analysis. Through thorough reading of the interviews, codes were developed by
the researcher to organize and delineate data by the major themes. All of the
codes were derived directly from the interview data. Some of these derived
codes reflect the barriers investigated in the preliminary study survey, while
others were entirely unique to the interview data. In this way, the researcher
coded the major content areas of the interviews and subsequently developed
more specific subcodes within each content area. Major codes were developed
to encompass the range of responses within the subcodes (Table 4.1.).
Variables such as age, sex and social class were not assumed to have analytic
relevance unless data showed them to be so (Berg 2001:253). The coded data
were then separated into social and structural barriers and analyzed for content.
While this separation was made for the purpose of analysis, much crossover
between social and structural barriers was noted and is reviewed in the
discussion chapter of this report.
57


Probes were used in the interviews where the researcher wanted to obtain
very specific information from the participants. Among the structural barriers,
time, referral resources, and the need for training were for the most part
emergent themes from the interviews that didnt require probing by the
researcher. Participants tended to mention time at the beginning of their
interview, and thus the course of the interview contained information about this
barrier. Referral resources and training needs were discussed by enough of the
participants that these barriers usually did not need to be probed.
Other potential structural barriers did not emerge naturally in the interview
process. In order to obtain data about the nature of some structural barriers, the
themes were the results of probes by the researcher. These included probes
about administrative barriers to change and ICD-9 coding for behavioral
interventions. In all, six of the participants were probed about administrative
barriers and six were probed about ICD-9 coding for behavioral interventions,
but the six probed were not necessarily the same for the two subjects. Probes
for ICD-9 coding were used in order to gain insight into whether monetary
concerns or reimbursement represented a barrier for implementation of the
behavioral interventions. The clinicians were asked, Would it help you to
implement the behavioral intervention if there was ICD-9 coding you could
use? Other probes were used in the interview process to gain further insight
into issues brought up by the clinician during the interview, and were thus
derived from the data. For instance, if the interview participant mentioned not
enough time as a barrier, the researcher would follow with, How much time
do you feel you would need to perform the behavioral interventions? Two
individuals were probed for information about referral resources in their clinic,
58


while many more participants brought the subject into context on their own.
Similarly, the issue of training was not probed for in the interviews.
Content Analysis
The quantitative survey questions created during the preliminary
conceptual and design phases of this study, prior to any fieldwork, were used as
a source to draw from in organization of the qualitative analysis (Patton 2002).
Themes that emerged in the interviews were compared to the quantitative
survey results. Beyond a strict comparison, new analytic insights emerged in
the qualitative data, providing further insight into the beliefs of the HIV
practitioners.
Content analysis was done in order to understand the individual
experiences participants had with the barriers, and to develop patterns among
the participants. These patterns helped to define the parameters of the barriers,
as well as the relationship that the barriers had to one another. Analyst-
constructed codes were created through analysis in order to identify and make
explicit themes that appeared to exist, but sometimes remained unperceived by
the interview participants. These codes, such as HIV stigma help to add depth
to the analysis and also allow for a study of the interrelation of barriers.
59


Table 4.1. Major themes and codes
Major Themes
ir:
A
I


2^i£&*£st siSs
(Social) Discomfort
l- '.V.
^(Social) Rple
mtm
, i
; !
< s

4 Codes
f * ^ *
Discomfort
Behavioral *
Biomedicine
JUde
I.O.S £ \\Uki A^JU

Physical
(Structural) Time
L.
!
(Structural fOther
JFime -
Resources
/Training :
| Structural Barriers. Money/ICD-9 Coding
, * ' . r Administrative-
Unusual Barriers
Frequency o||
... ^Quotes r-, |
86);:
4^m
4iijjp
i8&'\-
69 L"
20 :kf
17 ,
H
7 'Â¥'>

Codingnotdirectly £ '. Pa¥crifffiriettes *. 7 "V "!
related to major themes Participant
profile/stats/demographics
t*£&4jtsge?mwse&M


60


CHAPTER 5
FINDINGS
The themes that emerged from the interview data have been separated into
social and structural barriers for analysis. This differentiation helps accomplish
the first aim of this study: to gain insight into whether the barriers that clinicians
face to implementation of behavioral interventions with their HIV positive
patients are primarily social or structural in nature. Below, an analysis of social
barriers is followed by an analysis of structural barriers.
Social Constraints
Many of the barriers that emerged from the interviews were what I am
referring to as social constraints. These are individual-level and interactional
barriers that are related to the knowledge, beliefs, biases, attitudes, and
communication skills of the clinician, as well as interactional dynamics between
provider and patient. Because many of these barriers are very personal and part
of the individuals social reality, they are difficult to assess, and even more
difficult to change. In many cases, the worldview of the individual is altered in
order for them to personally identify and address these social barriers. While
there is a plethora of social barriers that can affect an individuals behavior, this
study highlights only the barriers for the clinician that are unique and/or
important for health communication between clinician and patient during the
medical encounter.
The clinician can have social barriers related to their views about the
biomedical system and their role as a practitioner in the field, and their beliefs
61


about disease and health especially as it relates to HIV. As well as
preconceived notions about their own professional identity, the clinician can
have beliefs about the role of the patient during the clinical encounter and what
constitutes proper behavior from both parties during this encounter. The
clinician may also encounter barriers that stem from their beliefs about the
causes of risk behaviors, especially in this circumstance when it relates to high-
risk behavior around drug abuse and sexual practices. Class interaction,
especially between clinicians and patients in a public clinic setting, can create
barriers. In the same way, differences in race, sexual identity and gender can
prove to be formidable social barriers. These common social barriers can lead
to a variety of issues that can effect health communication during the clinical
encounter, with most of these barriers manifesting in some form of discomfort
or a challenging of identity. Common issues surrounding social barriers include
provider discomfort, social stigma of HIV, patient discomfort, judgment or bias
about the patient, and difficulty discussing issues around sex and drugs.
Discomfort as a Social Barrier
Discomfort in one form or another was the most common barrier to
implementation of risk discussions that came up during the interviews for this
study. The types of discomfort encountered have been broken down into
several major themes: provider discomfort, judgment/bias, partners of positives,
HIV stigma, sex and drugs, and patient discomfort. These themes represent the
most prevalent issues that the clinicians feel they faced to implementation of
behavioral intervention during clinical encounters. While the preliminary
survey did not ask clinicians what their comfort level was with these issues,
62


17% of long-term survey participants reported that they felt staff is
uncomfortable with these issues.
Provider Discomfort
One of the most commonly stated barriers that arose in the interviews was
provider discomfort. Many clinicians felt discomfort when talking about issues
of sex practices and drug use with their patients. Clinicians also felt discomfort
talking to patients while a partner was in the room, and with discussing risk
behavior directly with the partner of an HIV-infected person. Because of this
reticence, many clinical encounters either lacked a discussion about these
important issues, or contained an abbreviated discussion with little or no follow
up or referrals. Some clinicians were aware of their low comfort level and
openly discussed their concerns in the interviews; nonetheless they encountered
difficulties in overcoming this barrier. Some clinicians identified a discomfort in
general with risk behavior discussions. This discomfort could lead to an
avoidance of the issues altogether.
Sometimes people feel discomfort discussing high risk
behavior, so sometimes they just dont ask patients about this, I
would say that [this is] a main barrier. [PROVIDER]1
Theres plain old failure of will where we dont make it the top
priority because we are excited about people getting healthier,
and it is hard to talk to people about the more delicate issues.
[PROVIDER]
1 At times during the interviews clinicians discussed barriers in the third person, seemingly
excluding themselves. They may feel more comfortable discussing barriers in this manner,
therefore not assigning the issues to them specifically. Alternately, they may have overcome
these barriers, but have witnessed the barriers in other clinicians.
63


Some [clinicians] are just not very comfortable with doing it.
They are learning to deal. [PROVIDER]
One clinician, while able to conduct a thorough risk behavior discussion,
had seen other clinicians truncate the discussion and thus render it meaningless.
Theres the personal comfort level. People need to become
comfortable talking about things that are scary, and it often
gets translated into Heres some condoms, or Are you using
condoms? or a few little things like that that are essentially
meaningless when it comes down to getting people to make
some behavior changes. [PROVIDER]
Sex is considered to be a very private matter by many people and sexual
activities can have strong moral implications. Clinicians at times became
embarrassed talking about sex, or felt that the discussion would embarrass the
patient. Because of this, sex was a difficult topic to broach during the clinical
encounter for some clinicians.
One clinician had observed this among some of the clinicians at her clinic.
Patients up to 21 years of age were treated in this pediatric setting, making a
large proportion of the population sexually active:
For the ones [clinicians] that are uncomfortable, Im not sure if
their morals or ethics come into question about those things.
Or, in talking with one, he said that that was the reason he went
into pediatrics, so that he wouldnt have to talk about sex, but
we have to do that with our population. [PROVIDER]
64


A key informant related this reticence to a cultural norm in our society
about sex:
I think its less the stigma of the disease and more the culture
of we dont talk about sex... I think were embarrassed to
talk about sex, and we think our patients will be embarrassed if
we talk to them about sex... .Sex is just something that people
dont want to talk about, and so they dont and their excuse is,
well they came in with a cough, and they came in with an
ache, and that took all of my 15 min.
[KEY INFORMANT]
A lack of training can be cause of discomfort when discussing these issues
with patients. Training can help clinicians overcome the barriers associated
with lack of knowledge or experience. Training can also help clinicians to
understand the risks and social behaviors that are prevalent in the population
they are treating. In the preliminary survey data, 50% (N=12) of clinicians in
the long-term follow up reported that they felt they needed more training.
Forty-six percent of clinicians in the long-term pilot follow-up felt that they
needed someone with experience to help them use these skills at their
workplace.
Having knowledge of lingo used in the population, knowledge of common
and current drug use and sex practices, and knowledge of the common places
and situations in which sex and drug risk behaviors take place can help the
clinician become comfortable with discussions of risk behaviors and
interventions. In turn, the patient will also feel more comfortable with the
clinician if they feel their provider understands their life situations and risk
behavior. In this way, knowledge is one of the cornerstones of health
communication. In the following interview excerpts, clinicians discussed their
views on knowledge and training:
65


...in my training [in medical school], I cant remember having
anything like a role play or whatever that would make it easier
and make me more comfortable asking questions related to
sexual activity. [PROVIDER]
Another barrier is knowledge, which is part of the comfort
issue [PROVIDER]
Provider comfort [is a main barrier]. A lot of providers arent,
and me to some extent are not comfortable... Most providers
have fairly boring backgrounds and havent been to bathhouses
and so talking to patients is somewhat not totally natural.
[PROVIDER]
One of our... physicians here... didnt know what top or
bottom meant. [That is] the language that you need to be able
to communicate. [PROVIDER]
One key informant identified three aspects of this provider discomfort
around lack of knowledge. This key informant contends that clinicians dont
know how to ask the questions, are concerned that they will receive responses
that they dont know how to deal with, and dont feel they have the resources to
deal with these problems if they do arise. This individual believes that training
is the solution to discomforts related to lack of knowledge: Providers became
more comfortable after they were trained; the big point is that they felt more
comfortable and asked more questions.
Clinicians dont quite feel comfortable about tailored
behavioral interventions; they dont know how to ask the
questions. An even bigger reason why providers that care for
HIV positive patients dont ask those questions is because they
are going to get responses that they dont know how to deal
with. They feel that if they start asking questions they will
open Pandoras box and will find out all of these horrendous
66


things that they are not going to know how to deal with or have
resources to deal with, so [they are] not going to ask.
[KEY INFORMANT]
Many clinicians were also concerned about the reaction that their patients
would have to questions about sex and drug use. This was a reflection of the
clinicians discomfort and not the patients discomfort. As one of the key
informants asserts, studies have shown that HIV patients want to have these
discussions. This individual contends that despite concerns, patients and
providers bond well:
The perception that the patient and clinician will bond poorly is
just that, it is a perception. They [the patients] consider this
part of their medical care and they want to have these
discussions. So, its purely a provider perception that patients
will respond poorly. Its not reality.
[KEY INFORMANT]
However, clinicians interviewed saw this as a very real barrier. One HIV
clinician provider, in referring to other clinicians stated:
They [clinicians] think it will get in their way: My patients
arent going to like it... This discomfort is coming from the
provider and not the patient. [PROVIDER]
Another HIV clinician provider felt the discomfort personally:
Sometimes when I ask these patients, they are just insulted that
I ask them. Im not trying to judge you, I just have to ask
sometimes, and you dont mention these things. But then I
feel terrible and Im less likely to ask the next person.
[PROVIDER]
67


Another key informant was also well aware of this barrier:
Some providers dont want to go there. Its like a mute point.
My patient is already infected; Im going to stigmatize and put
my patient on the spot, and that is going to be an
uncomfortable situation, an uncomfortable conversation...!
hear all the time from our STD course, where clinicians are not
asking difficult questions, are not probing into specific
behaviors... They are trying to find a way to ask these
questions without catching their patient population totally off
guard. The first step is to become personally comfortable.
[KEY INFORMANT]
Some clinicians did not feel personally comfortable with the actions of
their patients, or their own actions and identities. This can create a very real
barrier in which personal discomfort leads to an avoidance of the discussions.
Boredom, anxiety, hunger... thats a little more simplistic than
trying to figure out, why you would do something that would
hurt you or someone else. [PROVIDER]
Most physicians are very awkward about it, there is the
occasional gay physician that is good as long as they are OK
with their own gayness, and if not they are a true mess and
would be better out of the picture completely.
Another one [barrier], at least in this clinic, is class distinction.
We tend to be not from the same class as a lot of our patients;
we tend to be upper class or middle class, so those are real
barriers. [PROVIDER]
68


Judgment/Bias
For some clinicians, the barrier was not so much discomfort with the
issues, but rather personal judgments that they made about the patient. For
some, this meant that they had decided, without asking the patient, that the
patient did not have risk factors such as multiple partners, engaging in risky sex,
or using drugs. This can be a mistake that has grave consequences, since, for
instance, Mark Thrun MD asserts that 60% of MSM who report having a
committed/main partner, also have other partners (personal communication,
June 2005).
... my own biases may contribute to me glossing over the
subject, or not even addressing it in a consistent manner. I
might, for example, make an assumption [about] someone who
has told- me they are not sexually active, I might forget to check
in and see if thats changed. Or I might make an assumption
that when someone says they are monogamous in the past, that
they are still monogamous. I refer to that as my biases that
there is not risk there. [PROVIDER]
One HIV clinician provider grouped patients into two categories, those
who needed interventions and those who didnt. This can lead to missed
opportunities to do necessary interventions with the good group. Changes in
the patients can lead to the need for a behavioral intervention, and some
behaviors are well hidden, causing the need for the provider to look past their
basic judgment and ask the questions anyway.
Most of my patients, especially the ones that need behavioral
interventions, really have a lot of issues. Some patients are on
autopilot and are doing everything they should be. [PROVIDER]
One clinician believed that others had developed a fatalistic view of
behavior change, and thus did not believe that the patients were capable of
69


changing their risk behaviors. A key informant also believed this trend exists
among providers:
Clinicians [believe] that its not going to do any good anyway,
because they are not going to change all of these people.
[PROVIDER]
[Clinicians] Consider it soft [behavioral interventions], that its
not immediately important in that clinical visit. They also
think that its not necessary; what they say may not make an
impact anyway, so why bother that its a pointless exercise.
[KEY INFORMANT]
Partners of Positives (POP)
A few clinicians expressed an increased discomfort when discussing
issues about partners with their patients, or having risk discussions with partners
in the room. In the first quote below, physician discomfort was the main barrier
in these discussions, the clinician also felt hesitant to stigmatize the patient, and
wondered about issues of privacy and legality.
The delicacy issue that I believe is more at the root of the
barrier in terms of clinicians, and I can just talk for me and
other people can agree or not, is the issue of partners. I feel
shy about focusing on the issue of protecting partners because
it makes me feel like I am passing [judgment], accusing...
Basically its not about my patient anymore, but about how my
patient is dirty or broken, or dangerous. That feeling makes it
hard for me to have those discussions with patients... about
disclosure and about partner notification services in particular.
I dont really have a lot of trouble at this point with the sex and
drugs parts, its the parts about partners... The partner
notification section of the curriculum increased my fears on the
part of the patient and made me feel more like this was going
to be a punitive thing. My patients were going to be told that
they were dirty and bad... There was a lot of talk about the law
70


and criminalization... that part of it was not approached with a
good understanding of the audience. What their objections and
concerns were likely to be; those were not addressed.
[PROVIDER]
In this second passage, a clinician was faced with discussing risk
behaviors with the partner in the room, and again expressed discomfort with this
situation, the clinicians discomfort came also from a lack of knowledge of the
population.
Researcher: Im going to have you try and think about a time
that sticks out in your mind when talking about sexual and/or
drug practices with an HIV positive patient was really
challenging.
Respondent: I think that sometimes it can be more challenging
if a partner is in the room; Im thinking of a time when I
brought it up again [risk behaviors] and directed it to both
partners, Are you using condoms and engaging in risky sexual
behaviors?. That was a difficult scenario because they just
giggled and looked at each other, and made some joke that I
really didnt get. That was a situation where there were a few
different things: they were both African American men and so
in some ways it was a different culture. I dont necessarily
know the lingo in either of those cultures. Its not that I would
be expected to use it, but I dont always catch all of the
innuendos that go along with it, and dont always respond to
those innuendos. It was an instance where there was some sort
of underlying something; I didnt really know how to respond
to that scenario. I guess I could have asked the partner to leave
the room, and then asked the questions. [This] would have
probably been a good way to address it in some way, because
neither of them are going to be open about having other
partners with the other person in the room. [PROVIDER]
71


Sometimes family members being present can cause discomfort issues for
the provider:
I had a patient, who was a chronically ill patient, which always
came with his mom. His mom would know better about details
about his condition than he would himself, so I would say
taking a history from him would be very difficult. Having a
relative or significant other in the room would limit your
ability. [PROVIDER]
HIV Stigma
HIV has always been associated with stigma, and these feelings of stigma
can emerge during the medical encounter with the HIV positive patient. There
can be a disconnect between the feelings of the patient and those of their
provider, with the provider feeling there is stigma and the patient feeling
differently, or vice versa. Twenty-one percent (N=5) of clinicians in the long-
term survey believed that patient perception of stigma from a medical care
provider was a barrier to implementation of behavioral interventions. One key
informant remarks:
I think we as providers stigmatize HIV more than our patients
are stigmatized by it. Our patients deal with the fact that they
have HIV, [however] provider stigma around HIV is still there
and we need to educate them. Patient perception of stigma
from a medical care provider is really providers perception of
patient perception of stigma, as opposed to reality.
[KEY INFORMANT]
While the patient is unlikely to feel stigmatized during their medical
encounter, if there is a chance they do feel stigmatized, they are less likely to
seek medical care. Because of this, it is very important that the clinician makes
72


the patient feel comfortable, by feeling comfortable himself or herself with the
medical encounter. The same key informant also states:
When you survey people who are not in care with HIV, its
because of stigma. [KEY INFORMANT]
Some clinicians felt that the stigma came from both the patient and the
clinician; other clinicians asserted that the clinician has the power during the
medical encounter to either foster the stigma or erase it.
I think the stigmatization is definitely there, and I think it
works both ways. [PROVIDER]
Theres stigma, from both sides. [PROVIDER]
I think the provider has the responsibility to make the patient
either feel that stigma, or let them know that they are willing to
talk about whatever. It really comes across in the sexual
history thats taken. If you have providers that are non-
judgmental, that are empathetic and caring, that comes through
very quickly and allows the patient to feel comfortable in
disclosing information. Whereas, if your body language, or
facial features, or the wording that you are using gives off that
you are judging and disapproving [the patient can feel stigma].
[KEY INFORMANT]
I think [the provider] has the responsibility to either foster that
stigma by playing into it, or they can erase it very easily and
very quickly. I see what you are doing is so dangerous or
unhealthy, and since Im not engaging in the behavior I dont
understand how you would do it: thats where the stigma
comes into play when dealing with long-term behavior change.
[PROVIDER]
73


A clinician gave us an account of when they were faced with a situation in
which their patient felt stigmatized by them. This encounter made the clinician
more hesitant to ask risk behavior questions in the future.
I remember one person really got upset with me and said, what
do you think I am?! [I said] Im sorry that was not the
implication of the question I just gotta know so that I can
counsel you on stuff. He was like OK and calmed down. I
felt really bad, and it made me more nervous about asking
about it! Because its put upon them as a judgment statement
that they sleep around, like all gay men sleep around. No!.
Thats not what Im trying to tell you. [PROVIDER]
Patient Discomfort
Some providers felt that patients had discomfort with discussions of sex,
drug use, and the related behavioral interventions. This discomfort can possibly
emerge in the medical encounter for a few reasons. The patient may not feel
comfortable talking about sex and drug use because of the taboo nature of the
topic. Alternately, the patient may have a strong desire to please their provider
by good behavior, and may lie or fail to elicit information about their risk
behaviors. It would be possible for these discomforts on the part of the patient
to lead to a negative reaction by the patient during the encounter.
Some providers felt they had identified ways to overcome these barriers
by creating a comfortable and safe environment, and by not showing negative
reactions to the patients risk activities. While the patient can feel discomfort,
there is a chance that this discomfort is not real, but only perceived by the
clinician. It is impossible to tell this distinction by simply interviewing the HIV
clinician-providers. An in-depth study of the patients should be accomplished
in order to better understand this distinction. Regardless, the clinician can learn
74


how to reduce this discomfort, if it does in fact exist, and improve health
communication during the medical encounter.
... sometimes its things that they would rather not discuss so
much. Certainly its things that they get uncomfortable talking
about and without getting into too much detail they will
attempt to brush off or, side step, or just simply be agreeable
and nod their heads and not offer too much more than that.
[PROVIDER]
I think patients become noticeably uncomfortable when you
ask, do you use a condom every time, what kind of sex do you
have, and who do you have sex with? Sometimes theyre
hesitant to be completely honest. I think theres frequently
hesitancy on the patients part to admit that anything is
currently going on. [PROVIDER]
A key informant commented on this perceived reticence on the part of the
patient to have these discussions, and had some suggestions for providers.
Patients want to have discussions about whats going on in
their lives. They actually want to discuss their relationships
with their doc. They want to have the questions asked and they
are intimidated to bring up these issues with the doc, because
they are not comfortable having these conversations. So if we
as docs do not open up these issues and ask the questions, than
more than likely they are not going to occur. We need to send
the signal that it is OK to answer these questions. When
surveyed, over 90% [of patients] want to have these
discussions. [KEY INFORMANT]
The reticence on the part of the patient to discuss these issues can come
from the fact that they want to please their medical provider. The provider
represents an authority figure in their life. One poignant example of this comes
from a HIV case manager who works closely with patients in order to facilitate
75


adherence to drug therapies and help with life situations the patient may
encounter. This individual had noticed a marked difference in the patients
disclosure of risk behaviors with her/him and with their clinician providers.
I think that it is easier because I am not someone that they feel
they have to answer to because Im not their provider; they feel
that their doctor is their provider. They are more likely to say
to me that they didnt take them [their medications] because
they were out partying all night. They do have close
relationships with their providers, but I think that there is
always that guilt, and they dont feel that with me. They dont
feel any sense of responsibility like they do with their
providers, that they have to put their best foot forward, so they
are much more likely to tell me. They just dont care quite as
much about what I think as they do with their providers,
because they want them to think highly of them. For me, Im
just a ride there and pills or whatever. [HIV CASE MANAGER]
HIV clinician providers also recognized this trend with their patients:
Ill sometimes ask them, so is anything new, new partner,
anything? Sometimes theyll be open about it and sometimes
you can hear the hesitation. Sometimes they want to tell you
the right thing, but they dont want to lie to you. I think they
do want to please you, I think thats a huge thing. [PROVIDER]
As a society we place doctors up on this pedestal thing where,
oh my gosh, if the doc thinks that Im doing something bad!
An example is my mother; my mother will not test her sugar
for anything, (shes diabetic). When she goes into the doctor
she doesnt tell him that shes had a headache or that shes not
been eating so well because she doesnt want him to
disapprove of her behavior. Rather than talking about, Ive
had some struggles doing what you want me to do, she leads
him to believe that all is well. Then I get into the car and hear
for the next 20 min about how her sugar is really not so well.
[PROVIDER]
76


When you ask about their sexual practices, you have to keep
asking, Well, do you always use condoms, this week how
many times did you use condoms, how many times did you
have sex this week, how many of those times did you use
condoms? You have to really push at it, and I dont know if
its because they want to please you and they know that they
are not supposed to [have sex w/o condoms]... They dont
want to really tell you. [PROVIDER]
Patient discomfort can be particularly detrimental when it leads to anger,
which in turn causes the provider to hesitate before asking about risk behaviors
in future encounters.
You cant ask someone directly [questions related to risk
behaviors]... You can, but it puts them very much on the
defensive, like, what do you mean? And some of them still
dont have any insight into it; they dont know why they do it.
[PROVIDER]
However, the negative reaction of the clinician can create the barrier to the
intervention:
I remember watching a client interview... and [the clinician]
was just lecturing, You need to do this, and Im your doctor,
and you need to listen to me and follow my advice because I
know best. Totally not client centered at all, totally not
approaching it in a realistic way. I was just reading the patients
and they were like yeah right and who are you, uh huh. Man,
[the clinician] missed that whole thing completely! [PROVIDER]
While patient discomfort can exist, some providers were able to overcome
this barrier by establishing a safe environment for the patient to discuss difficult
issues, and by being non-judgmental. One provider referred to a study done
77


recently on different cities and the corresponding patient reporting of drug use
and unsafe sex.
One of the things found in the [city name] sample was a lot
more reported drug use and a lot more reported unsafe sex than
other places. I dont think [city name] is any different than
anywhere else, but if the patients are comfortable with their
providers than these conversations will take place. [PROVIDER]
Another provider had developed a unique situation within the medical
community: s/he makes house visits to his/her clients. It allowed him/her to
engage the patient while they were in a setting that they felt they had control
over and were familiar with.
Do you have a private setting?: we establish that before we
leave for the setting. We want to make sure the patient is
comfortable, so we also have another site we have for
counseling or testing. When I go [to the house]... I think for
the most part our clients like it. They talk more freely to us
and they will question things the doctors have said to them in
the clinic that they were scared to ask. Its not threatening and
they are not being rushed. A lot of times, they appreciate you
coming out and sitting down and talking with them about
things that they dont quite understand, or didnt quite get.
[PROVIDER]
Using a client-centered approach to the behavioral intervention can lessen
barriers of discomfort the patient may feel, even in a clinical setting:
I know in our clinic we have had patients leave that are very
surprised that the visit has gone as well as it has. Its very
individualized. [PROVIDER]
78


Sex and Drugs
An interesting theme that emerged from the interviews was the different
comfort levels associated with discussions of sexual activities versus drug use.
Many clinicians had noticed more hesitancy or reluctance on the part of their
HIV positive patients to discuss drug use. This reluctance may stem from the
fact that drug use is illegal, or the patients may feel another as yet undisclosed
discomfort with the issue.
Drug use is actually something that people try to avoid more
than the sexual practices topic. Just thinking about it, there are
instances where someone is known to be using [drugs] through
other sources, and when you ask them about that they will deny
it is actually happening. [PROVIDER]
I think patients are more likely to deny drug practices; most
patients are comfortable being open with sex practices.
[PROVIDER]
Sex, no I dont think Ive had a single patient object to my
asking about sexual practices and I do it a lot. Drugs,
occasionally, more often people of color might see a drug
question as accusatory. Ive had that happen just a couple of
times, especially when I was younger, and now I let people
know that its not shocking to me, and I am concerned.
[PROVIDER]
Sometimes theyre hesitant to be completely honest, less with
sexual behavior than with drug use. [PROVIDER]
.. .most people dont want to admit they are using drugs
because along with that they know that we know it also
involved unsafe sexual activities, especially with meth [use].
[KEY INFORMANT]
79


Another key informant had an interesting insight about stigma and risk
behavior discussions. Again this issue boils down to discomfort on the part of
the clinician associated with discussions of sexual activities. In contrast to the
above quotes, this individual feels that there was more discomfort surrounding
risky sexual activities than there was around drug abuse and mental health.
We dont have this attitude toward other things such as
substance abuse; we ask about substance abuse. The stigma
around asking about other things, such as mental health and
drug abuse, is not as strong as it is around asking about some of
the transmission risk behaviors.
[KEY INFORMANT]
Sometimes, if the patient was expecting questions about sex and drug use,
the discomfort was lessened:
Most patients are expecting that question, Are you currently
sexually active, how many partners do you have, what kind of
partners, what kind of sex do you have? I think they are
almost expecting that, so I feel fairly comfortable asking it.
[PROVIDER]
80


Role as a Social Barrier
Respondents overwhelmingly responded in the preliminary studies that
they felt the skills presented in the workshop were relevant to their role: 96%
(N=24) of clinicians in the long-term pilot survey. Compared to the survey
data, many interview participants responded differently. Some clinicians felt
that behavioral interventions were not an essential part of their role. Others felt
that they should not be solely responsible for the behavioral intervention, but
should instead have a stronger referral network in place. Some clinicians felt
unqualified to perform the behavioral interventions. Many clinicians
interviewed felt that these behavioral interventions were not an essential part of
biomedical care, and were easily trumped in a visit by more pressing medical
issues.
The difference between the survey data and the interview data can be at
least partially attributed to whether the clinicians felt that these behavioral
interventions were possibly relevant but not essential to their role as biomedical
clinicians. The survey asked clinicians if they felt the skills were relevant to
their role as a clinician, while clinicians in the interviews discussed how they
felt the behavioral interventions were not essential to their role as a biomedical
clinician. In order to address this barrier, the curriculum could place more of an
emphasis on the importance of these discussions to the health and well being of
the patients that receive them, thus boosting the argument for why these
interventions are in fact an important part of the role of the clinician.
Some clinicians commented, in general, that behavioral interventions may
not be their primary role as a biomedical provider:
Providers arent really sure that it is their role to implement
behavior change. There is a lack of a perception by providers
81


that it is something important in their scope of practice.
[PROVIDER]
I would say there are two types of HIV [clinicians], some will
only deal with HIV: they dont want to take care of their blood
pressure, they dont want to take care of their diabetes, and
they will send them back to their internist for that. I would say
that probably the majority just take care of it, most of us
remember our internal medicine enough to manage basic
things: most of us can manage diabetes, cholesterol, BP.
Rather than making them go to five appointments, you just
become their primary care doc, and then health maintenance
does start falling into that category. [PROVIDER]
Referral Network Needed
Some clinicians believed that a strong referral network was the solution,
and persons more specifically trained in these approaches should conduct
behavioral interventions. Twenty-nine percent (N-7) of clinicians in the long-
term pilot survey believed that lack of referral resources if the patient had
inadequate insurance coverage was a barrier. This barrier relates back to the
beliefs the clinician feels about their role during the clinical encounter. Many
clinicians felt their role was more that of a biomedical practitioner who treats
physical ailments than a behavioral interventionist. Despite this belief, the
curriculum was developed based on evidence that short behavioral interventions
delivered by clinicians during the clinical encounter are an effective tool for
prevention of HIV transmission risk behaviors among HIV infected patients
(Kamb et al. 1998). Because of this disconnect, there needs to be an emphasis
on the depth of behavioral intervention expected of clinicians.
I dont have time, and sometimes its truly enlightening to have
them [psych or social worker] deal with these psychiatric
82


issues. The psych or social worker will have a very detailed
history of what psychological problems they have, and whether
they have ever been abused... That would have never come out
in a visit... I think if a had a full hour not to discuss the
physical thing, just to discuss their mental stuff, than maybe,
but even then, I dont feel qualified to do a lot of the behavioral
counseling. I think theres a lot of stuff that goes into why they
are doing things and how to help them modify their behavior. I
dont feel qualified to do that, I can tell them to stop, but thats
not really the root of it. Some of them still dont have any
insight into it; they dont know why they do it. I think that it
takes a lot of really intricate training. They may have really
big psychological problems. I mean, the substance abuse plays
into it for a reason. It leads to the behavior, but again, why
they drink so heavily, or why they do drugs, or why they do
anything, I have no insight into it. I can diagnose depression
and thats about it, Im not trained in that. [PROVIDER]
I think if it were either dedicated time or time spent with
someone whose specific focus was going to be sex counseling
or substance use counseling that would be a more effective way
to go about it. I think that people identify in their minds what
the roles are of various people in the clinic or various
practitioners, and when they come to see their physician they
expect that there are certain things that they can talk about.
They may not always relate that physician visit with discussing
some of these other issues. If you had someone designated to
do harm reduction and met with clients depending on their
need, this probably would work much more effectively than
trying to bring it into a busy visit when there are other things
going on. Getting another health care provider involved to
focus more specifically on those issues would be my solution.
[PROVIDER]
We are a Ryan White funded facility so we have the whole
team: we have health educators and even a risk reduction
specialist. I think some of the barriers are historical, that they
[the clinicians] have been doing HIV care for a long time and
83


they feel they should just deal with the STD when they present
symptoms, not ask a whole lot of questions and just move on.
[PROVIDER]
Two key informants felt that the level of behavioral intervention expected
of the clinician needs to be clarified. They felt the more in-depth behavioral
counseling might be more the role of specifically trained behavioral intervention
professionals than the role of the clinician:
They need to bring up the topic; they need to do some risk
assessment. When it comes to the behavior change discussions
and counseling and that kind of thing, they need to then be able
to say, I want you to go talk to this person: a nurse, a case
manager a social worker, a health educator to discuss these
issues so that we can help you be safer, healthier and less
risky. [KEY INFORMANT]
Many of them will admit, and probably even say, well thats
why we have people that are more specifically trained for these
skills, thats why we have PCMs and social workers, health
educators, because thats not my role. And there is some truth
in that because I think we are asking providers to be catchalls
for their patients. Maybe we need to come up with a
management concept of people that we identify that have high
risk behaviors. Maybe we can bring in people that can handle
this level. [KEY INFORMANT]
Another key informant believed that it is in fact the clinicians role, and
they point out that studies have shown clinicians also feel it is their role. Again,
an emphasis needs to be placed in the curriculum on the level of behavioral
intervention expected by clinicians, thereby reducing this disconnect:
I think they believe that it is not their responsibility and this is
poor clinical care because in most clinical practice settings the
clinician is the one who drives behavior change.. .Are the docs
going to dismiss this as unimportant, and the answer is no.
84


actually looked at the data, and docs, nurses, NPs all increased
the number of times they had discussions with their patients
about risk. So training, education, and a little bit of practice
works for everybody. I think if you buy into the fact that
prevention in care should be the standard of care, I think
anybody can be effective at delivering messages.
[KEY INFORMANT]
Not Biomedical Care
In the last interview excerpt above, the key informant asserted that, If you
buy into the fact that prevention in care should be the standard of care, anybody
can be effective at delivering prevention messages. This next group of
interview excerpts point to a different point of view. While many clinicians
surveyed in the preliminary study believed these skills were relevant to their
role, and contended that prevention in care should be the standard of care, some
expressed different beliefs when interviewed. As discussed above, this can be
attributed to clinicians beliefs that these behavioral interventions are relevant,
but not essential to their role. During the clinical encounter, the patient and the
clinician expect certain things to take place, those things being congruent with
the biomedical culture. Within biomedicine, there is a focus on the physical
manifestations and the curing of disease. In this sense, preventative measures
such as behavioral interventions do not have the same level of importance to the
clinician as antiretroviral therapies and pain/sickness issues.
There are a number of different agendas depending on what
other problems the patient might have or what they may want
to talk about during a visit. Sometimes the prevention
message, safer sex issues, and drug use issues get triaged lower
on the list. I think thats one of the barriers, especially with
85


HIV positive patients, you often have a number of issues going
on and things that need to be discussed.
[PROVIDER]
Pain can be a chronic issue that can take up a lot of time,
especially if they are not taking their meds. I need to educate
them about that. Sometimes the patients dont know what
medication they are on, so we have to spend some time going
over that. Honestly, I usually just circle HIV and dont go
through all of the issues, which is something that maybe I
should be doing. [PROVIDER]
In order to change behavior, you have to have discussions
around it. Providers are not interested in spending time on
something that they consider fluff; they dont think that this is
a priority necessarily. They have clinical medical priorities
such as asking about medications and making sure that people
have their viral load and T-cells done. [That is] the line
between what providers consider strictly clinical and what they
consider fluff such as questions around substance abuse... I
think that providers dont prioritize this; they are not willing to
come to a training to learn about these things. They consider it
soft, its not immediately important in that clinical visit, they
also think that its not necessary, and that what they say may
not make an impact anyway, so why bother that its a pointless
exercise. [KEY INFORMANT]
Its always on my problem list, but it doesnt necessarily mean
that I really address anything. Most of the time I address
immunizations or colon screening, or pap or prostate; sort of
more standard basic things... What I call health maintenance
stuff gets thrown to the wayside because they have so many
active issues that you forget. You need to switch their regimen
and you need to spend you time counseling, and then you sort
of think about it later, like ugh, are they sleeping with
someone new? It doesnt dawn on you really to think about it
at that moment, because you are dealing with active things.
86


This may be active too, but its not necessarily a priority in
terms of what we are doing. [PROVIDER]
... youre concentrating on a couple different things, and
youre telling the patient, OK, put your hands above your
head, take a deep breath, hold your breath, swallow, (laughs),
all that. I think in that scenario that is where you get this, lets
really not focus on that or have that sort of conversation.
[PROVIDER]
Maybe there are queues like that someone has moved, or ended
a relationship, that would lead me to talk about these things,
but Im not disciplined to do this at every visit with every
patient. Im not disciplined to do healthcare maintenance at
every visit... [The patient] was having physical manifestations
of stress, and he wanted support, but I dont have the time, or
clinical skills to counsel him. Hes a MEDICAL patient, so I
really didnt have a counselor that I could plug him in to, or
make a commitment to find people for him to work with. What
I could not do for this man is feel that I connected him with a
clinician that could provide insight and skill to deal with this
situation that he said is killing him....(laugh).. .in a way that
the HIV is secondary! [PROVIDER]
87


Structural Constraints
In the context of this study, structural constraints are those that are
elements of the biomedical system and society at large, and the bureaucratic,
administrative elements thereof. While social constraints discussed earlier in
this study are represented by the personal beliefs and feelings of the
practitioners that lead to barriers to health communication, structural constraints
represent the larger material structures of society and more specifically the
biomedical system creating barriers for the clinician. Once the material or
structural changes to the biomedical system are made, it is much easier for the
clinician to follow through with individual behavioral changes.
Structural factors such as the amount of time a clinician is allotted to
spend with each patient, insufficient referral resources, and lack of opportunities
for training in health communication and behavioral interventions were strong
constraints for HIV clinician providers interviewed for this study.
Administrative barriers existed in the form of a lack of policy for consistent
implementation of behavioral interventions. Some clinicians felt they faced
monetary barriers in the form of a lack of reimbursement for their time and
services for performing behavioral interventions, but most felt that this was not
a barrier for them.
88


Time as a Structural Barrier
To begin the interviews, the researcher began with some version of the
following question: What do you think the barriers are to discussing sexual
practices and drug use with your HIV positive patients? Nine out of the eleven
HIV providers, the PCRS specialist, and two of the four key informants
involved in training and curriculum development mentioned time first in the
interview. In most of these cases, time as a barrier was mentioned in the first
sentence. Seventy-one percent (N=l 7) of clinicians in the long-term pilot
survey felt that lack of time was a barrier for them. This indicates that time was
a very strong and real barrier for these participants, and the barrier that
immediately came to mind for them. Participants were asked at some point
during the interview how much time they had per visit, and the times ranged
from 15 min to 30 min per patient, with most providers having 30 min for each
patient.
Researcher: What do you think the barriers are to discussing
sexual practices and drug use with your HIVpositive
patients?
One is probably time for providers. [PROVIDER]
My own personal barrier would be time. [PROVIDER]
On busy appointments when you have a lot to do, you dont
have time to talk or test for STDs: its usually lack of time.
[PROVIDER]
Time pressures are a barrier to most things. [PROVIDER]
89


Full Text

PAGE 1

BARRIERS TO CLINICIANS' IMPLEMENTATION OF BEHAVIORAL INTERVENTIONS RELATED TO SEXUAL PRACTICES AND DRUG USE OF PATIENTS LIVING WITH HIV by Christine Lenore Shure B.S., University of Missouri-Columbia, 1997 A thesis submitted to the University of Colorado at Denver and Health Sciences Center in partial fulfillment of the requirements for the degree of Master of Arts Anthropology 2006 r .J

PAGE 2

This thesis for the Master of Arts degree by Christine Lenore Shure has been approved by Susan Dreisbach }j uv.,_;, I o) Date

PAGE 3

Shure, Christine Lenore (M.A., Anthropology) Barriers to clinicians' implementation of behavioral interventions related to sexual practices and drug use of patients living with HIV Thesis directed by Professor Kitty K. Corbett ABSTRACT Clinicians' medical encounters with HIV positive patients represent a key but underutilized moment for discussing ongoing HIV and STD transmission risk. This thesis explores the main barriers that health care providers face in implementing behavioral interventions related to sexual practices and drug use with their HIV positive patients. These barriers to thorough and appropriate health communication are assessed in terms of being primarily social or structural in nature. In a project with CDC's National Network ofSTD/HIV Prevention Training Centers, semi-structured interviews with key informants (N=6) and HIV clinician providers (N=ll) provided data for the study. Among the social barriers, various forms of clinician discomfort with discussions surrounding sexual practices and drug use was the most frequent barrier. This discomfort stemmed from a lack of knowledge on the subject matter, the stigma of HIV, the possibility of a negative reaction from a patient, difficulty dealing with the partners ofHIV positive patients, and a reticence to cause the patient discomfort with these discussions. Conflicting beliefs about the role of the clinician as a behavioral interventionist was also frequently mentioned as a social barrier. Many clinicians felt that a more comprehensive referral network was needed to help them with these behavioral interventions. A number of clinicians also felt it was not the role of the biomedical clinician to perform behavioral interventions during the medical encounter. A lack of time was mentioned as a structural barrier by most of the participants interviewed, and looms as one of the largest barriers to implementation of the guidelines. Other structural barriers included

PAGE 4

administrative barriers to change, monetary reimbursement concerns, a need for more training, and the need for more comprehensive referral networks The five perceived attributes of innovations within Diffusion of limovations theory was used in order to analyze the barriers found in this study. Of the five attributes a lack of relative advantage looms large for this innovation, with some barriers also existing in compatibility with existing beliefs, observability, trialability, and complexity Quality improvement studies, health communication issues, and the interrelation of social and structural barriers are also discussed. This abstract accurately represents the content of the candidate's thesis I recommend its publication Signed

PAGE 5

DEDICATION Ibis thesis is dedicated to all first generation college students and their families.

PAGE 6

ACKNOWLEDGEMENT I would like to thank the Departments of Anthropology and Health and Behavioral Sciences for their support with my degree and this thesis. I would especially like to thank Kitty Corbett for her many words of wisdom that contributed to my thesis and to life in general. Finally, I would like to thank my husband, Joel Shure, for his undying support in my studies.

PAGE 7

TABLE OF CONTENTS l'ables ................................................................................ . .... xi CHAPTER 1. INTRODUCTION ....................................................... ........ 1 Objectives ............................ : ..................... : ................ 1 Aim 1 ................... ........................... . ..................... ........... 1 Aim 2 .......... ......... .... ................................................ . . ..... 2 Background ................................................................. 2 Significance ........... .... ................. .................................... 4 2. REVIEW OF THE LITERATURE ............................... ........... 8 HIV and the Biomedical System ......................................... 8 Prevention with Persons Living with HIV: Efficacy Studies ........ ..................................................... 9 Training to Overcome Missed Opportunities for HIV Prevention ...................................... ........................... 11 Barriers to Use of Behavioral Interventions in Practice ............................................................ 15 Quality Improvement Theory and Research ........................... 19 Diffusion of Innovations ....................................... 20 Quality Improvement Studies in Health Care ............. ... 26 vii

PAGE 8

The Anthropology of the Medical Encounter .. ...................... 30 Biomedicine as Ethnomedicine ................................ 32 Health Communication in Biomedicine ............................... 34 Highlights and Key Gaps in the Literature ..................... ..... 39 3. PRELIMINARY QUANTITATIVE DATA ............... ..... ..... .40 Methods .... ............................................................. .. 40 Participant Profiles ......................................... .... 41 Measurement of Variables in the Evaluation Instruments ........................................................ 43 Findings from the Post-Course Survey ...... ...................... .... .44 Findings from the Long-Teim Survey ............................. . .47 Discussion of Survey Results ....................................... . .49 4. METHODS ...................................................................... 51 Data Collection ........ ......... ........ .................... .. ..... .. 51 Course Details ................................................... 52 Human Subjects ................................................. 53 Sampling . ........................... .................... ........ 53 Data Management and Analysis ........................................ 56 Qualitative Data .............................................. .... 56 viii

PAGE 9

5. FINDINGS ..................................................................... .. 61 Social Constraints .............................................. ......... 61 Discomfort as a Social Barrier ................................. 62 Role as a Social Barrier ......................................... 81 Structural Constraints .................................................... 88 Time as a Structural Barrier .................................... 89 Other Structural Barriers: Administrative Monetary (ICD-9 codes), Training, and Referral Networks ........... 94 6. DISCUSSION ................................................................ 1 09 Introduction ............................................. : ................ 1 09 The Interrelation of Structural and Social Barriers ................. 110 The Role of Health Communication ................................. 113 Quality Improvement Studies Revisited .............................. 114 An Analysis of Results through Diffusion of Innovations Theory .................. ........................... 116 7. CONCLUSIONS ............................................................. l25 Implications of the Project. ....................................... .... 125 Recommendations ............................................. 126 Limitations of the Data ................................................ 127 Future Directions and Research Needs .............................. 129 ix

PAGE 10

APPENDIX A. INTERVIEW QUESTION GUIDE ........................................................................ l31 B. PRELIMINARY QUANTITATIVE DATA: PARTICIPANT PROFILE TABLES ......................................................... l32 BIBLIOGRAPHY ......................................................... : .............. 136 X

PAGE 11

LIST OF TABLES Table 3.1. Evaluation fonns used and time ofcompletion ................................. 41 3.2. Perceived barriers reported in the post-course evaluation .................... 45 3.3. Perceived barriers reported in the long-term evaluation ..................... .49 4.1. Major themes and codes ........................................... : ................ 60 B.l. Gender of course participants ....... : ............................................ 13 2 B.2 Profession of participants ......................................................... 132 B.3. Primary professional or functional role of participants ................... .. 133 B.4. Number of years practicing in their reported primary profession ........... 133 B.5. Facility/setting ................................................ ..................... ... .............. 134 B.6. HIV/AIDS services .............................................................. l35 Xl

PAGE 12

CHAPTER 1 INTRODUCTION Objectives HIV clinicians face many barriers to thorough health communication with their patients. This is especially true when the health communication involves discussions of behavioral risk factors related to sexual practices and drug use . Despite the barriers faced by HIV clinicians these conversations should take place during the medical encounter in order to help control the spread of HIV. These conversations are especially important since as many as one in three HIV -positive people continue unprotected sexual practices after learning they are HIV infected (Kalichman et al. 2001) This project has two aims: Aiml First, this thesis is designed to gain insight into whether the key barriers that clinicians face to implementation of behavioral interventions with their HIV positive patients are primarily social (individual level or interactional) or structural (system level or institutional) in nature. Identification of structural and social barriers in prevention conimunication is necessary to design effective interventions that are responsive to these barriers. It will also shed light on the bigger issue of patient physician communication within the biomedical system. Identifying barriers and challenges to implementation of the national guidelines will also help the National Network of STD/HIV Prevention Training Centers (NNPTC) better address the training needs of their course participants 1

PAGE 13

Aim2 Second, this thesis is designed to gain a more detailed understanding of the identified barriers and challenges that medical providers who serve persons living with HIV face in following national guidelines for incorporating behavior change interventions for STD and HIV prevention into medical visits with their HIV positive patients. This increased understanding is accomplished through the use of elements of Diffusion of Innovations theory as a framework for the discussion of the thesis results. The detailed understanding of these social and structural barriers will assist in implementing the changes that need to take place in the biomedical encounter in order to better facilitate these discussions. Background In 2003, the Centers for Disease Control and Prevention (CDC), the Health Resources and Service Administration (HRSA), the National Institutes of Health (NIH), and the HIV Medicine Association of the Infectious Disease Society of America (IDSA) put forth a list of recommendations for medical care providers in order to better incorporate HIV prevention into the medical care of persons living with HIV: Screen for transmission risk behaviors and sexually transmitted diseases, provide brief behavioral risk -reduction interventions in the office setting, refer selected patients for additional prevention interventions and other related services, and facilitate notification and counseling of sex and needle-sharing partners of infected persons (CDC 2003a). These recommendations stimulated the creation of a curriculum to train clinicians in dissemination of the guidelines during the medical encounter with 2

PAGE 14

persons infected with HIV. Following the CDC/HRSA/NIH/IDSA guidelines, in 2004 and 2005, the National Network of STD/HIV Prevention Training Centers (NNPTC) participated in the development of a "Prevention in Care" curriculum to train medical care providers ofHIV positive patients. The course aims to train HIV care providers to incorporate science-based intervention methods in order to help their patients reduce risk behaviors, even in highly constrained settings for care. The NNPTC is a network of ten regional centers funded by the CDC to train clinicians to diagnose, treat, manage, and prevent transmission of STD' s in the United States. As mandated by the CDC, NNPTC courses are regularly evaluated by an independent organization. The current pilot training curriculum, Prevention with Persons Living with HIV, has been evaluated by the Center for Research in Health and Behavioral Sciences at the University of Colorado at Denver and Health Sciences Center. This evaluation consisted of pre-course, post-course, and long-term survey data that allowed for an assessment of changes in knowledge, practice patterns, and attitudes of course participants. Some of this survey data were presented as a preliminary study in the review of literature section of this report. This preliminary data helped to guide the development of this current qualitative study, and has allowed for some comparisons between results. This thesis represents an addition to the existing Prevention with Persons Living with HIV pilot course evaluations. The NNPTC Evaluation Committee has endorsed my project; it will enhance their evaluation capacity beyond what their funding allows, and will address questions that they would like answered about whether their trainings make a difference to real world practice settings. 3

PAGE 15

The 'Prevention in Care' training course has been piloted and evaluated at six U.S. sites (Denver, St. Louis, Tampa, Baltimore, New York, and Oakland) during the spring and summer of2005. Participants who agreed to be part of long-term follow-up evaluation at a post-course time point were later contacted. When the curriculum is finalized, a CDC-approved version of the course and its evaluation materials will be utilized in trainings of several thousand clinicians at over 10 sites around the country. I was able to contact pilot course participants who agreed to provide long-term follow-up evaluation informati.on for semi structured interviews. The researcher conducted in-depth, semi-structured interviews with some of the clinicians who participated in the pilot of the curriculum. Topics in the interviews included barriers to implementing behavioral interventions during the medical encounter with HIV positive patients, the self-efficacy of the clinician in clinical encounters with HIV infected persons, clinicians' beliefs about patients' attitudes and willingness to change, and brief background information on the clinical setting and the clinician (See Appendix A). Significance The HIV virus was first recognized in 1981, and since then the virus has become responsible for a global AIDS pandemic. The HIV virus and AIDS are an unprecedented threat to global health. With the advent of antiretroviral drugs in the 1990s and the subsequent improvement of these treatments since then, people can live indefinitely with HIV. The antiretroviral drugs are, however, not the answer to the AIDS epidemic. Many citizens of underdeveloped 4

PAGE 16

countries do not have access to antiretroviral drugs, and suffer terribly from the burden of AIDS on their population. In the United States, while most individuals have access to antiretroviral therapies, not all take the drugs properly or agree to take the drugs at all. A staggeringly high number of HiV infected individuals are unaware of their viral status. At the end of2003, an estimated 1,039,000 to 1,185,000 persons in the United States were living with HIV/AIDS, with 24-27% undiagnosed and unaware of their HIV infection (Glynn 2005). The vast majority, of these infections were caused by sex and drug use practices, with male-to-male sexual contact, injection drug use, and heterosexual contact being the largest risk factors. Through 2003, only a little over 20,000 cases ofHIV in the US were caused by other risk factors, which include hemophilia, blood transfusion, perinatal, and risk not reported or identified (CDC 2003b). Most of these alternate risk factors have been greatly reduced in recent years with the implementation of better HIV testing methods for donated blood and better techniques for reducing perinatal infection. Currently HIV infection rates reflect primarily sex and drug : use behavioral risk factors. For this reason, the identification, control and modification of these risk factors is the key to control of the HIV/AIDS epidemic in the United States. As with any other epidemic, in order to control transmission of the disease, social or behavioral factors relating to its transmission must be addressed. In the U.S., the majority of persons who are aware of their HIV status are receiving medical care (NNPTC/AETC 2005:6). Because of this, the medical encounter with persons living with HIV represents the most important and in many cases the only moment in the biomedical patient/provider relationship in which behavioral information can be gathered and appropriate 5

PAGE 17

behavioral interventions can be disseminated. Various studies have shown that brief counseling sessions and short prevention messages may change patient behavior and reduce risky behavior (Thrun et al. 2003, Richardson et al. 2004, Fisher et al. 2002, NNPTC/ AETC 2005). Studies have also shown that most HIV care providers do not ask their patients about ongoing HIV transmission risk (Elford et al. 2000, Bull et al. 1999, Gardner et al. 2002, Duffus et al. 2003, NNPTC/ AETC 2005) Biomedical clinicians face many challenges in communication with their HIV positive patients. Their communication is affected by contextual factors and by personal and cultural meanings, as wen as both parties knowledge and attitudes that can hamper effective communication between patient and physician. The fact that HIV is highly stigmatized in society can also affect communication during the medical encounter. The confidence or self-efficacy of the physician in dealing with topics such as sex and drug use can affect what questions are asked and whether behavioral interventions are discussed during the medical encounter. Structural barriers present in the clinical setting can also affect communication between patient and clinician. Time pressures, administrative and political policies, and resources available to the clinician can all affect health communication. This is especially true when the clinician is under rigid time constraints or lacks adequate support from the administration to implement behavioral interventions. In order to effectively address the social modes of transmission of the HIV virus, researchers must identify the best ways to communicate HIV risk behavior information to at risk populations. While considerable effort has been: spent on aiming prevention messages toward persons not yet infected with the virus, the Centers for Disease Control and Prevention have recently identified 6

PAGE 18

HIV positive persons as an additional important target audience for prevention messages (CDC 2003a). In response to this, training programs have been developed by prevention training centers to facilitate in the dissemination of this important intervention. Studies have also been done to test the efficacy of dissemination of this intervention. Clinicians and researchers can have different goals for research of the clinical encounter (Stoeckle 1982). The researcher may be looking for regularities in clinician communication and the correlates that them. The clinician is more concerned with what works in care, and may be more interested in the language and questions that get the most information about disease. While these goals are not necessarily incongruent, they certainly can be. The insights from researchers should be useful so that practitioners can apply them in health care (Pendleton 1982). Keeping this in mind, this project was designed to produce outcomes that can be immediately useful to both HIV practitioners and curriculum developers. 7

PAGE 19

CHAPTER2 REVIEW OF THE LITERATURE This review of the literature covers important aspects of curriculum development and dissemination as they relate to HIV prevention interventions. The first section covers background information on HIV prevention programs and intervention dissemination leading to the goals of this study: identification of social and structural barriers to patient-clinician communication and an analysis of these barriers. Following this review ofbackground literature are three sections that relate to important goals of this study. The first of these sections is a review of quality improvement studies that help with the understanding of why clinicians may or may not implement an important intervention. Diffusion oflnnovations theory is reviewed in order to build a theoretical framework for detailed analysis of the social and structural barriers identified in this thesis. Next is a section titled 'The Anthropology of Biomedicine,' which is an analysis of the biomedical encounter through an anthropological perspective. This section provides an insightful anthropological perspective of the medical encounter. Finally, health communication in biomedicine is reviewed. This section looks at a model developed by a physician to frame the important aspects of health communication during the biomedical encounter. This section helps to shed light on the dynamics of patient-clinician interaction during the medical encounter. HIV and the Biomedical System In order to understand the barriers that clinicians face to implementation of the behavioral intervention s that they learn in the current curriculum, a 8

PAGE 20

review of past efficacy studies, studies on training, and studies on the barriers that clinicians face is reported below. Past efficacy studies ofHIV intervention counseling interventions have supported the development of the current curriculum based on findings that a series of brief counseling sessions can be an effective tool for disseminating HIV prevention messages (Kamb et al 1998, Thrun et al. 2003, Richardson et al. 2004, Fisher WA 2002) . A discussion of these efficacy studies is followed by a brief review of recent training studies related to STD/HIV prevention. These studies provide a useful to the present curriculum being disseminated, and show the past effectiveness of training as a means of disseminating behavioral interventions. A discussion then follows of the barriers that clinicians face to implementation of behavioral interventions. This discussion provides background for why clinicians may be experiencing the barriers that are outlined later in the findings of this thesis. Finally, a review of literature on HIV -related stigma provides background on this important barrier to communication. Prevention with Persons Living with HIV: Efficacy Studies Efficacy is an important measure because it shows the actual impact of an intervention on the individuals that received it. Efficacy studies are many times the first step in the creation of public health programs, followed by research into the potential dissemination of the innovation or its effectiveness. 9

PAGE 21

Project RESPECT In order to test the efficacy of brief counseling sessions to prevent infection with the HIV virus, a study was done by Kamb et al. (1998) to compare two interactive HIV /STD counseling interventions against didactic prevention messages typical of current clinical practice. One group of participants received enhanced counseling, one received brief counseling sessions, and two other groups received didactic messages similar to those currently found in clinical settings. The study was conducted at five public STD clinics around the US. Participants consisted of 5758 heterosexual, HIV negative patients aged 14 years or older who carne to the clinics for STD examinations. Reports showed that self-reported condom use was higher in both the enhanced counseling group and the brief counseling group than it was in either didactic message groups. At each of the five study sites, STD incidence was lower in the counseling intervention groups than in the didactic messages intervention groups. The researchers, Kamb et al. (1998), reached a few important conclusions. First, the study showed that brief counseling sessions during the clinical encounter could be as effective as enhanced counseling sessions that involve more time and multiple visits. The study also showed that because these effective counseling sessions were brief, they could easily be incorporated into the medical encounter with patients and implemented by clinicians, even in a busy public clinic environment. Finally the study showed that the incorporation of a client-centered approach to the brief counseling session, in lieu of more canned didactic messages, was a more effective method of reducing behavioral risk factors. 10

PAGE 22

Recent HIVISTD Efficacy Studies More recently, studies have been conducted to assess the effectiveness of clinician-initiated risk reduction counseling with HIV positive patients (Thrun et al. 2003, Richardson et al. 2004, Fisher WA 2002). Clinicians have only enough time during the clinical encounter to provide brief behavioral interventions Interventions presented to HIV clinicians should be time sensitive as well as effective. Richardson et al. (2004) did a study on the effect of brief safer-sex counseling by HIV medical providers, which tested the behavioral outcome of 585 HIV -positiveyersons. Two intervention arms were compared to control groups. One intervention arm received brief counseling emphasizing the negative consequences of unsafe sex, while the other intervention arm received brief counseling emphasizing the positive consequences of safer-sex. Unprotected anal and/or vaginal intercourse was reduced by 38% (P
PAGE 23

Morin 2002) Although clinical settings may provide opportunities for risk discussions, this potential is not always being fulfilled (Elford et al. 2000). Elford et al. (2000) did astudy with gay men in central London to determine if these men were discussing HIV risk reduction with their general practitioners. One thousand and four gay men who utilized one of five gyms in central London completed anonymous surveys for the study. The results showed that fewer than one-third of the men (30%) had discussed safer sex with a general practitioner. Of those who had discussed safer sex their general practitioner, one in six (17%) had found it difficult to do so. These results are worrisome since about one-quarter of respondents reported unprotected anal intercourse in the three months prior to the survey time. In a study of six public HIV clinics in California conducted by Marks et al. (2002), HIV -positive patients were asked if clinic providers ever talked with them about safer sex or disclosure oftheir HIV status. Overall, only 71% of patients surveyed reported that a clinical provider had ever talk;ed with them at least once about safer sex. Interestingly, men who have sex with men (MSM) and whites were less likely than other racially identified groups to receive HIV prevention messages. In light of the fact that MSM represent an important transmission risk group, these results are disturbing. Training in the skills and knowledge necessary to carry out behavioral interventions can help providers to eliminate many of these missed opportunities. Thrun et al. (2003) showed that both providers and patients respond favorably to curricula encouraging providers to incorporate prevention counseling into HIV care. In this study, a curriculum developed for incorporating HIV prevention into HIV care was taught in a workshop attended by HIV care providers who worked in a large urban HIV care clinic At the end 12

PAGE 24

of two years most providers believed that the training helped them feel comfortable initiating risk-reduction discussions. clinicians reported that their patients responded positively to the risk-reduction discussions. Henderson et al. (1999) evaluated nine HIVIAIDS training programs to assess the degree to which the programs produced positive healthcare systems changes to the delivery of HIV I AIDS care. This study is important because systems changes, or structural changes to the biomedical system, are necessary in order to implement new policies regarding HIV I AIDS delivery. About half ofthe training participants (55.5%) were able to provide at least one example of a systems change that had resulted from the training, with trainees having an average of 1.03 system change examples (Henderson et al. 1999). While changes in knowledge and skills of the individual healthcare practitioners can be readily achieved and measured, systems changes resulting from HIV I AIDS training can be more difficult to achieve and to assess. This important study shows significant system changes resulting from training and thus makes training an effective tool for this difficult purpose. Huba et al. (2000) and then Lalonde et al. (2002) assessed the long-term impact of HIV I AIDS education and training on clinician practice. These studies looked at the long-term training effects on general HIVIAIDS perspectives that clinicians have, health care provider service provision, and changes in procedures and operations at the health care system level. In this way, they assessed the individual changes as well as the overarching structural changes within the biomedical system as it pertains to HIVIAIDS care. Telephone interviews were conducted with 218 trainees 8 months after training. Generally, respondents rated the training experience in all three areas assessed as positive. Sustained changes resulting from the training were seen. There was increased 13

PAGE 25

knowledge on the part of the clinicians at the patient and system levels, and positive changes in provider-patient interactions. Some felt increased comfort, which is instrumental in fostering better health communication. Interesting is the fact that the system functioning domains were most positively associated with training, even more so than the patienf-care domains. This study, like the Henderson et al. (1999) study above shows that, while system changes resulting from HIV I AIDS training can be more difficult to achieve and to assess, these changes are in fact happening as a result of effective HIV I AIDS traip.ing. Closer to home, the Center for Research in Health and Behavioral Sciences (CRHBS) conducted a long-term course evaluation of the Denver STD/HIV Prevention and Training Center's 3-day STD/HIV Intensive Clinical Course (Devine et al. N.d). Clinician-participants' long-term retention of knowledge, skills, and practice patterns from the training were assessed. The study assessed participants who took the course between May 2001 and November 2004, thus spanning 3.5 years of courses, and data on about 100 participants for the pre-course assessment and around 60 participants for the post-course assessment. In the knowledge section, clinicians were given five case scenarios and posed questions on them. At six months post-course there was a significant and sustained 12.5% improvement in knowledge. Self reported skills improved and were sustained at six months in three categories: communication, technical, and diagnostic. At six months post-course, communication skills showed a 22.3% sustained improvement, technical skills showed a 54.0% sustained improvement, and diagnostic skills showed a 67.6% sustained improvement. Sustained improvements in knowledge and skills are easier to attain than sustained improvements in practice patterns. This is because, in general, behavior is difficult to change. In this study, practice 14

PAGE 26

patterns showed a slight improvement in most categories assessed, although most of these improvements were not significant. The researchers concluded that sustained improvements in practice patterns proved difficult to achieve. Despite this setback, this study has shown that training is effective at increasing and sustaining knowledge and skills among clinicians, which is the first step to changing provider behavior and practice patterns. In summary, while research has shown that prevention discussions are not taking place as often as they should, it has also shown that and training can positively affect many of these processes. The processes that can be positively affected by training include those on the individual provider level as well as systems changes in the broader biomedical context. Barriers to Use of Behavioral Interventions in Practice Training has been shown to be a valuable tool for changing provider practice patterns and attitudes toward behavioral risk reduction discussions with HIV -positive patients. While training has proven to be valuable, barriers to use of the learned knowledge in practice continue to exist. Knowledge is easier to influence than performance, which is easier to influence than patient outcomes: Programs influencing only predisposing factors, such as knowledge, attitudes, and skill, are less effective in improving performance than those influencing predisposing, enabling, and/or reinforcing factors (Umble and Cervo 1996: 167). While increased knowledge and skills are necessary to create practice changes, these changes will not take place if self-efficacy barriers or structural barriers in the clinical environment persist. 15

PAGE 27

Sometimes the most significant barrier to providers' use of skills in practice is their never having learned the skills to begin with. The AIDS Education and Training Centers (AETC) did a study to determine the characteristics of 'hard-to-reach' providers. Three key themes emerged which identified who was hard to reach: professional group designation, providers with low volumes ofHIV-infected clients, and HIV-expert clinicians who serve high volumes of HIV -infected clients (Bradley-Springer 2003 :28). Bradley-Springer et al. (2003) found that there were many reasons cited by these groups as to why they were unable or unwilling to receive training. Some of these issues were practice related, which included the stresses and limitations of a busy practice, not enough time to attend, the expense of education, and lack of administrative support. Isolation was also a barrier to training for these providers. Some were located in a rural location, were culturally isolated possibly because of a special minority group that they serve, or were even isolated in an urban area with lack of transportation or in a high-crime neighborhood. HIV and Stigma Illness can become a way that society labels a person; this is especially true for illnesses that are stigmatized. Stigma allows a disease to stamp powerfully disconfirming cultural meanings onto the sick person (Kleinman 1988). The sick person internalizes the stigma, further affecting them by mentally altering their illness experience Stigma can also affect the human rights of the individual, resulting in property loss, bodily injury or murder, loss of marriage rights and loss of access to healthcare as well as possibly leading to incarceration. Because of the powerful potential of the stigma of disease, illness 16

PAGE 28

takes on a vast array of psychological and societal meanings that are reflected in the illness narratives of the affected individual. HIV/AIDS since its very beginnings in the 1980's has been an extremely stigmatized disorder (RCAP 2005). ' HIV -related stigmatization is a process that also reinforces existing social inequalities based on race, gender, ethnicity, and sexual orientation" (Aggleton in RCAP 2005: 1 ). Individuals infected with HIV have been denied employment, housing, medical care as well as having numerous violent attacks against them Because of these social responses, individuals can face discrimination because of the violation of social norms. For instance, "fear of being blamed for violating sexual norms (i.e. heterosexual monogamy) can lead to non-disclosure to sexual partners if someone is infected" (RCAP 2005:1) Some countries such as Cuba and Vietnam have actually "rounded up" HIV positive individuals into camps to try to stop the spread of infection. In this way, HIV infection has become a reason for incarceration, thus creating extreme stigma against the illness. In many instances, HIV as no exception, illness becomes viewed as a type of cultural 'punishment' for a sin or evil acts that the affected individual must have performed. Religions and governmental systems can perpetrate this idea, and thus further ostracize the affected individuals. The cultural idea of "illness as punishment" further limits the credibility of the affected individual and allows for justification of human rights abuses. As an example, HIV is an extremely stigmatized disorder in Vietnam. HIV infection is considered a social evil that is punishable and worthy of discrimination. The government as well as the general public hold this belief. Because of these cultural beliefs, HIV prevention campaigns are geared around incarcerating the HIV infected as well as ending the evil' behaviors, such as 17

PAGE 29

drug use and prostitution that lead to HIV infection. Unfortunately, these government tactics have proven to be ineffective at stopping the spread of HIV, and the virus continues to spread at an alarming rate every year in Vietnam The stigma of this disease keeps individuals from seeking the proper medical care to treat their infection, thus making their illness worse. In the case of HIV in Vietnam, stigmatization has created a situation in which individual human rights have been violated, thus creating inequity .in care of sickness. In this type of inequity, certain diseases are 'acceptable' and treated likewise, while other diseases are stigmatized and individuals are denied treatment. While the United States has not implemented a formal policy against HIV positive individuals, civil rights violations are common and the disease remains a social stigma. Because of the negative social implications of HIV I AIDS, the medical encounter with persons living with HIV is a unique and especially difficult encounter within biomedicine. While any clinical encounter is full of personal bias, beliefs, and emotions possessed by the patient as well as the clinician, the medical encounter with HIV positive individuals is made more difficult with the addition of stigma associated with the condition. In a study done by Deborah Preston et al. (2004) on risk behaviors of rurally based men who have sex with men, stigma was found to be an associated risk. Stigma was predictive of higher risk behavior, and is thus a major factor with implications for lowered health status (Preston et al. 2004). Stigma can negatively affect an individual's self esteem and in many situations reduces social status (Goffman 1963). The beliefs and attitudes that clinicians bring into the medical encounter can have a stigmatizing effect on the patient. This can be a formidable barrier to patient-clinician communication. Alternately, a hyperawareness of potential stigma on the part of the clinician can create a 18

PAGE 30

situation during the medical encounter in which difficult topics are not breached lest stigmatization of the patient might happen. Both of these situations can be avoided if the clinician uses proper communication skills. An intervention that does not effectively address the stigma of HIV infection will be less useful in the treatment and prevention of individuals in need. Regardless of the type of barriers that providers face to using new skills in their practice, work can be done in order to overcome these barriers and to realize the full potential of trainings. This can be done by careful analysis of the barriers that clinicians face, and the revising of curricula to fully address these barriers. Quality Improvement Theory and Research The current study represents a quality improvement study as it is intended to improve health communication between providers and HIV positive patients, as well as help to identify and alleviate the current barriers to implementation of behavioral interventions with these patients. It is therefore important to review the relevant literature on quality improvement research and theory. Diffusion oflnnovations theory developed by Everett Rogers (2003) provides a useful theoretical framework for quality improvement research by analyzing why clinicians may or may not adopt the behavioral interventions proposed in this study. The five attributes of innovations outlined by Diffusion of Innovations theory are discussed as a useful tool for analyzing the results of this study. These sources of literature provide a useful background for understanding how the current study can contribute to the improvement of the curriculum it evaluates. 19

PAGE 31

All too often, clinical research that demonstrates the efficacy of a novel clinical process or innovation is not widely adopted and practiced by the target clinical community. There is a chasm between the knowledge of efficacious clinical guidelines and processes and the actual clinical practice of these healthcare improvements. Because of this gap, the successful implementation of new clinical guidelines for practice into the existing clinical environment should involve careful attention to the translation of research into practice. Without this last step, the value of the research is lost. Through quality I improvement analysis, my research attempts to understand why HIV providers who have learned valuable and efficacious behavioral interventions for their HIV -infected patients may not integrate these interventions into their practice Through this research, I hope to strengthen an already existing curriculum by analyzing data collected from the target community after they have received the intervention. In this way I hope to help close the gap between the clinical research that went into developing this intervention and the clinical practice of these guidelines. Diffusion of Innovations Robert Sanson-Fisher et al. (2004) assert that the behavioral sciences can make a substantial contribution to the process of changing providers' behavior. Behavioral change theories should be used to develop theoretical frameworks that explain how behavioral change in the clinical setting is achieved. Accurate measurement of relevant clinical behavior should also take place in order to properly evaluate the effectiveness of an intervention. This can be accomplished through direct observation of clinic activities, as well as 20

PAGE 32

examination of the accuracy of self-report by the healthcare providers. Finally, Sanson-Fisher et al. (2004) argue that there should be an in-depth examination of the incentives and barriers that clinicians face to implementation of guidelines, and thorough evaluation of the intervention. One behavioral theory that has proven useful for an in-depth examination of the incentives and barriers that clinicians face, and is utilized in this current study, is Diffusion oflnnovations Theory developed by Everett Rogers (2003) Diffusion of Innovations Theory is a useful theoretical base in ql,Ullity improvement research because it is instrumental in identifying and explaining why a certain intervention is or is not being adopted, who is adopting the intervention, and how it is changed during the process of being adopted by the community (Schoenwald and Hoagwood 2001 ). Everett Rogers notes that an assumption is often made that after an innovation is developed, and its efficacy and effectiveness demonstrated, widespread adoption and uptake will occur automatically (Rogers 2003). This is an assumption that can lead to a failure of adoption and failure to sustain use of proven efficacious and effective interventions. Much time and money can be wasted on the development of innovations that never get fully realized in their target community. Diffusion of Innovations Theory helps to identify the predictors of adoption as well as the barriers to adoption of an innovation (Rogers 2003). Diffusion of Innovations Theory proposes that the attributes or characteristics of an innovation most likely to affect the speed and extent of the adoption and diffusion process can be identified (Rogers 1983, Oldenberg and Parcel 2002). It provides a useful paradigm for understanding how new interventions are introduced and adopted within the medical system. The theory consists of various constructs, which combine to form the overarching doctrine. 21

PAGE 33

These elements include: the definition of diffusion of innovations, the innovation-decision process, attributes of innovations and their rate of adoption, adopter categories, diffusion networks, and change agent . While these elements combine to provide a clear and thorough picture of all aspects of the innovation process, the constructs can be used separately to understand specific aspects of an innovation. The researcher has utilized the basic concepts of diffusion as well as Roger's Attributes of Innovations in order to analyze the results of this study. Rogers defines diffusion as: the process by which an innovation is communicated through certain channels over time among members of a social system, with the aim being to maximize the exposure and reach of innovations, strategies, or programs (Rogers 2003). Rogers' above definition of diffusion contains crucial elements in the diffusion of new ideas. There are four elements of this definition as is defmed and elaborated upon below. (1) Innovation-an idea, practice, or object that is perceived as new by an individual (Rogers 2003:12). The innovation need not be a new idea to the individual, but the attitude, knowledge ab9ut, and/or decision about action regarding the idea must be new. In the present study, the innovation is the guidelines for conducting behavioral interventions with HIV positive patients. The concept is not new, although the recent emphasis paid to this form of patient-provider communication is new. (2) Communication Channels these are the means by which messages get from one individual to another (Rogers 2003: 18). In the case of the intervention in this study, the communication channels would be the training course as well as the clinicians 22

PAGE 34

that have undergone the training. The patient-clinician medical encounter is also a communication channel. The concepts of health communication discussed later in this review of literature are pertinent to an understanding of the clinician as a 'communication channel'. (3) Time this involves the rate of adoption for the individual as well as the larger social system. Rogers has developed mathematical formulas that can help to predict this rate of adoption. While the rate of adoption is an important concept, the present study was not designed to explore this area of diffusion theory. ( 4) Social this is a set of interrelated units that are engaged in joint problem solving to accomplish a common goal (Rogers 2003:23). In the case of the present study, the units are individuals; and in some sense the units can also be the clinics or even health organizations. Diffusion research is probably best known for facilitating an understanding of the individual by assigning them to different adopter categories. It is also a valuable tool for analyzing characteristics of innovations. In this study the researcher has used the latter construct in the analysis of the study results. The Five Attributes of Innovations Everett Rogers has outlined five attributes of innovations that can help to explain the rate of adoption. The attributes are: relative advantage compatibility, complexity, trialability, and observability. The researcher has explored how these attributes help to explain the barriers that clinicians report they encounter during implementation of the guidelines. The relative advantage of an intervention and its compatibility with existing ideas tend to be the most 23

PAGE 35

important attributes of innovations when considering their ability and rate of adoption. Complexity, trialability, and observability do also play a role in the adoption of many innovations. While barriers related to these attributes may not be present in an innovation, if the potential adopters perceive them as being present it is detrimental to the adoption process. Relative advantage. Relative advantage is defined as, "the degree to which an innovation is perceived as better than the idea it supersedes"(Rogers 2003: 15). There are subdimensions of relative advantage which include economic profitability, low initial cost, a decrease in discomfort, social prestige, a saving oftime and effort, and immediacy of reward (Rogers 2003: 223). An innovation will be adopted more quickly if all or most of these subdimensions are advantageous. An innovation may not be adopted at all if there are large barriers of this nature. Compatibility. Compatibility is defined as, "the degree to which an innovation is perceived as consistent with the existing values, past experiences, and needs of potential adopters" (Rogers 2003: 240). Compatibility ofthe innovation reduces uncertainty with the advantages of adoption and fits well with the individual's current situation. Rogers asserts that an innovation can be compatible or incompatible with sociocultural values and beliefs, previously introduced ideas, and/or client needs for the innovation (Rogers 2003: 240). When introducing an innovation to a new culture, one can experience reduced adoption because of conflict with deeply held religious or cultural beliefs. / 24

PAGE 36

Innovations are also more easily adopted when they are familiar or compatible with previously held ideas. Rogers asserts that, "old ideas are the main mental tools that individuals utilize to assess new ideas and give them meaning" (Rogers 2003: 243). When an innovation is very compatible, there is a reduced need for behavior change. Furthermore, when an innovation supersedes the previous idea or institution, there is more incentive to adopt. Finally, successful adoption of an innovation requires that the innovation meets a felt need of the individual. In the case of prevention inn<;>vations, the individuals may not feel a need for the innovation, thus making it more difficult for the change agents to implement. Complexity. An innovation should be easy to understand and use. Training courses should consist of short, interactive and experiential sessions with straightforward concepts and simple intervention messages. The potential adopter can feel frustrated if the innovation is too complex, or they may not remember the important elements of the innovation. Tria/ability. An innovation tends to be more successful if it can be experimented with on a limited basis (Rogers 2003) Moreover, if an innovation can be re-designed through trial by the adopter in order to fit individual needs or comfort, than it has a better chance of being adopted. Observability. This relates to the degree to which the results of an innovation are visible to others (Rogers 2003). In the case of prevention innovations, the observability of the outcome is not always visible. HIV/AIDS 25

PAGE 37

prevention is one example of this, especially prevention with positives in which the outcome is the prevention of HIV transmission to partners of HIV positive individuals. The five perceived attributes of innovations are revisited in the discussion section of this study in light. of the barriers that clinician participants report they have when attempting to the guidelines for incorporating behavioral interventions into the medical encounter with patients living witl;l HIV. Quality Improvement Studies in Health Care The ultimate goal of quality improvement studies in health care is to improve health outcomes. One major aspect of this literature focuses on the improvement of health care provider trainings in order to optimize the chances of desired practice changes taking place. Although there is a wealth of research on the efficacy of clinical guidelines, there is continuing evidence of the failure to translate clinical research findings into existing practice. As a result of this, it is thought that 30% to 40% of patients do not receive treatments of proven effectiveness (SansonFisher, Grimshaw, and Eccles 2004). Many times the wrong choice of intervention is made. For example, Lisa Bero (1998) found that the most common method of intervention is passive dissemination of information, such as a conference publication or mailing of educational material, despite the fact that this method has been shown to be generally ineffective and only results in small practice changes (Bero et al.l998). 26

PAGE 38

She asserts that: the choice of intervention should be guided by the evidence on the effectiveness of dissemination and implementation strategies, the characteristic of the message, the recognition of external barriers to change, and the preparedness of the clinicians to change (Bero et al.1998:5). In order to tackle the difficult topic of changing provider practice patterns and behavior Wyszewianski and Green (2000) have developed a framework for improvement of training focused on changing clinicians' practice patterns. Although most traditional clinician training focuses on knowledge, Wyszewianski and Green assert that there is a need to also have behavior oriented training especially for hard-to-reach providers. While the knowledge oriented training involves an evidence-based approach, namely by disseminating information through continuing education and guidelines, the behavior-oriented approach adds a focus on general behavior change strategies. These strategies include rewards, strong incentives for positive behavior change, and the removal of obstacles to change The researchers conclude that while this framework still needs to be empirically tested, the broader that it proposes will most likely prove useful. In a quality improvement study of methodology, a systematic review of 14 continuing medical education activities (CME), Davis et al. (1999) found that the CME courses that relied on didactic sessions, in contrast to CME sessions that involved participant activity and practice of skills were not as effective at changing physicians' performance. From these results the researchers contend that methodologies involving only didactic sessions should receive less credit and funding than fully integrative and interactive sessions. This study shows 27

PAGE 39

that the methodology of delivery is as important to successful dissemination of the intervention as the information being covered in the intervention. In a quality improvement analysis, Bero et al. (1998) did an assessment of systematic reviews of interventions and found that many of these reviews were not assessing important factors such as cost effectiveness of the intervention and the successful reaching of target audiences. In this sense, not only are many interventions missing their target audience, the reviews of the effectiveness of these interventions are also missing valuable opportunities for critique and improvement . In a review of the implementation of clinical practice guidelines, Davis et al. ( 1997) looked at various factors that effect the adoption of guidelines They found that the adoption of guidelines was affected by certain qualities of the guidelines, characteristics of the health care professional, characteristics of the practice setting, regulations, and patient factors. Davis et al. used Everett Rogers' Diffusion of Innovations theory to explain why these types of factors would affect adoption. If the clinical guidelines coincided with and represente d only a small change to the existing beliefs and values of the clinician and patient, and represented only a small change to the existing structural aspects of the clinical environment, they were much more likely to be adopted. The researchers contended, "the adoption of any innovation or the dissemination of new medical knowledge should be considered in a holistic, contextual manner" (Davis et al. 1997:6). Interactive guideline dissemination such as community based interventions and practice-based interventions were found in this study to be more effective than didactic passive dissemination of guidelines. Traditional didactic lecture-based CME were least effective and reminder systems, academic detailing, and multiple interventions were most effective, with audit 28

PAGE 40

and feedback also being a useful tooL These results mirror later results found by Davis et al. (1999) and Wyszewianski and Green (2000) described earlier in this section in relation to CME and training activities. Cabana et al. (1999) did a study analyzing why clinicians fail to follow clinical practice guidelines. They assert that little is known about how physicians change their practice methods in response to guidelines, and that in many cases guidelines have little effect on changing physician behavior. In an. analysis of the guideline-related barriers reported in numerous studies from the mid sixties through 1999, Cabana et al. (1999) found two important trends. First, physicians were more likely to report that guidelines were difficult to follow when asked about them in theory than they were when asked about specific guidelines What this tells us is that one of the crucial steps for guideline adoption for clinicians is getting the clinicians to at least try the guideline. Secondly, the researchers found that it was easier for clinicians to follow guidelines that recommended adding a new behavior than it was for them to follow a guideline recommending the elimination of an established behavior. This basic trend holds true for human behavior change in most settings, and therefore it is not surprising that it was also found to be true with clinician practice change. The principles of Diffusion of Innovations theory once again come into context, with guidelines being more successful that require less drastic changes. Davis et al. contend that there are "complex intrapersonal, interpersonal, and professional educational variables that affect the physician-learner" (Davis et al. 1999:873). The quality improvement studies described above reflect this assertion. What these studies above show is that in order to be fully effective, quality irriprovemi:mt research should cover all aspects of the intervention 29

PAGE 41

delivery process, including the method of delivery, barriers faced by the clinician-participants in implementation, and the reviews of the processes themselves. Covering one aspect of possible quality improvement research, my research focuses on the barriers clinicians face to implementation of behavioral interventions. Since these behavioral interventions are based on guidelines, a focus on quality improvement research related to guideline dissemination is in order. This study does just this, utilizing Diffusion of Innovations theory as a framework for analyzing results in this study. The Anthropology of the Medical Encounter A review of anthropological approaches to the medical encounter is provided below. Medical anthropology has helped the researcher to understand the social and structural intricacies of patient-clinician interaction during the biomedical encounter. The review of medical anthropology in this section also provides some background for understanding the origins of Diffusion of Innovations theory. Compatibility with the cultural beliefs of the social system is important for adoption of innovations (Rogers 2003). The earliest roots of diffusion research are in anthropological studies, and current anthropological research commonly contains elements of diffusion theory. The communication approach [found in diffusion research] is utilized by a number of disciplines, including anthropology, the products of which are seen in the government communication institutes, such as the national institutes of health and the centers for disease control (Rogers 2003). Anthropological theory allows for an in depth analysis of the material or physical aspects of humans and human systems as well as the nonmaterial 30

PAGE 42

aspects of culture, behavior and consciousness. Within anthropological theory, materiality is assumed to be only one aspect of a larger system (Hahn and Kleinman 1983 :308), and social relations are simultaneously a material and an ideational reality (Godelier in Hahn and Kleinman 1983). Medical anthropology, in particular, studies human health systems in their broad social and cultural contexts (Barfield 1997). There are three fundamental premises of medical anthropology that are outlined by Thomas Barfield in his Dictionary of Anthropology .(1997), which make this discipline useful for gaining a comprehensive understanding of current health communication issues within the biomedical encounter. First, illness and healing are best understood holistically, in the contexts of human biology and cultural diversity . Illness and disease cannot be fully understood in the context of biomedicine alone, because this context is focused primarily on the physiological aspects of illness. Second, disease is both influenced by human behavior and requires biocultural adaptations. Across the board, through numerous studies spanning several decades, the social sciences have shown this premise to be true. Finally, and perhaps the most important premise for my present study, is the assertion that "the cultural aspects of health systems have important pragmatic consequences for the acceptability, effectiveness, and improvement of health care" (Barfield 1997 :316). The successful diffusion of innovations can only take place once the cultural aspects of health systems are understood Health communication between patient and clinician depends upon this understanding. 31

PAGE 43

Biomedicine as Ethnomedicine In a medical anthropology text from the 1970's, ethnomedicine is defined as "the beliefs and practices relating to disease which are the products of indigenous cultural development and are not explicitly derived from the conceptual framework of modem medicine" (Bauwens 1978). This assumption that biomedicine is somehow separate from its surrounding culture is untrue. As with any other system of medicine, biomedicine is a product of the culture that created it, and is thus a sociocultural system. In current definitions of ethnomedicine, biomedicine is always included. Biomedical theory emerged out of positivist notions that the world, or human conditions, can be explained fully through materialist or concrete theory (Waitzkin 1991). Within this framework, nonmaterial events such as consciousness and culture hold little or no weight (Hahn arid Kleinman 1983). Health care, however, is a context or culture within which all health communication takes place (Pettegrew 1982). In the U.S., this context is primarily biomedical, and thus to understand healthcare in the U.S. one should have a firm understanding of the cultural.foundations of the biomedical system. Although biomedicine sees itself as a fully scientific discipline that is separate from morals, ethics and the social it is in fact just the opposite. Biomedicine is an ethnomedical system that is created out of social values and norms and operates within these limitations. It is subject to the same biases as any other sociocultural system, and thus these biases, beliefs, and actions should be analyzed. The medical encounter, being & social relation, should be assessed not only as a biomedical diagnostic tool, but also as a human social interaction. A wealth of studies spanning all of the social science disciplines 32

PAGE 44

and decades of research have shown that human behavior is one of the main detei:minants of health outcomes. Health simply cannot be achieved when the consciousness, actions, and social setting of the individual is not taken into account. The medical e?counter inevitably includes a world of sociocultural issues, from the patient as well as the clinician, that need to be properly acknowledged and managed in order for healthcare to be successfully administered: The [medical] interview is not simply an interpersonal communication skill or technique, but a meeting of persons full of emotions, attachments, ideologies, values, rights, duties, politics, and conflicts (Billings and Stoeckle 1999:ix). In biomedicine, communication is limited a great extent to the medical encounter, which is rich in context and meaning that can hamper effective communication between the patient and physician. The medical encounter represents the most important and in many cases the only moment in the biomedical patient/provider relationship that behavioral information can be gathered and appropriate behavioral messages can be disseminated. The clinician brings into the clinical encounter as much 'social and personal baggage' as the patient. For this reason, it is important to conduct in depth studies of the clinician's beliefs about aQd barriers to proper health communication during the medical encounter. My study does just this by examining the individual as well as structural barriers clinicians face to implementing behavioral interventions with their HIV -infected patients. 33

PAGE 45

Health Communication in Biomedicine One of the main aims of this study is to identify and understand the social and structural barriers to patient-clinician communication and the interrelation of these barriers in an attempt to improve this interaction The researcher has found it useful to look at a health communication model that a clinician created to describe the different aspects of this communication. This model helps the current study by illustrating what the different factors are for health communication to take place between patient and clinician, and where barriers to this communication are likely to occur. Diffusion of Innovations is essentially about communication: "Diffusion is a special type of communication that is concerned with messages that are new ideas" (Rogers 2003:5). Trying to change the biomedical model, or how clinicians communicate with their patients, is essentially a type of social change. Communication is necessary for social change (Rogers 2003). There are two points of communication that this thesis is concerned with. The first communication is between the trainer and the clinician during the introduction of the innovation. The second point of communication is between the clinician and the patient during the dissemination of the innovation. Both of these communication points are important to the success of the public health intervention. Health communication is not confined to the patient-physician interaction, but covers all aspects of the healthcare context in which communication takes place (Pettegrew 1982) Some of these interactions include the human-machine interface of health information and the dissemination of community-wide health messages, as well as the medical encounter with patients. Health 34

PAGE 46

communication as a social science field is an important component ofthe broader field of health education and training, one aspect of which is the teaching of communication skills to healthcare practitioners and administrators. Effective patient-physician communication in the health care context is essential for the collection of vital patient information and the dissemination of important behavioral interventions. Effective communication does not, however, always take place during the medical encounter. This lack of effective communication is compounded by a lack of recognition of the on the part of clinicians. When problems arise with others in the health care context, many practitioners tend to blame personality differences, work pressures, or the organizational bureaucracy-anything but communication and their role in making it succeed or fail (Pettegrew 1982:4). Clinicians hold the key to effective communication with their patients; however, this incorrect placement of blame on outside factors out of the control of the clinicians can have an effect on the feelings of self-efficacy that the clinicians possess about their ability to effectively communicate. Many clinicians believe that their main role as a medical provider is to provide biomedical diagnostics and treatment of disease, placing behavioral interventions low in priority within their practice. Despite this belief, the role of the clinician involves both the medical diagnostics and treatment of disease as well as communication of information about the patient's condition, personal support of the patient, and prevention education interventions. Awareness of interpersonal processes that take place during the clinical encounter can be useful for overcoming barriers to communication (Stoeckle 1982). To facilitate discussion and understanding ofthe importance ofthe 35

PAGE 47

interpersonal process in the medical encounter, John Stoeckle ( 1982) has outlined a list of medical tasks that clinicians should address during clinical encounters. This list of medical tasks illustrates the inherent role of the interpersonal process within the medical encounter: 1. Medical and psychological diagnosis and treatment. 2. Communication of infoimation about illness, diagnosis and treatment. 3. Personal support of patients. 4 Optimal maintenance of the chronically ill, in which adherence to treatment is important. 5. Prevention of disease and disability through education, persuasion, and preventative treatment-health education. Medical schools tend to focus on the learning of specific skills and knowledge, much of this knowledge being biological or chemical in nature (Stoeckle 1982). The main tasks ofthedinical encounter involve communication skills as much, if not in some cases more, than they involve technical expertise. A strong awareness of the interpersonal process allows the clinician to thoroughly and effectively accomplish all of the tasks that need to take place during the clinical encounter. The interpersonal process includes the clinician's awareness of what they are bringing into the clinical encounter and communicating to the patient, as well as the patient's perspective, requests, and expectations (Stoeckle 1982).In order to fully communicate with the patient: The clinician must understand the patient's expectations of medical aid; their attributions, attitudes, and self-treatment of illness; their family, social networks, and community resources for coping; their personality style and non-verbal behavior (Stoeckle 1982: 13). When treating HIV positive patients in particular, a firm knowledge of the patient dynamics above and healthy communication skills are essential for 36

PAGE 48

gathering knowledge from the patient about sexual practices and drug use, and for developing behavioral interventions for the patient. Every patient is different, which means that every medical encounter will also be different. The clinician should develop tlle ability to communicate in a client-centered manner, instead of relying solely on universal prevention messages that may not cover the patient's intervention needs. Tailored behavioral interventions are an essential part of patient care, and the delivering of these interventions demand that the clinician can thoroughly communicate about a wide range of personal topics with the patient during the medical encounter. Interpersonal communication skills involve awareness of many aspects of the patient's life and condition as well as an awareness of the personal biases, ethics, knowledge, attitude, body language, and appearance that the clinician brings into the medical encounter. "Practitioners need to understand the interpretation of the nonverbal cues of others as well as how those others form impressions of the practitioners" (Pendleton 1982:22). The clinician should also recognize structural aspects of communication that can affect the clinical encounter These include aspects of the clinical atmosphere such as decor and dress, physical space, time, location, and administrative policies (Stoeckle 1982). While the clinician may not have control over some of these structural aspects of communication, they are nonetheless responsible in the eyes of the patient for creating a comfortable atmosphere within these structural confmes. Failing to create a physical atmosphere in which communication can take place can greatly hamper the interpersonal process. 37

PAGE 49

In order to effectively negotiate these personal aspects of communication, the clinician must first accept their role in the medical encounter as that of more than a biomedical diagnostician and treatment specialist. The medical training clinicians receive focuses mainly on learning biomedical facts. Continuing education and communication training programs are essential for covering issues on interpersonal communication that are essential for an effective medical encounter. Communication trainings can only be effective once there is a full understanding of the barriers and challenges that clinicians.feel they face in the interpersonal process, including their perception of their role in the process. A discussion of health communication issues in light of the study results is accomplished in the discussion section of this report. 38

PAGE 50

Highlights and Key Gaps in the Literature The above literature provides insight for understanding why clinicians may not implement the current behavioral intervention; the researcher hopes to close some ofthe gaps present in this literature with the present study. The above resources also address the main aims of this study. The current study differentiated between the social and structural barriers encountered by clinicians to health communication about behavioral interventions during the biomedical encounter. This study is filling a key gap in the current literature regarding this important distinction. A discussion of the interrelationship of social and structural barriers is also presented. While barriers clinicians face are separated into social and structural categories in the findings section, there is a discussion about the interrelationship of social and structural barriers in the discussion section of this study. Prevention with persons living with HIV is a relatively novel concept that has just recently been addressed by the CDC and disseminated to local health departments. Because of this, the current evaluation of the pilot curriculum represents one of only a few studies that have been done on this important area of quaiity improvement health research. Concepts of quality improvement research are revisited in the discussion section of this study in light of the current findings. In particular, constructs from Diffusion of Innovations theory are explored in the discussion section with regards to the study results of the barriers that clinicians face to implementation of the guidelines. 39

PAGE 51

CHAPTER3 PRELIMINARY QUANTITATIVE DATA This chapter consists of quantitative survey results on barriers clinicians face to implementation of guidelines for implementing behavioral interventions with HIV positive patients during the clinical encounter. The survey results provided useful preliminary data ofbarriers and challenges clinicians face with implementation of guidelines for conducting behavioral interventions with HIV infected patients. This survey data were useful in the development of the interview guide used in this study. Methods The data collected were from participants' evaluation forms for a pilot of a course on 'Incorporatmg HIV prevention into the medical care of persons living with HIV'. At each pilot course, staff explained and distributed the surveys and collected them post-course. An on-line long-term evaluation survey was also distributed electronically to course participants approximately 3 months after the course took place. 40

PAGE 52

Table 3.1. Evaluation forms used and time of completion Evaluation Forms Participant information form Time of completion 1 1 : Post-Course & Long-Term Eval. Prior to course Module 1: Risk Assessment I Mod 1 Pre-course Prior to course in all and STD Screening I form sites ________ u...,..r...,..se---'. __j j Module 2: Brief Risk 1 d . After Module 1 in j Reduction Interventions Baltimote; I Mod2 Pre-course Prior to course in other I I form sites I ,l Mod2Post-course .J, Imm ___ e-d-ia-te-ly-_ .-a-ft-er-I I form Module 2 ---------.... -... ----'-----.......... Module 3: Partner Services ... I I 1 and Referrals I I After Module 2 in Baltimore; Mod3 Pre-course Prior to course in other form 5 sites . ... .... .. .. .. Mod3 Post-course Immediately after form Module 3 Long-Term .... .... A,,, 'M, "'' ....... ...... Combined survey for all three 3-4 months after course modules completion ... .. ... . .. Participant Profiles For this evaluation, only course participants who were clinicians were included in the data. There were a total of sixty-four clinician participants in the post-course survey, and twenty-two clinician participants in the long-term follow-up survey. Both sets of survey participants are comprised mainly of physicians: 61.7% N=37 of post-course participants, and 66.7% N=l4 oflong-41 I

PAGE 53

term participants. Most participants also work in the functional role of care provider/clinician: 57.1% N=36 ofpost-course and 66.7% N=14 oflong-term. Some participants work in a different role other than as a clinician, with 4 8% N=3 of post-course and 9.5% N=2 of long-term participants working in education/training, 3 2% N=2 of post-course and 4 8% N=l oflong-term participants working as administrators, and 4.8% N=l oflong-term participants working as a case manager. Thirty-five percent N=22 of post-course and 14.3% N=3 of long-term participants did not specify their primary role. Post-course survey participants reported working in their primary profession for an average of 11.65 years, and long-term participants reported working in their primary profession for 10.95 years The majority of participants work in a hospital or hospital-based clinic setting, with 68.8% N=44 of post-course and 54.5% N=12 oflong-term participants working in this setting. The rest of the participants work in public settings including: community or migrant health, community based organizations, STD/Family planning, community mental health center, and correctional facility. An exception to this is 6 3% N=4 of post-course and 4 .5% N=1 of long-term participants work in a solo/group private medical practice, and 1.6% N = 1 of post-course and 4.5% N = l oflong-term participants work in a HMO/managed care setting. Just over two-thirds of participants in both surveys reported that they provide all HIV services throughout the course of the disease to their HIV infected patients, with 68.8% N=44 of post-course and 68.2% N =15 oflongterm participants reporting this. 42

PAGE 54

The participants in the interviews for the qualitative thesis represent a subset of the individuals that participated in the training course as well as key informants involved in the curriculum development. Measurement of Variables in the Evaluation Instruments All quantitative data were entered into a database, managed, and analyzed using the Statistical Package for the Social Sciences (SPSS), Microsoft Excel, and Microsoft Word. Descriptive statistics were generated from the evaluation instrument data, some of which were used to understand the big picture of the barriers and challenges faced by clinicians in implementing behavioral interventions with persons living with HIV. Post-course, and then again at approximately three months after having completed the course, participants were asked to assess how strongly they agreed or disagreed about the influence of specific barriers to their implementation of behavioral interventions during the medical encounter with HIV positive patients. The participants chose one of four rankings for each variable: strongly disagree, disagree somewhat, agree somewhat, or strongly agree. For analysis, categories were collapsed: strongly disagree and disagree somewhat were collapsed together as were agree somewhat and strongly agree. 43

PAGE 55

Findings from the Post-Course Survey The pilot course was divided into three modules: Module 1-Risk Screening, Module 2-UniversalPrevention Messages and Tailored Behavioral Interventions, Module 3-Partner Counseling and Referral Services. Post-course barriers data were collected at the completion of each module, and was collected three times during the workshop. While most participants completed all three modules, some participants only attended one or two of the modules. These data allow us to analyze the barriers to implementation of specific aspects of the guidelines. 44

PAGE 56

Table 3.2. Perceived barriers reported in the post-course evaluation Please indicate if you think the following will be a barrier to your ability to use fully the skills and information presented in this workshop. I B I M d le l(Ml) Module 2(M2) I Module 3(M3) arners 0 u ______ j Total I J The skills are not I 8 I relevant to role I I need more training I 17 I . I need someone with experience to help me use these skills at my workplace -4. .. -. .. . Lack of support from other staff Concerns about privacy : 6 and confidentiality ; -. ._:j Lack of referral resources if patient has inadequate insurance coverage . ... oO '' .. T 34 I Lack of time v . . -. Administrative barriers to change ... J Staff are uncomfortable 18 with these issues -. .. 8 Patient perception of stigma from a medical care provider* v A ....... . .. .. -*data for this barrier were not collected in the post-course instrument 45

PAGE 57

As shown in Table 3.2, in all three modules at the post-course data point, the majority of respondents felt that a lack of time (Ml :74%; M2:75%; M3:62%) was a main barrier to implementation of the guidelines. In all three modules, lack of time is the only barrier that was felt by over 50% of respondents. While the rest of the barriers were reported by fewer than 50% of participants, the frequencies are still high enough in many instances that roughly one-third or more participants experienced them. A number of felt that they needed more training, and this need was felt in all three modules (M1:37%; M2:48%; M3:39%). Administrative barriers to change were also felt by roughly one-third or more participants (Ml :44%; M2: 32%; M3:31 %). Some barriers were felt strongly in one or two modules, but not in all three. -In these instances, one-third or more of participants in one or more modules felt these issues were barriers to their practice. Many participants in modules 2 and 3 and a moderate number in module 1, felt that they needed someone with experience to help them use these skills at their workplace (Ml :24%; M2:40%; M3:38%), and they felt they had a lack ofreferral resources if the patient had inadequate insurance (Ml :20%; M2: 47%; M3:37%). A number of participants in modules 1 and 3 felt that staff are uncomfortable with these issues (Ml :39%; M3:38%)., while only a moderate amount in module 2 (23%) felt this way. Concerns about privacy and confidentiality were felt by 40% of participants in module 3, but by a small amount of participants in the other modules (Ml:l3%; M2:13%). Lack of support from other staff in implementation of the guidelines was felt most strongly in module 3 (32%), with fewer participants in modules 1 and 2 seeing this as a strong barrier (Ml :22%; M2: 14% ). The vast majority of respondents in all three modules (Ml :83%; 46

PAGE 58

/ M2:94%; M3:87%) felt that the skills presented in the workshop are relevant to their role as a clinician. Findings from the LongTerm Survey In the long-term follow-up evaluation, the barriers survey was presented to participants as a comprehensive evaluation, and represents the overall feelings the participants had about all three modules. While direct comparisons between post-course and long-term data are impossible because of the differing data collection techniques, general trends can be established. Lack of time (71% N= 17) was sustained as the main barrier to implementation of the guidelines, and the vast majority of respondents (96% N=23) still felt that the skills presented in the workshop are relevant to their role. Only 12% (N=3) of respondents had concerns about privacy and confidentiality, and only 17% (N=4) felt that staff is uncomfortable with these issues, both of which show decreases from the post-course data. Fifty percent of participants believe that they need more training, which represents an increase from the post-course data for all three modules. There is also an increase in individuals who feel that they need someone with experience to help them use these skills at their workplace (46% N = 11 ), and a slight increase in participants who believe they lack support from other staff(34% N=8). There is a slight decrease in participants who feel they encounter administrative barriers to change (29% N = 7) 29% of participants still believe they lack referral resources if the patient has inadequate insurance coverage. A new barrier was added to this survey instrument that was not present in the post-course 47

PAGE 59

instrument: Patient perception of stigma from a medical care provider. This barrier was reported by 21% (N=5), or about one quarter, of respondents. 48

PAGE 60

Table 3.3. Perceived barriers reported in the long-term evaluation Discussion of Survey Results Some useful comparisons between surveys have been made above in the survey data. These preliminary survey findings affirm that barriers exist and are highly relevant to effective dissemination of the guidelines. Additionally, these survey data have provided a comparison to my qualitative thesis results, and has helped to frame the questions asked in my interview guide. There are a few 49

PAGE 61

fundamental limitations to this data. As is the nature of survey data, it is difficult to understand how the participants are interpreting each barrier in the survey. The question arises as to how meaningful this survey data can be, given this limitation. This is one reason why it is important to elaborate upon this survey work by conducting in-depth interviews with curriculum participants. As will be discussed in more detail in the results there are many themes that do not loom large in the survey data that do so in the interview data. One possible reason for this is that it is impossible to elaborate views in the data, while the interviews allow a more in-depth and nuanced analytical approach. Finally, due to data collection issues, it is difficult to make a comparison between the above data sets. For instance, while the pilot data have a post course and long-term component, the post-course data were collected in three separate modules, and the long-term data in one comprehensive survey. Participant information is complete for the post-course data, but is spotty for the long-term set. Thus participant profiles are impossible to compare. Some comparisons between survey results and qualitative results have been made in the results section to follow. The preceding limitations hinder powerful comparisons between the surveys and between the surveys and qualitative results. The researcher has decided to include these comparisons keeping these limitations in mind. 50

PAGE 62

CHAPTER4 METHODS Data Collection This study examines the barriers and challenges that HIV clinician providers face in implementation of guidelines for conducting behavioral interventions with their HIV -infected patients. My thesis is based on data collected from interviews that I conducted as part of the larger NNPTC curriculum development and evaluation process. The pilot curriculum course took place at six sites in 2005 around the continental United States: Baltimore, New York, St. Louis, Denver, Oakland, and Tampa. The pilot curriculum was conducted by the National Network of Prevention Training Centers in collaboration with the AIDS Education and Training Centers through funding from the Centers for Disease Control and Prevention in Atlanta, Georgia. The data collected for this study consist of semi-structured interviews with key informants involved in the curriculum development process, and semi structured interviews with HIV providers who participated in the pilot of the curriculum. The data were collected over a two-month period during the summer of 2005. The interview data were supplemented with data from a preliminary study that presents survey results on barriers clinicians face to implementation of guidelines for implementing behavioral interventions with HIV positive patients during the clinical encounter. The literature resources enhanced the collect e d data by giving a background on the theory and preliminary studies used to initially develop the guidelines and curriculum, and gave insight into the most recent trends in HIV infection and behavioral interventions The survey 51

PAGE 63

provided useful preliminary data of barriers and challenges clinicians face with implementation of guidelines for conducting behavioral interventions with HIV infected patients. The survey data were useful because the participants came from the same pilot curriculum study as the participants that completed interviews for this study. This study has been designed and the sampling method has been chosen in order to fill in a key gap in the research as yet conducted on the medical encounter with persons living with HIV. This study is also designed to maximize rapid dissemination of the research findings. Critical findings with important public health implications will be transmitted to the Denver STD/HIV Prevention Training Center and the CDC for public dissemination and discussion. Course Details The course consists of three one-hour modules: Module 1: Risk Assessment and STDScreening Module 2: Brief Risk Reduction Interventions Module 3: Partner Services and Referrals The modules for this course are usually done one after another and are approximately one hour in length. Module 1 consists of information on how to conduct a risk assessment during the medical interview with a person living with HIV. This module also addresses the common STDs that are found in high-risk individuals and how to identify and treat those STDs. Module 2 covers the steps in conducting a brief risk reduction intervention with a person living with HIV. In this module clinicians are instructed on how to ask 52

PAGE 64

questions about high-risk behavior related to sexual behavior and drug use. It is emphasized in this module that these interventions are to be very brief, lasting only a couple of minutes, and are often accompanied by follow-up intervention questions at later medical visits. The clinicians also learn how to create a client centered, individually tailored risk-reduction plan for the patient to follow. Module 3 consists of on partner services and referrals. Clinicians are instructed on the different ways that a HIV positive person can disclose their status to a partner. This module also covers referral resources foJ," the clinician in case a patient needs more in-depth counseling than what can be provided during a brief interaction between patient and clinician. Human Subjects All research was reviewed and approved by the Human Subjects Research Committee at the University of Colorado at Denver and Health Sciences Center on May 19, 2005 (Approval #2005-115). Sampling For this study, seventeen participants were recruited Purposive criterion sampling was accomplished in order to select participants. This was a convenient way to select participants that couhi give a wealth of information on their experiences in the curriculum development process or participation in the pilot curriculum. The participants consisted of eleven clinicians involved in the care of HIV positive patients, three key informants involved in curriculum development, one key informant involved in clinical training, one key informant 53

PAGE 65

who is a 'Partner Counseling and Referral Services' specialist, and one HIV case manager key informant who works closely with clinicians. The eleven clinician participants were selected for interviews based on the following criteria: (1) They currently provide clinical care to HIV -infected patients, (2) They see five or more HIV positive patients per week, (3) They participated in the pilot course curriculum (one exception was a clinician who was very familiar with the course curriculum, but had not actively participated in it) Some of the key informants were chosen for this study based on their involvement with curriculum development and dissemination. The PCRS specialist and the HIV case manager were chosen for this study because they work closely with HIV providers and because of their professional background were able to give a unique perspective to the issues. Of the clinicians interviewed in this study, six were physicians, two were nurse practitioners, and three were RN' s. All of the clinicians interviewed worked in a public setting, with all working in a health department or public hospital setting. Ten clinicians worked in a HIV /ID clinic, and one clinician worked in a STD clinic. All clinicians saw at least five HIV positive patients a week with most seeing more, and some seeing as many as 40 to 50 patients a week. Seven of the clinicians were female and four were male. These clinicians represent high-volume HIV clinical care providers who have received the pilot course training on, or in the case of one clinician are knowledgeable about, the clinical guidelines for incorporating HIV /STD prevention into the medical care of persons living with HIV. The participants for this study were recruited from the pool of 11 0 clinicians and non-clinicians who attended one of six pilot training courses on incorporating HIV prevention into the medical care of persons living with HIV. 54

PAGE 66

Sixty-eight of the pilot participants were clinicians Many of the clinician participants in the pilot were invited to the event because of their lengthy experience and high standing in the field of HIV care. These individuals had the ability to be effective role models for the larger community ofHIV clinicians, and had the ability to give valuable feedback to the curriculum developers about the training. About three months after the courses took place, a long-term evaluation survey was electronically disseminated to all fifty-one clinician participants that filled out post-course evaluation instruments in the courses. Respondents to this survey were asked if they could be contacted for an in.terview about the barriers to implementation of the guidelines If the participant responded 'yes,' the researcher then contacted them. It was then determined, based on the above criteria, if they qualified for the study. Of the twenty-two clinician participants that filled out the long-term evaluation forms, eleven of the clinician participants agreed to a follow-up interview and were later contacted. In addition, some non-clinician respondents were interviewed as key informants because they worked closely with HIV clinicians, and could provide a unique perspective on the questions at hand. Key informants involved in curriculum development were chosen and contacted by the researcher based on their high level of expertise and involvement. An interview was either set up in person, if possible, or over the phone. The interview process began with an introduction about the project by the researcher and a discussion of informed consent. Consent was obtained either by signature on a consent for or verbally. To open up the interview, the participants were asked a general question about barriers that they face to implementation of the guidelines for incorporating HIV prevention into the 55

PAGE 67

medical care of persons living with HIV. In some of the interviews, probes were used to gather information about specific barriers. Data Management and Analysis Qualitative Data In depth, semi-structured open-ended interviews were used in order to better understand the barriers and challenges to implementing behavioral interventions in the clinical care of persons living with HIV through an analysis ofthe subjective experiences ofthe clinicians. Interviews were conducted with key informants involved in the curriculum development process, HIV provider clinicians who participated in the pilot curriculum course, and a IDV counselor who works closely with clinical staff. The question guide that was used in the interviewing is found in Appendix A Time efficiency is a main concern for this project because clinicians tended to have little spare time, and preferred to have interviews last only 15-20 minutes. The semi-structured interviewing method was chosen because it minimizes the variation among interviews, creates a highly focused interview that is time-efficient, and facilitates analysis by making responses easy to find and compare (Patton 2002). While the question guide was used in the interview process, much liberty was used to probe pertinent topics and issues that arose but were not anticipated when the interview guide was written. Key informants were routinely asked to respond to barriers topics that clinicians reported in the preliminary survey and clinician participant interviews, while the clinician participant interviews were kept open-ended in order to capture the individual barriers felt by these individuals. 56

PAGE 68

Clinicians were interviewed either by phone or, when possible, in person. All interview participants read and either signed a consent form or gave a verbal consent. The consent form given to all participants was approved by the human subjects research committee, as was the option of providing only verbal consent. The interviews were recorded either using a telephone recording device or a digital voice recorder. The recorded interview material was then transcribed, coded and analyzed for emerging themes using Atlas ti. Coding The interviews were transcribed verbatim and then loaded into AtlasTi for analysis. Through thorough reading of the interviews, codes were developed by the researcher to organize and delineate data by the major themes. All of the codes were derived directly from the interview data. Some of these derived codes reflect the barriers investigated in the preliminary study survey, while others were entirely unique to the interview data. In this way, the researcher coded the major content areas of the interviews and subsequently developed more specific subcodes within each content area. Major codes were developed to encompass the range of responses within the subcodes (Table 4.1.). Variables such as age, sex and social class were not assumed to have analytic relevance unless data showed them to be so (Berg 2001 :253). The coded data were then separated into social and structural barriers and analyzed for content. While this separation was made for the purpose of analysis, much crossover between social and structural barriers was noted and is reviewed in the discussion chapter of this report. 57

PAGE 69

Probes were used in the interviews where the researcher wanted to obtain very specific information from the participants. Among the structural barriers, time, referral resources, and the need for training were for the most part emergent themes from the interviews that didn't require probing by the researcher. Participants tended to mention time at the beginning of their interview, and thus the course of the interview contained information about this barrier. Referral resources and training needs were discussed by enough of the participants that these barriers usually did not need to be probed. Other potential structural barriers did not emerge naturally in the interview process. In order to obtain data about the nature of some structural barriers, the themes were the results of probes by the researcher. These included probes about administrative barriers to change and ICD-9 coding for behavioral interventions. In all, six of the participants were probed about administrative barriers and six were probed about ICD-9 coding for behavioral interventions, but the six probed were not necessarily the same for the two subjects. Probes for ICD-9 coding were used in order to gain insight into whether monetary concerns or reimbursement represented a barrier for implementation of the behavioral interventions. The clinicians were asked, "Would it help you to implement the behavioral intervention if there was ICD-9 coding you could use"? Other probes were used in the interview process to gain further insight into issues brought up by the clinician during the interview, and were thus derived from the data. For instance, if the interview participant mentioned 'not enough time' as a barrier, the researcher would follow with, "How much time do you feel you would need to perform the behavioral interventions"? Two individuals were probed for information about referral resources in their clinic, 58

PAGE 70

while many more participants brought the subject into context on their own Similarly, the issue of training was not probed for in the interviews. Content Analysis The quantitative survey questions created during the preliminary conceptual and design phases of this study prior to any fieldwork, were used as a source to draw from in organization of the qualitative analysis (Patton 2002). Themes that emerged in the interviews were compared to the quantitative survey results Beyond a strict comparison, new analytic insights emerged in the qualitative data, providing further insight into the beliefs of the HIV practitioners. Content analysis was done in order to understand the individual experiences participants had with the barriers, and to develop patterns among the participants. These patterns helped to define the parameters of the barriers, as well as the relationship that the barriers had to one another. Analyst constructed codes were created through analysis in order to identify and make explicit themes that appeared to exist but sometimes remained unperceived by the interview participants. These codes, such as 'HIV stigma' help to add depth to the analysis and also allow for a study of the interrelation of barriers. 59

PAGE 71

Table 4.1. Major themes and codes :-60

PAGE 72

CHAPTERS FINDINGS The themes that emerged from the interview data have been separated into social and structural barriers for analysis. This differentiation helps accomplish the first aim of this study: to gain insight into whether the barriers that clinicians face to implementation of behavioral interventions with their HIV positive patients are primarily social or structural in nature. Below, an analysis of social barriers is followed by an analysis of structural barriers. Social Constraints Many of the barriers that emerged :Irom the interviews were what I am referring to as "social constraints." These are individual-level and interactional barriers that are related to the knowledge, beliefs, biases, attitudes, and communication skills of the clinician, as well as interactional dynamics between provider and patient. Because many of these barriers are very personal and part of the individual's social reality, they are difficult to assess, and even more difficult to change. In many cases, the worldview of the individual is altered in order for them to personally identify and address these social barriers. While there is a plethora of social barriers that can affect an individual's behavior, this study highlights only the barriers for the that are unique and/or important for health communication between cli:q.ician and patient during the medical encounter. The clinician can have social barriers related to their views about the biomedical system and their role as a practitioner in the field, and their beliefs 61

PAGE 73

about disease and health especially as it relates to HIV. As well as preconceived notions about their own professional identity, the clinician can have beliefs about the role of the patient during the clinical encounter and what constitutes proper behavior from both parties during this encounter. The clinician may also encounter barriers that stem from their beliefs a'9out the causes of risk behaviors, espe
PAGE 74

17% of long-term survey participants reported that they felt staff is uncomfortable with these issues. Provider Discomfort One of the most commonly stated barriers that arose in the interviews was provider discomfort. Many clinicians felt discomfort when talking about issues of sex practices and drug use with their patients. Clinicians also felt discomfort talking to patients while a partner was in the room, and with discussing risk behavior directly with the partner of an HIV -infected person. Because of this reticence, many clinical encounters either lacked a discussion about these important issues, or contained an abbreviated discussion with little or no follow up or referrals. Some clinicians were aware of their low comfort level and openly discussed their concerns in the interviews; nonetheless they encountered difficulties in overcoming this barrier. Some clinicians identified a discomfort in general with risk behavior discussions This discomfort could lead to an avoidance of the issues altogether. Sometimes people feel discomfort discussing high risk behavior, so sometimes they just don't ask patients about this, I would say that [this is] a main barrier. [PROVIDER]1 There's plain old failure of will where we don't make it the top priority because we are excited about people getting healthier, and it is hard to talk to people about the more delicate issues. [PROVIDER] 1 At times during the interviews clinicians discussed barriers in the third person, seemingly excluding themselves They may feel more comfortable discussing barriers in this manner, therefore not assigning the issues to them specifically. Alternately, they may have overcome these barriers, but have witnessed the barriers in other clinicians. 63

PAGE 75

Some [clinicians] are just not very comfortable with doing it. They are learning to deal. [PROVIDER] One clinician, while able to conduct a thorough risk behavior discussion, had seen other clinicians truncate the discussion and thus render it meaningless. There's the personal comfort level. People need to become comfortable talking about things that are scary, and it often gets translated into "Here's some condoms," or "Are you using condoms?" or a few little things like that that are essentially meaningless when it comes down to getting people to make some behavior changes. [PROVIDER] Sex is considered to be a very private matter by many people and sexual activities can have strong moral implications. Clinicians at times became embarrassed talking about sex, or felt that the discussion would embarrass the patient. Because of this, sex was a difficult topic to broach during the clinical encounter for some clinicians. One clinician had observed this among some of the clinicians at her clinic. Patients up to 21 years of age were treated in this pediatric setting, making a large proportion of the population sexually active: For the ones [clinicians] that are uncomfortable, I'm not sure if their morals or ethics come into question about those things. Or, in talking with one, he said that that was the reason he went into pediatrics, so that he wouldn't have to talk about sex, but we have do that with our population. [PROVIDER] 64

PAGE 76

A key informant related this reticence to a cultural norm in our society about sex: I think it's less the stigma of the disease and more the culture of'we don't talk about sex' ... I think we're embarrassed to talk about sex, and we think our patients will be embarrassed if we talk to them about sex .... Sex is just something that people don't want to talk about, and so they don't and their excuse is, "well they came in with a cough, and they came in with an ache, and that took all of my 15 min". [KEY INFORMANT] A lack of training can be cause of discomfort when discussing these issues with patients. Training can help clinicians overcome the barriers associated with lack of knowledge or experience. Training can also help clinicians to understand the risks and social behaviors that are prevalent in the population they are treating. In the preliminary survey data, 50% (N=12) of clinicians in the long-term follow up reported that they felt they needed more training. Forty-six percent of clinicians in the long-term pilot follow-up felt that they needed someone with experience to help them use these skills at their workplace. Having knowledge of lingo used in the population, knowledge of common and current drug use and sex practices, and knowledge of the common places and situations in which sex and drug risk behaviors take place can help the clinician become comfortable with discussions of risk behaviors and interventions. In turn, the patient will also feel more comfortable with the clinician if they feel their provider understands their life situations and risk behavior. In this way, knowledge is one of the cornerstones ofhealth communication. In the following interview excerpts, clinicians discussed their views on knowledge and training: 65

PAGE 77

... in my training [in medical school], I can't remember having anything like a role play or whatever that would make it easier and make me more comfortable asking questions related to sexual activity. [PROVIDER] Another barrier is knowledge, which is part of the comfort issue [PROVIDER] Provider comfort [is amain barrier]. A lot of providers aren't, and me to some extent are not comfortable ... Most providers have fairly boring backgrounds and haven't been to bathhouses and so talking to patients is somewhat not totally natural. [PROVIDER] One of our ... physicians here ... didn't k.riow what top or bottom meant. [That is] the language that you need to be able to communicate. [PROVIDER] One key informant identified three aspects of this provider discomfort around lack of knowledge. This key informant contends that clinicians don't know how to ask the questions, are concerned that they will receive responses that they don't know how to deal with, and don't feel they have the resources to deal with these problems if they do arise. This individual believes that training is the solution to discomforts related to lack of knowledge: "Providers became more comfortable after they were trained; the big point is that they felt more comfortable and asked more questions". Clinicians don't quite feel comfortable about tailored behavioral interventions; they don't know how to ask the questions. An even bigger reason why providers that care for HIV positive patients don't ask those questions is because they are going to get responses that they don't know how to deal with. They feel that if they start asking questions they will open Pandora's box and will find out all of these horrendous 66

PAGE 78

things that they are not going to know how to deal with or have resources to deal with, so [they are] not going to ask. [KEY INFORMANT] Many clinicians were also concerned about the reaction that their patients would have to questions about sex and drug use. This was a reflection of the clinician's discomfort and not the patient's discomfort. As one of the key informants asserts, studies have shown that HIV patients want to have these discussions. This individual contends that despite concerns, patients and providers bond well: The perception that the patient and clinician will bond poorly is just that, it is a perception. They [the patients] consider this part of their medical care and they want to have these discussions. So, it's purely a provider perception that patients will respond poorly. It's not reality. [KEY INFORMANT] However, clinicians interviewed saw this as a very real barrier. One HIV clinician provider, in referring to other clinicians stated: They [clinicians] think it will get in their way: "My patients aren't going to like it" ... This discomfort is coming from the provider and not the patient. [PROVIDER] Another HIV clinician provider felt the discomfort personally: Sometimes when I ask these patients, they are just insulted that I ask them. 'I'm not trying to judge you, I just have to ask sometimes, and you don't mention these things'. But then I feel terrible and I'm less likely to ask the next person. [PROVIDER] 67

PAGE 79

Another key informant was also well aware of this barrier: Some providers don't want to go there. It's like a mute point. "My patient is already infected; I'm going to stigmatize and put my patient on the spot," and that is going to be an uncomfortable situation, an uncomfortable conversation .. .I hear all the time from our STD course, where clinicians are not asking difficult questions, are not probing into specific behaviors ... They are trying to find a way to ask these questions without catching their patient population totally off guard. The first step is to become personally comfortable. [KEY INFORMANT] Some clinicians did not feel personally comfortable with the actions of their patients, or their own actions and identities. This can create a very real barrier in which personal discomfort leads to an avoidance of the discussions. Boredom, anxiety, hunger ... that's a little more simplistic than trying to figure out, why you would do something that would hurt you or someone else. [PROVIDER] Most physicians are very awkward about it, there is the occasional gay physician that is good as long as they are OK with their own gayness, and if not they are a true mess and would be better out of the picture completely. Another one [barrier], at least in this clinic, is class distinction We tend to be not from the same class as a lot of our patients ; we tend to be upper class or middle class so those are real barriers. [PROVIDER] 68

PAGE 80

Judgment/Bias For some clinicians, the barrier was not so much discomfort with the issues, but rather personal judgments that they made about the patient. For some, this meant that they had decided, without asking the patient, that the patient did not have risk factors such as multiple partners, engaging in risky sex, or using drugs. This can be a mistake that has grave consequences, since, for instance, Mark Thrun MD asserts that 60% of MSM who report having a committed/main partner, also have other partners (personal communication, June 2005) . . my own biases may contribute to me glossing over the subject, or not even addressing it in a consistent manner. I might, for example, make an assumption [about] someone who has told me they are not sexually active, I might forget to check in and see if that's changed. Or I might make an assumption that when someone says they are monogamous in the past, that they are still monogamous I refer to that as my biases that there is not risk there. [PROVIDER] One HIV clinician provider grouped patients into categories, those who needed interventions and those who didn't. This can lead to missed opportunities to do necessary interVentions with the 'good' group. Changes in the patients can lead to the need for a behavioral intervention, and some behaviors are well hidden, causing the need for the provider to look past their basic judgment and ask the questions anyWay. Most of my patients, especially the ones that need behavioral interventions, really have a lot of issues. Some patients are on autopilot and are doing everything they should be. [PROVIDER] One clinician believed that others had developed a fatalistic view of behavior change, and thus did not believe that the patients were capable of 69

PAGE 81

changing their risk behaviors. A key informant also believed this trend exists among providers: Clinicians [believe] that it's not going to do any good anyway, because they are not going to change all of these people. [PROVIDER] [Clinicians] consider it soft [behavioral interventions], that it's not immediately important in that clinical visit. They also think that it's not necessary; what they say may not make an impact anyway, so why bother that it's a pointless exercise. [KEY INFORMANT] Partners of Positives (POP) A few clinicians expressed an increased discomfort when discussing issues about partners with their patients, or having risk discussions with partners in the room. In the first quote below, physician discomfort was the main barrier in these discussions, the clinician also felt hesitant to stigmatize the patient, and wondered about issues of privacy and legality. The delicacy issue that I believe is more at the root of the barrier in terms of clinicians, and I can just talk for me and other people can agree or not, is the issue of partners. I feel shy about focusing on the issue of protecting partners because it makes me feel like I am passing Oudgment], accusing ... Basically it's not about my patient anymore, but about how my patient is dirty or broken, or dangerous. That feeling makes it hard for me to have those discussions with patients ... about disclosure and about partner notification services in particular. I don't really have a lot of trouble at this point with the sex and drugs parts, it's the parts about partners ... The partner notification section of the curriculum increased my fears on the part of the patient and made me feel more like this was going to be a punitive thing. My patients were going to be told that they were dirty and bad . There was a lot of talk about the law 70

PAGE 82

and criminalization ... that part of it was not approached with a good understanding of the audience. What their objections and concerns were likely to be; those were not addressed. [PROVIDER] In this second passage, a clinician was faced with discussing risk behaviors with the partner in the room, and again expressed discomfort with this situation, the clinician's discomfort came also from a lack of knowledge of the population. Researcher: I'm going to have you try and think about a tirrie that sticks out in your mind when talking about sexual and/or drug practices with an HIV positive patient was really challenging. Respondent: I think that sometimes it can be more challenging if a partner is in the room; I'm thinking of a time when I brought it up again [risk behaviors] and directed it to both partners, 'Are you using condoms and engaging in risky sexual behaviors?'. That was a difficult scenario because they just giggled and looked at each other, and made some joke that I really didn't get. That was a situation where there were a few different things: they were both African American men and so in some ways it was a different culture. I don't necessarily know the lingo in either of those cultures. It's not that I would be expected to use it, but I don't always catch all of the innuendos that go along with it, and don't always respond to those innuendos. It was an instance where there was some sort of underlying something; I didn't really know how to respond to that scenario. I guess I could have asked the partner to leave the room and then asked the questions. [This] would have probably been a good way to address it in some way, because neither of them are going to be open about having other partners with the other person in the room. [PROVIDER] 71

PAGE 83

Sometimes family members being present can cause discomfort issues for the provider: I had a patient, who was a chronically ill patient, which always came with his mom. His mom would know better about details about his condition than he would himself, so I would say taking a history from him would be very difficult. Having a relative or significant other in the room would limit your ability. [PROVIDER] HIVStigma HIV has always been associated with stigma, and these feelings of stigma can emerge during the medical encounter with the HIV positive patient. There can be a disconnect between the feelings of the patient and those of their provider, with the provider feeling there is stigma and the patient feeling differently, or vice versa. Twenty-one percent (N=5) of clinicians in the long term survey believed that patient perception of stigma from a medical care provider was a barrier to implementation of behavioral interventions. One key informant remarks: I think we as providers stigmatize HIV more than our patients are stigmatized by it. Our patients deal with the fact that they have HIV, [however] provider stigma around HIV is still there and we need to educate them. Patient perception of stigma from a medical care provider is really providers' perception of patient perception of stigma, as opposed to reality. [KEY INFORMANT] While the patient is unlikely to feel stigmatized during their medical encounter, if there is a chance they do feel stigmatized, they are less likely to seek medical care. Because of this, it is very important that the clinician makes 72

PAGE 84

the patient feel comfortable, by feeling comfortable himself or herself with the medical encounter. The same key informant also states: When you survey people who are not in care with HIV, it's because of stigma. [KEY INFORMANT] Some clinicians felt that the stigma came from both the patient and the clinician; other clinicians asserted that the clinician has the power during the medical encounter to either foster the stigma or erase it. I think the stigmatization is definitely there, and I think it works both ways. [PROVIDER] There's stigma, from both sides. [PROVIDER] I think the provider has the responsibility to make the patient either feel that stigma, or let them know that they are willing to talk about whatever. It really comes across in the sexual history that's taken. If you have providers that are non judgmental, that are empathetic and caring, that comes through very quickly and allows the patient to feel comfortable in disclosing information: Whereas, if your body language, or facial features, or the wording that you are using gives off that you are judging and disapproving [the patient can feel stigma]. [KEY INFORMANT] I think [the provider] has the responsibility to either foster that stigma by playing into it, or they can erase it very easily and very quickly. 'I see what you are doing is so dangerous or unhealthy, and since I'm not engaging in the behavior I don't understand how you would do it': that's where the stigma comes into play when dealing with long-term behavior change. [PROVIDER] 73

PAGE 85

A clinician gave us an account of when they were faced with a situation in which their patient felt stigmatized by them. This encounter made the clinician more hesitant to ask risk behavior questions in the future. I remember one person really got upset with me and said, 'what do you think I am'?! [I said] 'I'm sorry that was not the implication ofthe question I just gotta know so that I can counsel you on stuff'. He was like 'OK' and calmed down. I felt really bad, and it made me more nervous about asking about it! Because it's put upon them as a judgment statement that they sleep around, like all gay men sleep around. 'No! That's not what I'm trying to tell you'. [PROVIDER] Patient Discomfort Some providers felt that patients had discomfort with discussions of sex, drug use, and the related behavioral interventions. This discomfort can possibly emerge in the medical encounter for a few reasons. The patient may not feel comfortable talking about sex and drug use because of the taboo nature of the topic. Alternately, the. patient may have a strong desire to please their provider by good behavior, and may lie or fail to elicit information about their risk behaviors. It would be possible for these discomforts on the part of the patient to lead to a negative reaction by the patient during the encounter. Some providers felt they had identified ways to overcome these barriers by creating a comfortable and safe environment, and by not showing negative reactions to the patients' risk activities. While the patient can feel discomfort, there is a chance that this discomfort is not real, but only perceived by the clinician. It is impossible to tell this distinction by simply interviewing the HIV clinician-providers. An in-depth study of the patients should be accomplished in order to better understand this distinction. Regardless, the clinician can learn 74

PAGE 86

how to reduce this discomfort if it does in fact exist, and improve health communication during the medical encounter . . sometimes it's things that they would rather not discuss so much. Certainly it's things that they get uncomfortable talking about and without getting into too much detail they will attempt to brush off or, side step, or just simply be agreeable and nod their heads and not offer too much more than that. [PROVIDER] I think patients become noticeably uncomfortable when you ask, 'do you use a condom every time, what kind of sex do you have, and who do you have sex with'? Sometimes they're hesitant to be completely honest I think there's frequently hesitancy on the patient's part to admit that anything is currently going on. [PROVIDER] A key informant commented on this perceived reticence on the part of the patient to have these discussions, and had some suggestions for providers. Patients want to have discussions about what's going o"n in their lives They actually want to discuss their relationships with their doc. They want to have the questions asked and they are intimidated to bring up these issues with the doc, because they are not comfortable having these conversations. So if we as docs do not open up these issues and ask the questions, than more than likely they are not going to occur. We need to send the signal that it is OK to answer these questions. When surveyed, over 90% [of patients] want to have these discussions. [KEY INFORMANT] The reticence on the part of the patient to discuss these issues can come from the fact that they want to please their medical provider. The provider represents an authority figure in their life. One poignant example of this comes from a HIV case manager who works closely with patients in order to facilitate 75

PAGE 87

adherence to drug therapies and help with life situations the patient may encounter. This individual had noticed a marked difference in the patients' disclosure of risk behaviors with her/him and with their clinician providers. I think that it is easier because I am not someone that they feel they have to answer to because I'm not their provider; they feel that their doctor is their provider. They are more likely to say to me that they didn't take therri [their medications] because they were out partying all night. They do have close relationships with their providers, but I think that there is always that guilt, and they don't feel that with me. They don't feel any sense of responsibility like they do with their providers, that they have to put their best foot forward, so they are much more likely to tell me. They just don't care quite as much about what I think as they do with their providers, because they want them to think highly of them For me, I'm just a ride there and pills or whatever. [HIV CASE MANAGER] HIV clinician providers also recognized this trend with their patients: I'll sometimes ask them, 'so is anything new, new partner, anything'? Sometimes they'll be open about it and sometimes you can hear the hesitation. Sometimes they want to tell you the right thing, but they don't want to lie to you. I think they do want to please you, I think that's a huge thing. [PROVIDER] As a society we place doctors up on this pedestal thing where, 'oh my gosh, if the doc thinks that I'm doing something bad' An example is my mother; my mother will not test her sugar for anything, (she's diabetic). When she goes into the doctor she doesn't tell him that she's had a headache or that she's not been eating so well because she doesn't want him to disapprove of her behavior. Rather than talking about, 'I've had some struggles doing what you want me to do', she leads him to believe that all is well. Then I get into the car and hear for the next 20 min about how her sugar is really not so well. [PROVIDER] 76

PAGE 88

When you ask about their sexual practices, you have to keep asking, 'Well, do you always use condoms, this week how many times did you use condoms, how many times did you have sex this week, how many of those times did you use condoms'? You have to really push at it, and I don't know if it's because they want to please you and they know that they are not supposed to [have sex w/o condoms] ... They don't want to really tel(you. [PROViDER] Patient discomfort can be particularly detrimental when it leads to anger, which in turn causes the provider to hesitate before asking about risk behaviors in future encounters. You can't ask someone directly [questions related to risk behaviors]. . You can, but it puts them very much on the defensive, like, 'what do you mean'? And some of them still don't have any insight into it; they don't know why they do it. [PROVIDER] However the negative reaction of the clinician can create the barrier to the il)tervention: I remember watching a client interview ... and [the clinician] was just lecturing, 'You need to do this, and I'm your doctor, and you need to listen to me and follow my advice because I know best'. Totally not client centered at all, totally not approaching it in a realistic way. I was just reading the patients and they were like 'yeah right and who are you, uh huh'. Man, [the clinician] missed that whole thing completely! [PROVIDER] While patient discomfort can exist, some providers were able to overcome this barrier by establishing a safe environment for the patient to discuss difficult issues, and by being non-judgmental. One provider referred to a study done 77

PAGE 89

recently on different cities and the corresponding patient reporting of drug use and unsafe sex. One of the things found in the [city name] sample was a lot more reported drug use and a lot more reported unsafe sex than other places. I don't think [city name] is any different than anywhere else, but if the patients are comfortable with their providers than these conversations will take place. [PROVIDER] Another provider had developed a unique situation within the medical community: s/he makes house visits to his/her clients. It allowed. him/her to engage the patient while they were in a setting that they felt they had control over and were familiar with. 'Do you have a private setting?': we establish that before we leave for the setting. We want to make sure the patient is comfortable, so we also have another site we have for counseling or testing. When I go [to the house] ... I think for the most part our clients like it. They talk more freely to us and they will question things the doctors have said to them in the clinic that they were scared to ask. It's not threatening and they are not being rushed. A lot of times, they appreciate you coming out and sitting down and talking with them about things that they don't quite understand, or didn't quite get. [PROVIDER] Using a client-centered approach to the behavioral intervention can lessen barriers of discomfort the patient may feel, even in a clinical setting: I know in our clinic we have had patients leave that are very surprised that the visit has gone as well as it has. It's very individualized. [PROVIDER] 78

PAGE 90

Sex and Drugs An interesting theme that emerged from the interviews was the different comfort levels associated with discussions of sexual activities versus drug use. Many clinicians had noticed more hesitancy or reluctance on the part of their HIV positive patients to discuss drug use. This reluctance may stem from the fact that drug use is illegal, or the patients may feel another as yet undisclosed discomfort with the issue. Drug use is actually something that people try to avoid more than the sexual practices topic. Just thinking about it, there are instances where someone is known to be using [drugs] through other sources, and when you ask them about that they will deny it is actually happening. [PROVIDER) I think patients are more likely to deny drug practices; most patients are comfortable being open with sex practices. [PROVIDER] Sex, no I don't think I've had a single patient object to my asking about sexual practices and I do it a lot. Drugs, occasionally, more often people of color might see a drug question as accusatory. I've had that happen just a couple of times, especially when I was younger, and now I let people know that it's not shocking to me, and I am concerned. [PROVIDER] Sometimes they're hesitant to be completely honest, less with sexual behavior than with drug use. [PROVIDER] ... most people don't want to admit they are using drugs because along with that they know that we know it also involved unsafe sexual activities, especially with meth [use]. [KEY INFORMANT] 79

PAGE 91

Another key informant had an interesting insight about stigma and risk behavior discussions. Again this issue boils down to discomfort on the part of the clinician associated with discussions of sexual activities. In contrast to the above quotes, this individual feels that there was more discomfort surrounding risky sexual activities than there was around drug abuse and mental health. We don't have this attitude toward other things such as substance abuse; we ask about substance abuse The stigma around asking about other things, such as mental health and drug abuse, is not as strong as it is around asking about of the transmission risk behaviors. [KEY INFORMANT] Sometimes, if the patient was expecting questions about sex and drug use, the discomfort was lessened: Most patients are expecting that question, 'Are you currently sexually active, how many partners do you have, what kind of partners, what kind of sex do you have'? I think they are almost expecting that, so I feel fairly comfortable asking it. [PROVIDER] 80

PAGE 92

Role as a Social Barrier Respondents overwhelmingly responded in the preliminary studies that they felt the skills presented in the workshop were relevant to their role: 96% (N=24) of clinicians in the long-term pilot survey. Compared to the survey data, many interview participants responded differently. Some clinicians felt that behavioral interventions were not an essential part of their role. Others felt that they should not be solely responsible for the behavioral intervention but should instead have a stronger referral network in place. Some clinicians felt unqualified to perform the behavioral interventions. Many clinicians interviewed felt that these behavioral interventions were not an essential part of biomedical care, and were easily trumped in a visit by more pressing medical issues. The difference between the survey data and the interview data can be at least partially attributed to whether the clinicians felt that these behavioral interventions were possibly relevant but not essential to their role as biomedical clinicians. The survey asked clinicians ifthey felt the skills were relevant to their role as a clinician, while clinicians in the interviews discussed how they felt the behavioral interventions were not essential to their role as a biomedical clinician. In order to address this barrier the curriculum could place more of an emphasis on the importance of these discussions to the health and well being of the patients that receive them, thus boosting the argument for why these interventions are in fact an important part of the role of the clinician. Some clinicians commented, in general, that behavioral interventions may not be their primary role as a biomedical provider: Providers aren't really sure that itis their role to implement behavior change. Ther. e is a lack of a perception by providers 81

PAGE 93

that it is something important in their scope of practice. [PROVIDER] I would say there are two types ofHIV [clinicians], some will only deal with HIV: they don't want to take care of their blood pressure, they don't want to take care of their diabetes, and they will send them back to their internist for that. I would say that probably the majority just take care of it, most of us remember our internal medicine enough to manage basic things: most of us can manage diabetes, cholesterol, BP. Rather than making them go to five appointments, you just become their primary care doc, and then health maintenance does start falling into that category. [PROVIDER] Referral Network Needed Some clinicians believed that a strong referral network was the solution, and persons more specifically trained in these approaches should conduct behavioral interventions. Twenty-nine percent (N=7) of clinicians in the long term pilot survey believed that lack of referral resources if the patient had inadequate insurance coverage was a barrier. This barrier relates back to the beliefs the clinician feels about their role during the clinical encounter. Many clinicians felt their role was more that of a biomedical practitioner who treats physical ailments than a behavioral interventionist. Despite this belief, the curriculum was developed based on evidence that short behavioral interventions delivered by clinicians during the clinical encounter are an effective tool for prevention of HIV transmission risk behaviors among HIV infected patients (Kamb et al. 1998). Because of this disconnect there needs to be an emphasis on the depth of behavioral intervention expected of clinicians. I don't have time, and sometimes it's truly enlightening to have them [psych or social worker] deal with these psychiatric 82

PAGE 94

issues. The psych or social worker will have a very detailed history of what psychological problems they have, and whether they have ever been abused ... That would have never come out in a visit ... I think if a had a full hour not to discuss the physical thing, just to discuss their mental stuff, than maybe, but even then, I don' t feel qualified to do a lot of the behavioral counseling. I think there's a lot of stuff that goes into .why they are doing things and how to help them modify their behavior. I don't feel qualified to do that, I can tell them to stop but that's not really the root of it. Some of them still don't have any insight into it; they don't know why they do it. I think that it takes a lot of really intricate training. They may have really big psychological problems. I mean, the substance abuse plays into it for a reason. It leads to the behavior, but again, why they drink so heavily, or why they do drugs, or why they do anything, I have no insight into it. I can diagnose depression and that's about it, I'm not trained in that. [PROVIDER] I think if it were either dedicated time or time spent with someone whose specific focus was going to be sex counseling or substance use counseling that would be a more effective way to go about it. I think that people identify in their minds what the roles are of various people in the clinic or various practitioners, and when they come to see their physician they expect that there are certain things that they can talk about They may not always relate that physician visit with discussing some of these other issues. If you had someone designated to do harm reduction and met with clients depending on their need this probably would work much more effectively than trying to bring it into a busy visit when there are other things going on. Getting another health care provider involved to focus more specifically on those issues would be my solution. [PROVIDER] We are a Ryan White funded facility so we have the whole team: we have health educators and even a risk reduction specialist. I think some of the barriers are historical, that they [the clinicians] have been doing HIV care for a long time and 83

PAGE 95

they feel they should just deal with the STD when they present symptoms, not ask a whole lot of questions and just move on. [PROVIDER] Two key informants felt that the level of behavioral intervention expected of the clinician needs to be clarified. They felt the more in-depth behavioral counseling might be more the role of specifically trained behavioral intervention professionals than the role of the clinician: They need to bring up the topic; they need to do some risk assessment. When it comes to the behavior change discussions and counseling and that kind of thing, they need to then be able to say, 'I want you to go talk to this person: a nurse, a case manager a social worker, a health educator to discuss these issues so that we can help you be safer, healthier and less risky' [KEY INFORMANT] Many of them will admit, and probably even say, 'well that's why we have people that are more specifically trained for these skills, that's why we have PCM's and social workers, health educators, because that's not my role'. And there is some truth in that because I think we are asking providers to be catchalls for their patients. Maybe we need to come up with a management concept of people that we identify that have high risk behaviors. Maybe we can bring in people that can handle this level. [KEY INFORMANT] Another key informant believed that it is in fact the clinician's ro l e, and they point out that studies have shown clinicians also feel it is their role. Again, an emphasis needs to be placed in the curriculum on the level of beha vio ral intervention expected by clinicians, thereby reducing this disconn ect: I think they believe that it is not their responsibility and this is poor clinical care because in most clinical practice settings the clinician is the one who drives behavior change ... Are the docs going to dismiss this as unimportant, and the answer is 'no'. 84

PAGE 96

actually looked at the data, and docs, nurses, NP's all increased the number of times they had discussions with their patients about risk. So training, education, and a little bit of practice works for everybody. I think if you buy into the fact that prevention in care should be the standard of care, I think anybody can be effective at delivering messages. [KEY INFORMANT] I Not Biomedical Care In the last interview excerpt above, the key informant asserted that, 'If you buy into the fact that prevention in care should be the standard of care, anybody can be effective at delivering prevention messages.' This next group of interview excerpts point to a different point of view While many clinicians surveyed in the preliminary study believed these skills were relevant to their role, and contended that prevention in care should be the standard of care, some expressed different beliefs when interviewed. As discussed above, this can be attributed to clinicians' beliefs that these behavioral interventions are relevant, but not essential to their role. During the clinical encounter, the patient and the clinician expect certain things to take place, those things being congruent with the biomedical culture. Within biomedicine, there is a focus on the physical manifestations and the curing of disease. In this sense, preventative measures such as behavioral interventions do not have the same level of importance to the clinician as antiretroviral therapies and pain/sickness issues There are a number of different agendas depending on what other problems the patient might have or what they may want to talk about during a visit. Sometimes the prevention message, safer sex issues, and drug use issues get triaged lower on the list. I think that's one of the barriers, especially with 85

PAGE 97

HIV positive patients, you often have a number of issues going on and things that need to be discussed. [PROVIDER] Pain can be a chronic issue that can take up a lot of time, especially if they are not taking their meds. I need to educate them about that. Sometimes the patients don't know what medication they are on, so we have to spend some time going over that. Honestly, I usually just circle 'HIV' and don't go through all of the issues, which is something that maybe I should be doing. [PROVIDER] In order to change behavior, you have to have discussions around it. Providers are not interested in spending time on something that they consider fluff; they don't think that this is a priority necessarily. They have clinical medical priorities such as asking about medications and making sure that people have their viral load artd T -cells done. [That is] the line between what providers consider strictly clinical and what they consider fluff such as questions around substance abuse ... I think that providers don't prioritize this; they are not willing to come to a training to learn about these things. They consider it soft, it's not immediately important in that clinical visit, they also think that it's not necessary, and that what they say may not make an impact anyway, so why bother that it's a pointless exercise. [KEY INFORMANT] It's always on my problem list, but it doesn't necessarily mean that I really address anything. Most of the time I address immunizations or colon screening, or pap or prostate; sort of more standard basic things ... What I call health maintenance stuff gets thrown to the wayside because they have so many active issues that you forget. You need to switch their regimen and you need to spend you time counseling, and then you sort of think about it later, like 'ugh, are they sleeping with someone new'? It doesn't dawn on you really to think about it at that moment, because you are dealing with active things. 86

PAGE 98

This may be active too, but it's not necessarily a priority in terms of what we are doing. [PROVIDER] ... you're concentrating on a couple different things, and you're telling the patient, 'OK, put your hands above your head, take a deep breath, hold your breath, swallow, (laughs)', all that. I think in that scenario that is where you get this, 'let's really not focus on that or have that sort of conversation'. [PROVIDER] Maybe there are queues like that someone has moved, or ended a relationship, that would lead me to talk about these things, but I'm not disciplined to do this at every visit with every patient. I'm not disciplined to do healthcare maintenance at every visit ... (The patient] was having physical manifestations of stress, and he wanted support, but I don't have the time, or clinical skills to counsel him. He's a MEDICAL patient, so I really didn't have a counselor that I could plug him in to, or make a commitment to find people for him to work with. What I could not do for this man is feel that I connected him with a clinician that could provide insight and skill to deal with this situation that he said is killing him .... (laugh) ... in a way that the HIV is secondary! [PROVIDER] 87

PAGE 99

Structural Constraints In the context of this study, structural constraints are those that are elements of the biomedical system and society at large, and the bureaucratic, administrative elements thereof. While social constraints discussed earlier in this study are represented by the personal beliefs and feelings of the practitioners that lead to barriers to health communication, structural constraints represent the larger material structures of society and more specifically the biomedical system creating barriers for the clinician. Once the material or structural changes to the biomedical system are made, it is much easier for the clinician to follow through with individual behavioral changes. Structural factors such as the amount of time a clinician is allotted to spend with each patient, insufficient referral resources, and lack of opportunities for training in health communication and behavioral interventions were strong constraints for HIV clinician providers interviewed for this study. Administrative barriers existed in the form of a lack of policy for consistent implementation of behavioral interventions. Some clinicians felt they faced monetary barriers in the form of a lack of reimbursement for their time and services for performing behavioral interventions, but most felt that this was not a barrier for them. 88

PAGE 100

Time as a Structural Barrier To begin the interviews, the researcher began with some version of the following question: "What do you think the barriers are to discussing sexual practices and drug use with your HIV positive patients?" Nine out of the eleven HIV providers, the PCRS specialist, and two of the four key informants involved in training and curriculum development mentioned time first in the interview. In most of these cases, time as a barrier was mentioned in the first sentence. Seventy-one percent (N= 17) of clinicians in the long-term pilot survey felt that lack of time was a barrier for them. This indicates that time was a very strong and real barrier for these participants, and the barrier that immediately came to mind for them. Participants were asked at some point during the interview how much time they had per visit, and the times ranged from 15 min to 30 min per patient, with most providers having 30 min for each patient. Researcher: "What do you think the barriers are to discussing sexual practices and drug use with your HIV positive patients? One is probably time for providers. [PROVIDER] My own personal barrier would be time. [PROVIDER] On busy appointments when you have a lot to do, you don't have time to talk or test for STDs: it's usually lack of time. [PROVIDER] Time pressures are a barrier to most things. [PROVIDER] 89

PAGE 101

I think that one [barrier] is time to get to all of the issues that a patient may come in with. [Many times] they come to your office with something that they bring up. [PROVIDER] Primary barriers. . [would be] time that the clinician has in a typical outpatient visit, and competing priorities which is another way of saying time. This [barrier] would be on the part of the clinician and the patient. [PROVIDER] Time is an element that is certainly difficult to incorporate especially in the setting you're in and the amount of patients that you have. [PROVIDER] Time is a major barrier, but I don't think it's a true barrier. Instead [it is] a perceived barrier that clinicians don't have time to really implement behavior change. [PROVIDER] One thing is time; there is never quite enough time to go into an in-depth discussion on that sort of a thing. [PROVIDER] First and foundational challenge to clinicians in implementing these guidelines is the time factor. [PCRS SPECIALIST] I think time is one that everyone identified as an issue, and time is an issue. [KEY INFORMANT] Time as a perceived barrier. [KEY INFORMANT] Two HIV clinicians did not mention time at the beginning of the interview. One mentioned it a little later on. The other clinician (quoted below) had to be prompted by the researcher to discuss time. I think it's an excuse to get off the hook, period! I think if you are working within HIV, it can be interwoven into everything you do. You don't have to be in clinician mode and then behavioral change mode; you incorporate all of that. Particularly when you move into the private sector, you are 90

PAGE 102

seeing fewer sick visits, and more well visits. Clinicians will get into talking about whatever they are interested in, and time really becomes an excuse in a lot of ways because they are always behind. How did they get behind? Figuring out how to treat pneumococcal pneumonia? They know how to do that, they got behind talking about whatever. [PROVIDER] How much time is needed? When asked if they had enough time to implement behavioral interventions, the answers were mixed, with some providers feeling time was sufficient, and others feeling that more time was needed. More interesting is the fact that other barriers were often mentioned when the participant was asked how much time would be necessary to implement behavioral interventions. One participant remarked that 'no amount of time would be sufficient,' indicating that other barriers must be present for them. In fact, many of the participants mentioned another barrier instead of or in addition to time when asked this question. Researcher: As far as time, do you feel that if you had more time than you could discuss these issues? In this quote, while the clinician felt they had enough time, they also felt that other clinicians may find the time limit a barrier. Depends on the visit, but on average it takes 3-4 minutes. We have 30 minutes for every appointment so I feel that there is a way to do it, but I'm not sure there is consensus from the other provider in my office. [PROVIDER] 91

PAGE 103

Issues of patient-clinician communication and interaction came up as a concern for this next clinician-participant. The clinician felt time was putting a strain on the patient-clinician relationship, not allowing the patient to discuss issues that are important to them. Yes! Well, I don't think initiating or establishing a first step takes a lot of time, and I feel very confident in my skill to do that. I feel the place that takes time is doing a behavioral assessment and allowing the patient to really talk, that's what's time consuming. I sometimes feel like I have to not cut people off, but rein them in a little bit, and that feels bad to me. I need to redirect and say, 'OK we need to move forward with this', when they may really want the opportunity to talk more. [PROVIDER] Doing a full physical takes a lot of time. Just moving around the room to do that takes 20 minutes and then that leaves 1 0 minutes. I guess it can be done during the course of doing everything, but it's interrupted time. When you are listening to someone's heart or looking into someone's ears it's interrupted time as opposed to a consistent discussion or an open discussion. [This] is needed more than just a cursory, 'Oh well, make sure you use a condom every time', but an actual conversation. [PROVIDER] This next participant brought into question the role and focus of the biomedical practitioner, and the need for a referral network was brought up. This brings into question whether they really perceived time as the barrier or some other factor I've thought about this before, and I think that other than having more time, because I think that whatever amount of time you give somebody it is going to get used up, I think if it were either dedicated time or time spent with someone whose specific focus was going to be sex counseling or substance use 92

PAGE 104

counseling that would be a more effective way to go about it. [PROVIDER] The role of the biomedical clinician is brought again into question here with these two participants. The behavioral intervention fell low in priority compared to physical concerns. We are a primary care clinic, so we are trying to address a lot' of issues. We are also trying to address hypertension and cholesterol, vaccines, healthcare screening, and for our patients who are in their 50's prostate and colon. We have a 25 point checklist, but most of it doesn't need to be done on every visit. [However if] most of it is done ... a half hour can go pretty quickly. [PROVIDER] I would say 'yes', because sometimes people come in with an acute illness or problem that needs to be taken care of right away and prevents us from doing health maintenance stuff. [PROVIDER] Lack of qualification or specialized training, as mentioned in the previous section on role, was brought up as a barrier. I think if a had a full hour not to discuss the physical thing, just to discuss their mental stuff, than maybe, but even then, I don't feel qualified to do a lot of the behavioral counseling. [PROVIDER] I would say somewhere between five and ten minutes would be good. Because there are things that you have to start talking about, and talk about it for a couple of minutes before people really ... are going to be open about it, or willing to actually to step in. [PROVIDER] In summary, while time was mentioned as a barrier at the very beginning of almost all of the interviews, elaborations on this barrier usually led to an explanation of another barrier. That was the case in the examples above where 93

PAGE 105

participants were asked how much time would be sufficient for implementation of behavioral interventions. Other Structural Barriers: Administrative, Monetary (JCD-9 codes), Training, and Referral Networks Administrative Barriers to Change As mentioned before, this barrier was often probed for in the interviews with the clinicians and key informants involved in curriculum development and training. The reason for this was that administrative barriers to change were considered a barrier by 29% (N=7) of clinicians in the long-term pilot survey. Consistent themes emerged among the interview participants. First, a lack of administrative will or buy-in was noted by some of the interviewees and key informants. This lack of buy-in by the administration can be because of political or social beliefs, or a reticence to implement a new policy. Many participants linked administrative barriers to time, citing that the administration decides how much time is allowed per visit. Various clinicians cited monetary concerns, which were linked to the lack of ability to be reimbursed for the interventions The next quotes reflect the feeling that clinician-participants and key informants had about the power of administrative will as a barrier to implementation of behavioral interventions. In the first interview segment, the participant believed that the lack of administrative will lies in the difficulty with implementation of a new policy. The next clinician felt that the barriers were more on a macro level and lie in the political and social ideals that guide our 94

PAGE 106

culture. This is especially the case with the current abstinence only policy encouraged by our government. A key informant also felt macro-level political and social powers could become a formidable barrier for clinicians. Researcher: Do you think that there are any administrative barriers to change that would cause people to keep from implementing these guidelines? ... trying to institute a new policy about how things are instituted in a clinical setting. Clinic administrators may not see it as a good utilization of the resources, [they may] need more data. [PROVIDER] ... one of the training sites they went to in [a state] would not allow literature about masturbation. I don't think any physician around here fear for their job when discussing issues with patients, I know I haven't and I'm always dancing out on the edge with that one. I fit this into the I don't have time to discuss these subjects, well, get specific, your boss said that you don't have time to talk about safe sex? Or, is the legislature firing you? We work for the city, we may as well be in Cuba, the University will be a little tighter, and in the private sector you have no one telling you what to do really, so it's really not going to happen. [PROVIDER] The powers that be won't let me do it: my boss, our supervisors, our organization will get in the way. Also, if the organization .doesn't buy into the importance of it, they are not going to allow you to do it. So if you work in a conservative clinic that believes in abstinence only, you won't be able to ask these questions. Family planning clinics are getting the message of abstinence first in their counseling practice. [KEY INFORMANT] 95

PAGE 107

In response to the same question above, the next two participants, one clinician and one key informant, saw the administrative barrier more as the level of importance that is placed on these behavioral interventions by the clinic or hospital administrators. I think there can be, it depends on the facility and the buy in of the administration. Clinicians will be more on board feeling that they have the ability, or even feeling that they have to change [if they have buy-in from the administration]. Ifthe administration does not see this as a priority or a need, than it gets pushed to the wayside It's like, 'Oh well, there's another directive or suggestion', but it's ignored. The administration really sets the tone for how things are looked at and are perceived. [PROVIDER] One of the biggest administrative barriers is the administration. If you don't get buy in from the clinic director that this is important it isn't going to be important. In places where the clinic director, the principal investigator, or the leadership in the clinic didn't buy in the providers didn't rank this as a high priority. When prevention in care is seen as important, where it is the mantra, people will do it. The curriculum needs to include the clinic directors and personnel. [KEY INFORMANT] This participant remarks that behavioral interventions were not part of the protocol sent down to them from the administration, which it seems was a signal to them that there was not administrative buy-in. We don't have a protocol, there's not institutional mandate, it's not part of our CQI, and I'm not audited: ya know, things that would keep me consistently screening people. [PROVIDER] 96

PAGE 108

Time was once again invoked as a barrier in the discussion of adminis!fative barriers. Understaffing which leads to a time crunch and mandatory patient scheduling were administrative barriers involving time issues My first thought is that it would be linked to time, there might be administrative barriers to that. [PROVIDER] An administrative barrier to more time. Our director is also an HIV provider and he is very amenable to the situation and the need to discuss behavioral things, especially risky behaviors. But at the same time, as far as administrative, we work for a large university, so we can't necessarily change visit time, or numbers, or that sort of thing... Right now we're understaffed, we don't have enough medical assistants. Patients tend to get into the room 1 0 minutes late and then your time is cut down to 20 minutes to do what you would have had to do. Then your next patient is coming and you don't want to make them 20 minutes late. That is an administrative issue because it's harder to hire somebody, and it has to be approved, when they only give two weeks [notice]. [PROVIDER] Patient scheduling where you are not allowed enough time. The time element again plays into the administrative, in that you have a minimum ainount to patients that you have to see within a given time constraint. [PROVIDER] Monetary concerns in the form of reimbursement from insurance, not having sufficient money for staffing, and dependence on state and federal funding were discussed by the following clinician-participants and key informant. In addition to monetary concerns, the key informant also mentioned administrative will, links to time, staffing and bureaucratic red tape as barriers. Insurance stuff like it would be ideal if I could bring my patients' partners in to do stuff, but I can't because of where I'm at. They don't qualify necessarily for the benefits there, so 97

PAGE 109

I couldn't see them. And what if they don't have any insurance, well then you are going to just farm them out somewhere else, and hope that they get screened. [PROVIDER] One [barrier] is reimbursement, billing ... Reimbursement would make it easier, not just for the behavioral part but also the screening. [PROVIDER] [We rely on funding from the] CDC. The responsibility needs to come out of the community and you don't need to be sucking at the federal tit or the state tit, you need to be independent. Look at all the money the gay community pours into AIDS walks and all of that. The issue isn't the money, it's there, but we have become so dependent on the government. [PROVIDER] The administrative issues are related to a lot of things: it's related to not enough money to hire a nurse, social worker, case manager, somebody that does that sort of education. Administrative issues are related once again to time, because the MD and NP don't have a lot of time. Neither do the social workers that currently work in that place. Then I think there is an administrative will to determine that this is important enough to spend your resources on. Whether it's time or money or space, people are going to have to have some space in order to do this counseling. They are going to have to be willing to have patients come back, on a fairly regular basis, to get some support or counseling as they go through the change process; it is an investment. [KEY INFORMANT] The preceding interview segments have helped to better understand what clinicians believe are the administrative barriers to implementation of behavioral interventions. These segments have also helped to link barriers such as time and monetary concerns to larger administrative concerns. The buy-in of 98

PAGE 110

the administration seems to be important for implementation of new interventions as well as the macro level social and political environment. ICD-9 Coding Six clinician-participants and three key informants were asked about the usefulness of ICD-9 coding for the implementation of behavioral interventions. This was a useful probe because it further explores the importance of reimbursement and monetary concerns that clinicians might view as barriers. Also, the ICD-9 coding would help the clinicians to justify time spent doing behavioral interventions, therefore this probe gives further insight into whether time spent is really a major barrier for clinicians. In a larger context, this probe helped us to understand how structural barriers can affect the individual actions of clinicians, and whether these barriers were being addressed in the curriculum. The clinician-participants were evenly split between 'Yes', 'No', and 'Maybe' as to the effectiveness that ICD-9 coding would have on the implementation of behavioral interventions. Researcher: Would coding on the ICD-9 help you in implementation of behavioral interventions? Two participants felt that the cpding would be helpful, citing that it would alleviate reimbursement issues, one contending that it had been used effectively for tobacco counseling. You betcha! It puts a face on it. We all talk about doing things that we don't get documented ... if there was good coding for them and user friendly documents to check off, and if it was driven by reimbursement [it would be easier]. [PROVIDER] 99

PAGE 111

Yes it would help, there's a similar thing that is used for tobacco counseling. [PROVIDER] This participant thought the coding might be helpful, but then noted that health maintenance can already be coded for and was not used frequently at her clinic. This clinician seems to lack buy-in that the behavioral interventions were valuable and necessary, which was keeping him/her from doing them regularly. It might, health maintenance now is an ICD-9 code, so you can bill for it. When I was a resident that was one of the big pushes, that health maintenance had to be on your problem list regardless of what they came in for. They may have come in for a broken foot, but you had to put health maintenance. It may be someone young that's healthy that broke their foot but you still have to write 'Advised patient to wear sunscreen, to wear their seatbelt, not drink and drive' ... the sort of things that you think are common sense, but have to document so that you can get reimbursed, regardless of why you came in. Now I don't know how much that is done, you can try, it's always on my problem list, but it doesn't necessarily mean that I really address anything. Most of the time I address immunizations or colon screening, or pap or prostate, sort of more standard basic things. [PROVIDER] This participant felt that a more important barrier for clinicians was the lack of importance or understanding of the behavioral interventions than reimbursement and time justification concerns I don't know, maybe. The key point is trying to get the providers to understand the importance of behavioral interventions, until you get that, the codes won't matter. [PROVIDER] Again with this clinician, the ICD-9 coding was not considered effective. 100

PAGE 112

Honestly I usually just circle HIV and don't go through all of the issues, which is something that maybe I should be doing, but that probably wouldn't change it for me. [PROVIDER] Key informants were mixed on their feelings about the effectiveness of coding for reimbursement. In the first interview segment, the informant believed that the 'time is money' issue would be helped by the coding, allowing the clinicians to get paid for and justify their time spent. Oh sure, because any time a clinician can demonstrate 'I'm doing this and it's a payback issue' than it becomes more tolerable for them. 'If I can get some reimbursement for this than I will definitely spend some time doing it', which makes it sound very mercenary and I don't mean for it to sound that way, but time is money in these practices, and the payback for Medicare/Medicaid is minimal anyway. If your practice is at risk of going down the tubes because of funding, then yeah, I'm going to fix the ingrown toenail and not ask about your sex life. [KEY INFORMANT] The next two key informants didn't believe that the coding could be effective, citing that clinicians didn't tend to feel strongly about reimbursement in this context, and the behavioral interventions would not be valued highly by the insurance companies. It would help ... [but] clinicians already think that this is their responsibility, and I don't think many of them will list reimbursement as a high priority for what would get them to do it. [KEY INFORMANT] Counseling has never been valued enough by insurance companies to allow providers to bill for their time and effort. If they could bill for it they would probably be much more receptive, but it is still difficult subject matter. [KEY INFORMANT] 101

PAGE 113

Time and money are in many ways combined structural barriers, especially when time spent needs to be justified for monetary compensation. Despite this claim, clinicians and key informants were tom as to the effectiveness of the coding system as a method of barrier alleviation in this respect. The more important issue seems to be the buy-in of the clinicians as well as the buy-in of the insuranc.e companies and public health facilities that behavioral interventions were an important use of the clinician's time. Training Fifty percent (N=12) of clinicians in the long-term pilot survey reported that they felt they needed more training. Training in the form of continuing education and classes for practicing clinicians can lead to an increase in knowledge and a decrease in discomfort, enabling clinicians to better implement . new interventions. Training can address structural barriers in a way that allows the clinician to better understand these barriers and know ways around them. Training during medical school is also an important factor that can help to alleviate future barriers. Both types of training were addressed in the following interview excerpts. This first clinician addressed the issue of a lack of training in the topic of behavioral interventions within biomedical education. Really [there's] no good organized training that physicians get, so each person approaches the situation differently, just based on how they were trained. In general in my training I can't remember having anything like a role play that would make it easier and make me more comfortable asking questions related to sexual activity ... So, it's very individual how you build up the skill: some people might ask about sexual activity, some 102

PAGE 114

people might ask about sexual preference, some might go into detail about type of sexual activity and practice and some not. [PROVIDER] The next clinician addressed his/her lack of qualification to perform behavioral counseling. S/he felt that 'a lot of intricate training' needed to take place in order to deal properly with the risk behaviors of some high risk populations. While s/he had attended the training curriculum addressing behavioral interventions, s/he still felt that more training was needed, maybe beyond the scope of his/her biomedical practice. I don't feel qualified to do a lot of the behavioral I think there's a lot of stuff that goes into why they are doing things and how to help them modify their behavior. I don't feel qualified to do that, I can tell them to stop, but that's not really the root of it. Some of them still don't have any insight into it; they don't know why they do it. I think that it takes a lot of really intricate training. They may have really big psychological problems. I mean, the substance abuse plays into it for a reason. It leads to the behavior, but again, why they drink so heavily, or why they do drugs, or why they do anything, I have no insight into it. I can diagnose depression and that's about it, I'm not trained in that. [PROVIDER] The next two clinicians felt that training can be effective, whether the clinicians were aware of their barriers or not. Training was effective at raising awareness and implementing things; I was aware of the barriers going into it. [PROVIDER] I've seen where docs come off really preachy, really 'I know best', and they don't feel that they really need to learn new techniques ... And I think it's about training; I think if we could actually get docs in training to see that, we could teach them a way to do all of that. It's just a matter of tweaking a few 103

PAGE 115

things; the problem is change is hard for a lot of people. [PROVIDER] A key informant commented that studies have shown that training is effective at changing clinician behavior and increasing comfort level. While many clinicians agreed with this assessment, other clinicians such as the one quoted above felt that the curriculum did not qualify them to deal with the more serious behavioral issues with their patients We looked at the data and docs, nurses, NP' s all had an increase in the amount of time they had discussions, so the training works for everybody. We presented the data on the providers, 3 months from the prevention with positives course, providers became more comfortable after they were trained. The big point is that they felt more comfortable and asked more questions ... We trained a broader base than just docs and nurse practitioners, because we have nurses not just NP's but nurses that are primary providers. The patient may have a relationship with the social worker than they do with me [a MD]. If the social worker has been trained and knows about this concept; they may be able to deliver these prevention messages and change behavior better than I because they have a better relationship. It's important I think to include the entire clinic .. Are the docs going to dismiss this as unimportant? The answer is 'no'. We actually looked at the data, and docs, nurses, NP's all increased the number of times they had discussions with their patients about risk. Training, education and a little bit of practice work for everybody. [KEY INFORMANT] Referral Resources Twenty-nine percent (N=7) of clinicians in the long-term pilot survey felt that they lacked referral resources if the patient had inadequate insurance coverage. Some clinician-participants in the interviews also felt that they 104

PAGE 116

lacked referral resources, especially for patients that need ongoing or in-depth assistance. The issue of time came into the discussion, as some clinicians felt that they didn't have time to properly address the complex behavioral issues that some of the patients present. They became hesitant to bring up behavioral risk issues during the medical encounter because they lacked the referral resources to plug the patient into when a complex problem arose. At the beginning of the interview, this clinician-participant was asked what the main barriers were to implementing behavioral interventions with their HIV -positive patients. Lack of appropriate resources was the second barrier cited by the clinician. My own personal barriers would be time, second would be my frustration about lack of appropriate referral resources, third would be my own biases that may contribute to me glossing over the subject, or not even addressing it in a consistent manner, and fourth would be that we don't have a protocol or there's not institutional mandate, or it's not part of our CQI. I'm not audited, ya know, things that would keep me consistently screening people. [PROVIDER] Researcher: Resources in what areas? I know we have PCRS [Partner Counseling and Referral Services] available, but what we don't have is adequate therapists to deal with patients that need that level of support. Neither do we have PCM [Prevention Case Management] [for those] that need ongoing assistance. [PROVIDER] Once again, the issue of a lack of resources came up when this clinician was asked about a time when it was difficult talking to a patient about sex and drug issues. The clinician also mentioned having a lack of time and clinical skills to counsel the patient. This vignette was reproduced in its entirety in 105

PAGE 117

order to show the magnitude of the behavioral issues that the patient had presented and the reaction of the clinician to these issues. There was a patient I saw today, who I have known for a long time, who is very socially isolated. His issues were he wanted counseling about safer sex practices, he wanted to talk to me about the different things he is doing with his new partner, and he wanted me to quantify for him the risk associated with each behavior. He told me that he had a separate need of support around disclosure to this partner, and that he hadn't told. He needed to talk because he was in love with this person, but the person wasn't committed to him because the [the partner] was heterosexual ... [His partner told him]'I can't make a commitment to you'. He was having physical manifestations of stress, and he wanted support, but I don't have the time, or clinical skills to counsel him ... And he's a MEDICAL patient, so I really didn't have a counselor that I could plug him in to, or make a commitment to find people for him to work with. I could immediately find information about risk reduction, and I could tell him that there were people that could work with him around disclosure ... But what I could not do for this man is feel that I connected him with a clinician that could provide insight and skill to deal with this situation that he said is killing him ... [laugh]. .. in a way that the HIV is secondary. That's very hard to accept. [PROVIDER] While this next clinician had some mental health resources available, s/he felt that patients sometimes did not properly utilize these resources. S/he also believed that there were not enough of these resources available. Researcher: Do you have a lot of referral resources for the more behavioral aspects? We have mental health; we're pretty booked, most of my patients see them [mental health] every few months. We don't have a lot of weekly or biweekly talk therapy. We could use 106

PAGE 118

more of these resources. Mostly it is the patient who doesn't really believe in that or want to do that. [PROVIDER] When asked about time as a barrier, this clinician discussed the need to have providers available who deal specifically with behavioral risk issues instead of relying on the clinician. Researcher: Do you feel that if you were given more time during the actual visits that that would help the issue, or would something else be better? If you had someone designated to do harm reduction, and met with clients depending on their need; that probably would work much more effectively than trying to bring it into a busy visit when there are other things going on ... Getting another HCP [Health Care Provider] involved to focus more specifically on those issues would be my solution. [PROVIDER] Another clinician transitions from discussing time as a barrier to discussing the need for specialists to deal with psychological and social problems. I don't have time, and sometimes it's truly enlightening to have them deal with these psychiatric issues. The psychological or social worker will have a very detailed history of what psychological problems they have, and whether they have ever been abused, etc ... That would have never come out in a visit. [PROVIDER] 107

PAGE 119

A key informant also felt that while clinicians should bring up the topic, a specialist should do detailed behavior change discussions. This informant felt the curriculum should give clinicians the skills to address behavior change topics, but should also give them the skills to identify and contact appropriate referral resources for the patient. They need to bring up the topic; they need to do some risk assessment. But when it comes to the behavior change discussions, counseling and that kind of thing: they need to then be able to say, 'I want you to go talk to this person': a nurse, a case manager a social worker, a health educator to discuss these issues so that we can help you be safer, healthier and less risky. [KEY INFORMANT] Another key informant addressed this issue by explaining that a lack of referral resources to deal with complex patients was creating a reticence among clinicians to bring up important issues. They [the clinicians] feel that ifthey start asking questions they will open Pandora's Box and find out all of these horrendous things that they are not going to know how to deal with or have resources to deal with ... 'So I'm not going to ask'. We don't have this attitude toward other things such as substance abuse; we don't not ask about substance abuse Lack of referrals is a big barrier; lack of support with complex patients. [KEY INFORMANT] 108

PAGE 120

CHAPTER6 DISCUSSION Introduction This study has identified and explored many barriers and challenges that medical providers who serve persons living with HIV face in incorporating behavior change interventions (i.e. STD and HIV prevention) into medical care with their HIV positive patients. All of the participants in this study who discussed the impact of the training did so in a positive manner; the participants discussed all of the training topics, exercises, and materials in a positive light. Despite this trend, many participants felt that they needed more training in order to implement the guidelines. Other participants felt that they were not qualified to implement these behavioral interventions, which also indicates a need for more training. In addition to these direct assertions by some participants about training, most participants interviewed expressed various and continuing barriers to implementation of the guidelines after having taken the course. These factors that emerged from the interviews indicate that more training may possibly be needed in order to help clinicians overcome formidable barriers. The presence of continuing barriers also indicates that certain areas may need more emphasis in the curriculurri or a change in training format in order to be more effective. This study provides insight into whether the barriers and challenges that clinicians face to implementation of behavioral interventions with their HIV positive patients are primarily social or structural in nature. Some of these barriers are inherent to the current structure of biomedicine in the U.S and a change in the biomedical model is necessary in order to fully alleviate these 109

PAGE 121

common barriers to health communication. Other barriers are primarily social in nature, encompassing the personal beliefs, values, and attitudes of the clinicians and interactional dynamics of provider-patient encounters. A discussion follows surrounding the study results and existing literature in the areas of structural and social barriers, health communication, and quality improvement. Diffusion of Innovations theory is used in order to better understand why these interventions may not be easily adopted. This discussion also sheds light on the bigger issue of patient-clinician interaction within the biomedical system. The Interrelation of Structural and Social Barriers In my analysis of the barriers and challenges that clinicians face to implementation of behavioral interventions into the medical encounter with persons living with HIV, the barriers that emerged from the interview data were separated out into two categories: structural barriers and social barriers. While I have found this activity to be useful for classification, I also recognize that some of the barriers that were identified are neither of a completely structural nature or related completely to social issues. Because of the inherent dualism in many of these barriers, they are best understood at times through an integrated view of the structural and social elements of the barriers. Overall, clinicians interviewed for this study reported more social constraints than structural ones In many instances, structural constraints had to be probed for by the researcher. When the participants discussed structural constraints, the context of the discussion often led to social constraints. 110

PAGE 122

Similarly, discussions of social constraints often contained structural elements. The researcher found that while social barriers seem to be more important to participants, structural barriers are also present. Also, many issues discussed during the interviews involved a mixture of social and structural elements. Below I have outlined several situations discovered in the interview data in which structural and social barriers were in some way combined or related. As mentioned before, discomfort in one form or another was the most common barrier to implementation of risk discussions that came up during the interviews for this study. An overarching structural barrier, which has created an environment that fosters discomfort, can cause this social barrier. For instance, some institutional policies that were discussed by clinicians and key informants in the interviews, like a policy prohibiting discussions about masturbation, abstinence only campaigns, 'just say no' drug campaigns, and the illegal status of drug use, can make discussions about risky sexual behavior and drug use very uncomfortable for the clinician. In some cases, these discussions could lead to litigation or other consequences for the clinician engaging in them. Stigmatization of behaviors or disease, especially related to HIV transmission, can also be created by these institutional and political policies. In this way, a structural barrier is not only in place but has created a social barrier, thus compounding the problem and decreasing the likelihood of behavioral interventions taking place. Another compounded barrier related to discomfort surrounds the training of the clinician. A lack of training in implementation of behavioral interventions leads to a lack of knowledge on important issues and techniques, thus creating discomfort for the clinician. A lack of proper referral resources can make the clinician hesitant to engage in difficult or detailed conversations Ill

PAGE 123

about risk behavior with their patients that they won't have the ability to follow upon. A lack of time may seem at first to be a wholly structural barrier; but the issue of time is also related to the individual decisions made by the clinician as to what is important to cover during their clinical encounter. If the clinician does not believe that behavioral interventions should be part of their clinical role, than they are much less likely to allow time for these activities. Along with experiencing this very individual barrier, the structural reality of limited time during the clinical encounter is an issue that is on the macro-level of institutional policy, which the clinician has little control over. Time constraints can also create discomfort for the clinician by placing a strain on patient clinician communication. This is especially true when patients are not given enough time to discuss the issues they feel are important to them, and the clinician is not given enough time to focus properly on a behavioral intervention. Many of the clinicians interviewed for this study felt that behavioral interventions were not an essential part ofbiomedical care; they didn't feel that these interventions were part of their role as biomedical clinicians. While the perceived role of the clinician lies in the deeply held beliefs and values that the clinician holds about biomedicine, these beliefs originated from the institution of biomedicine. In this way, beliefs about the role of the clinician have both social and structural components. 112

PAGE 124

The Role of Health Communication In order to overcome the common barriers clinicians face to implementation ofbehavioral interventions with their HIV positive patients, an improvement in health communication and the interpersonal process must take place. Despite the fact that many of the clinicians interviewed for this study believe that their role as a healthcare practitioner is first and foremost a biomedical one, many of the tasks that need to take place during the biomedical encounter involve health communication and a strong use of interpersonal skills. It is impossible for the biomedical practitioner to have a fully successful medical visit with a patient without communicating about sociocultural issues. The role of the biomedical clinician involves understanding and addressing the sociocultural factors that cause disease as much as it does the treatment of the physical ailments of disease. Behavioral interventions related to STD/HIV prevention given during the medical encounter should be considered by all HIV providers to be an essential partofbiomedical care. This discrepancy can be addressed in the training curriculum, but it should also be addressed in the training received in medical schools. As mentioned previously, lack oftime emerged as a main structural barrier to implementation ofbehavioral interventions during the medical encounter for those clinicians interviewed. However, the behavioral interventions presented in the training course were designed with clinical time constraints in mind, and can be delivered within a few minutes. Although the clinician may not be able to control the amount of time he or she is allotted with each patient, the clinician does have the ability to increase the effectiveness of health communication by not appearing 'rushed' or rushing the patient. This atmosphere will facilitate a 113

PAGE 125

more thorough risk discussion with the patient, in a minimal amount of time since the patient will feel comfortable and will hesitate less. One clinician interviewed noted that a risk discussion could easily take place throughout the medical visit during the examination and other processes. Behavioral risk discussions can be more easily delivered if health cornmunication becomes integrated into the duration of the visit, instead of as a 'special segment' added onto the beginning or end of the visit. With health communication taking place throughout the medical visit a lack of time is no longer a barrier to communication of behavioral interventions. Quality Improvement Studies Revisited Despite the demonstrated efficacy of short behavioral interventions during the clinical encounter (Kamb et al.l998), the dissemination and adoption of guidelines in clinical practice remains a challenge. In order to overcome this challenge, barriers to adoption must be evaluated in detail and from different angles. The analysis completed in this study addresses some of the assessments recommended by Bero (discussed in Chapter 2 above): it identifies some of the external barriers to change that clinicians face, and it helps to assess the preparedness to change of the clinicians. While external or structural constraints were not emphasized in participants' statements as niuch as social constraints, external or structural barriers to change were nevertheless present with most clinicians interviewed. Structural barriers such as administrative barriers to change, a lack of or inadequate referral resources, and a need for more training were mentioned by few participants, while a lack of time was mentioned by almost all. 114

PAGE 126

Historically, and in this current study, time was mentioned as a main barrier to change. This historically prominent structural barrier is difficult to change because it is part of the existing structure of the biomedical system. The study results also facilitate an assessment of the preparedness of clinicians for practice change. Since the interviews for this study were done a couple of months after the clinicians attended the training workshop, participants had time to return to their clinical practice setting and experiment with implementation of the new behavioral interventions. The detailed social barriers that were reported in the interviews allow for an assessment of the level of preparedness that clinicians have for practice change after taking the course. Clinician discomfort was the most reported and elaborated upon barrier in the interviews. This barrier is an indication that clinicians may not feel personally prepared to implement these interventions. The need for more training on these issues expressed by some of the participants also indicates that the clinicians may not be fully prepared for their own behavioral change. Wyszewkianski and Green (2000) assert that knowledge-oriented training that involves dissemination of information through continuing education and guidelines may not be fully effective at changing provider clinical practice. They believe that through the use of researched behavior change strategies, practice change is more likely to happen. Fully didactic sessions have also been found to be less effective than educational sessions that involve high levels of participant activity (Davis et al. 1999) Behavior change strategies, especially interactive ones, may be especially useful at alleviating the discomfort that clinicians feel with implementation of behavioral interventions. These strategies are used in the current training, and should continue to be used. 115

PAGE 127

An Analysis of Results through Diffusion of Innovations Theory One behavior change theory that has proven very useful to quality improvement research, and is continuously cited within the quality improvement body of literature, is Everett Roger's Diffusion oflnnovations theory. As discussed in the review of literature, Diffusion of Innovations theory is useful because it assists researchers in understanding facilitators of the adoption of an intervention as well as barriers to adoption of the intervention. With this theory, researchers can begin to understand why an intervention is adopted or not by a certain community or with a certain target group. This study is an evaluation of one aspect of a curriculum created in response to guidelines disseminated by the CDC (Centers for Disease Control and Prevention). The guidelines were created based on epidemiological studies showing a need for HIV prevention among persons living with HIV, and the reported efficacy of short prevention messages delivered by clinicians to HIV infected patients during the clinical encounter. Training is used in this intervention because it is a proven method for facilitating implementation of important clinical guidelines, as the studies in the review of literature show. Participant response to training is therefore an important theme to revisit. What this study tells us is what barriers participants experienced when they returned to their workplaces after training to implement what they had learned, and how participants felt about those barriers. This gives us an idea of the ways that the guidelines were adopted in clinical practice and what barriers are still in place. 116

PAGE 128

The Five Attributes of Innovations The five attributes of innovations within Diffusion of Innovations theory are used in order to better understand the barriers found in this study. In an analysis ofRoger's attributes of innovations in light of the thesis results, we can begin to understand the barriers that were reported by the clinician participants. Of the five attributes outlined in the review of literature, a lack of relative advantage looms large for this innovation, with some barriers also existing in compatibility with existing beliefs, observability, trialability, and complexity. An analysis of the reported barriers through the use ofthis construct will help with identification of challenging elements of the current innovation. A focus on these challenges in the curriculum development and training processes could make the current innovation more successful. Relative advantage. A low relative advantage is the most important barrier to the adoption of the current innovation. The subdimensions of relative advantage include economic profitability, low initial cost, a decrease in discomfort, social prestige, a saving of time and effort, and immediacy of reward (Rogers 2003:223) Most of these subdimensions of relative advantage are not met by the current innovation. First of all, there is no economic profitability related to the implementation of the behavioral intervention. While there is no initial cost to the implementation of the guidelines, the lack of incentive is an issue. Some clinician participants felt that ICD-9 coding for reimbursement for the behavioral intervention would be an incentive for implementing the innovation. Rogers notes that incentives, monetary and otherwise, have been successful in 117

PAGE 129

speeding up the diffusion of innovations in the fields ofhealth, medicine and family planning (Rogers 2003). The addition of an incentive for clinicians may be useful for the present innovation. Alternately, a mandate may help to increase the rate of adoption and overcome this barrier. Mandates for adoption are a mechanism through which the system exerts pressure on an individual to recognize the relative advantage of an innovation (Rogers 2003: 240). Mandates are particularly useful in preventive innovations, such as the present innovation, because the relative advantage of preventive innovations is characteristically low. A decrease in discomfort is an important subdimension of relative advantage, however discomfort in one form or another was the most frequently discussed barrier in the interviews. Clinician discomfort with discussions of sex and drug use is probably the most important barrier to the implementation of the current innovation. Discomfort is an unfortunate byproduct of the guidelines that is caused by the nature of the behavioral intervention message proposed. Because of this, it is important for the training to focus heavily on all aspects of this important barrier. As outlined in the results section of this report, clinicians had discomfort for a variety of reasons. First, clinicians felt discomfort with discussions about sex and drug use, which stemmed from their lack of knowledge about the subject, as well as the illegal status of drugs and the taboo nature of sex discussions. Clinicians also felt as if they were unfairly judging their patients by engaging in the risk discussions. The social stigma of HIV naturally caused discomfort surrounding the risk discussions, as did the clinician perception of patient discomfort with these discussions. These areas of discomfort should be focused on in training on these guidelines. 118

PAGE 130

While social prestige ca!l increase the relative advantage of adoption of an innovation, there is no social status advantage to the implementation of the guidelines in this study. The behavioral intervention happens in a private setting, the medical encounter, which is an interpersonal experience that offers no increase in social status for the clinician. Moreover, some clinicians iiiterviewed felt a decrease in social capital if they were found by the patient to be ignorant of the most current sexual practices and drug use trends. This may compromise the clinicians feeling of status and authority. Another important subdimension of relative advantage involves an innovation saving time and effort for the adopter. The clinicians do not save time or effort in adoption of this innovation. A lack of time was a major theme that emerged from the interviews. The amount of time that clinicians have with a patient has varied throughout history, with less time being allotted in recent years than in the past. The demands of insurance companies for increased productivity are partly accountable for this trend, as is the changing culture of biomedicine and American society at large. American culture has become increasingly fast-paced during the end ofthe 20th and the beginning ofthe 2151 centuries. The current curriculum has taken these structural changes into account with the creation of short behavioral interventions that can easily be disseminated within the time constraints of the modem biomedical practitioner. This being said, clinician-participants overwhel!Dingly felt that time was a barrier to the dissemination of these interventions. In order to address the ongoing concerns of the clinician participants in this curriculum an increased emphasis must be placed on the short amount of time that these behavioral interventions really require 119

PAGE 131

An immediacy of reward can increase the relative advantage of adoption of an innovation However, there is no immediacy of reward for clinicians since the innovation is preventative and behavior change will happen slowly for the patients over time. Preventative innovations, defined as "a new idea that an individual adopts in order to lower the probability of some unwanted future event," characteristically have a low rate of adoption (Rogers 2003:234). This would be the case for HIV I AIDS prevention innovations such as the present one. This is because a preventative innovation has a delayed reward that may or may not happen. Furthermore, the reward may be invisible, such as is the case with not contracting HIY I AIDS (Rogers 2003). In the current study, HIV clinicians will probably never see the product of the innovation since it in many ways affects the partners of HIV positive patients and not the patients themselves. For these reasons, it is important to make the goals of the innovation very specific for the target community when implementing a prevention innovation. Showing the results of the innovation to trainees is another way to unmask the hidden outcome of the innovation. Compatibility. Compatibility is defined as "the degree to which an innovation is perceived as consistent with the existing values, past experiences, and needs of potential adopters" (Rogers 2003: 240). Several aspects ofthe current innovation are important to this area, probably the most important aspect being the issue of change. Rogers asserts, "old ideas are the main mental tool that individuals utilize to assess new ideas and give them meaning" (Rogers 2003: 243). Change is difficult for individuals in any situation, clinicians being no exception to this. 120

PAGE 132

Large changes are more difficult for individuals to implement than smaller changes. One way to understand these continuing barriers is to assess if the clinicians feel that the changes they are being asked to implement are very large. As discussed in chapter 2, Davis et al.(1997), using Everett Rogers' Diffusion of Innovations theory found that if clinical guidelines represent a large change to the existing beliefs and values of the clinician, and/or are large changes to the existing structural aspects of the clinical environment, than they are less likely to be adopted. The current curriculum was developed to represent a small change in clinical practice, and one that is easily implemented. Short behavioral interventions delivered during the clinical encounter should not represent a large change to clinician practice. If the clinicians feel that these changes are large, they are much less likely to be adopted. Clinician discomfort, and beliefs about the role of the clinician are two main themes that emerged from the interviews, which show that clinicians feel that tlie changes they are being asked to make are large ones. Clinician discomfort in one form or another was the most common barrier to implementation of risk discussions that came up during the interviews for this study. Clinicians mentioned many reasons for this discomfort all of which centered on the content of the risk discussions. In light of the fact that the intervention was set up to be a small change in practice, with the addition of short behavioral interventions, this level of discomfort brings up an important question: Do clinicians feel that this is a big practice change? Is this partly why they feel the high level of discomfort that they do? If this intervention represents a large change in the beliefs, values, and routine practice patterns of the clinicians, it is possible that this causes discomfort with these discussions. The curriculum should take time to address the beliefs and values of the 121

PAGE 133

clinicians in an attempt to alleviate some of this discomfort. Trainings tailored toward these prejudices and biases would be helpful in overcoming this comfort barrier. The more knowledge and practice dealing with these issues that the clinician possesses, the more comfort they will feel addressing these difficult issues. Clinicians' beliefs about the role of the biomedical clinician in behavioral interventions was another major theme that emerged from the interviews. In the interviews, many clinicians felt that behavioral interventions were simply not part of their role as a clinician, emphasizing that it is not an essential part of biomedical care. Clinicians also felt that they lacked the proper qualifications to perform the behavioral interventions, and lacked the proper referral resources for complicated behavioral issues that arose. This barrier, like discomfort, is a social barrier that stems from the beliefs and values held by the clinician, in this case, the beliefs and values surrounding biomedicine. As predicted by Diffusion oflnnovations theory, if this intervention requires a large change in the beliefs and values of the clinician surrounding biomedicine, then this may account for the persistent role barrier. Historically, biomedicine or allopathic medicine has focused on the treatment of physical manifestation of disease. "Phenomena relating to health and sickness only become 'real' when they can be objectively observed and measured under [scientific] conditions" (Helman 2001 :79). Physical abnormalities and physical states of disease and health have been the focus of biomedicine instead of the social and cultural background of the patient (Helman 2001). The current intervention requires clinicians to conduct behavioral interventions with their patients. These interventions are related to the sociocultural risk factors that the patient may possess, and are not traditionally a focus of the biomedical 122

PAGE 134

practitioner. In this way, the intervention is a large change from the traditional biomedical practice. This being said, the last few decades of biomedicine have seen many successful additions of behavioral interventions to the biomedical encounter, including those related to sunscreen use, smoking cessation, and proper diet to name a few. By reviewing these successful interventions, the current curriculum could possibly improve dissemination strategies. Observability. Innovations are diffused faster if the results have a high degree of observability. Clinicians give the prevention with positives message in this study to patients during the medical encounter, but clinicians do not have a chance to readily observe other clinicians during the medical encounter, and do not have the ability to directly observe the intervention taking place. Role plays, which take place during the training, help to overcome this barrier. Alternately, video footage of a clinical encounter in which the identities of the individuals were concealed would be a way for clinicians to experience the guidelines in real practice. Tria/ability. The trialability of an innovation relates to the ability of the innovation to be experimented with on a limited basis, thus reducing the long term commitment and/or effects of the innovation. While not a major concern in this intervention, trialability explains some of the discomfort barrier that many clinicians felt. The ability to 'try out' the innovation without possible negative consequences can reduce discomfort with the initiation of an innovation. Some clinicians were concerned that discussions of sexual practices and drug use with their HIV positive patients would be met with a negative 123

PAGE 135

reaction. A few of the clinicians had actually experienced a negative reaction concerning these discussions from one or a few of their patients. The clinicians were concerned that a negative reaction would damage the long-term relationship with the patient, thus decreasing the trialability of this intervention. These feelings can cause the clinicians to abandon future experimentation with the intervention. In order to combat this factor, trainings should give the clinician tools that they can use in case a negative reaction does occur, and should work to dispel the myth that a negative reaction will dissolve the relationship. In most cases, a negative reaction from the patient can be resolved with a brief discussion about the importance of the intervention. Complexity. The current intervention was designed as straightforward, short messages that cari be easily delivered by the clinician during the medical encounter. The level of complexity for this type of intervention should be low. Nonetheless, some clinicians felt they needed more training in order to properly implement the interventions. The need for referral networks was also frequently mentioned, indicating that the clinicians felt they needed help with these messages. While the clinician should be able to deliver these intervention messages within a couple of minutes during the encounter, the majority of clinicians also felt that a lack of time was a main barrier for them. These factors are an indication that the clinicians felt the current innovation is complex enough to require additional elements in order to be properly and successfully delivered. In order to alleviate these concerns, the trainings should emphasize the most important elements of the innovation, removing extraneous information that may be helpful, but not necessary. 124

PAGE 136

CHAPTER 7 CONCLUSIONS Implications ofthe Project This project has various implications for understanding and practice. First and foremost, it is hoped that this study will help the National Network of Prevention Training Centers (NNPTC) in the curriculum development and evaluation process. This project has helped to develop a better understanding of the interrelationship of social and structural barriers to practice changes. This analysis has redefined key barriers by identifying the interrelationship of the social and structural elements present within the barriers. Recognition of thl.s interrelationship between these barriers will help to create more effective trainings and intervention methods that target all of the various aspects of these important barriers. This project has also shed some light on the bigger issue of patient clinician interaction within the biomedical system. Many of the barriers and issues that emerged from the interviews are felt in all areas of biomedicine, and are not simply limited to HIV care. Many other types of clinicians face the issues of a lack of time, discomfort with difficult topics, and conflicting beliefs about their role as a biomedical clinician. Finally, an analysis of the barriers through the use of constructs found in Diffusion of Innovations theory has helped to understand these barriers and how they can be addressed. Diffusion of Innovations theory has proven very useful for this study in understanding the specific barriers faced by preventive innovations. 125

PAGE 137

Recommendations Perhaps the largest concern related to this intervention is whether it will be widely adopted and implemented. First and foremost is the question of whether awareness of the innovation will diffuse widely. There are a few factors that may help to increase the diffusion rate. First, the researcher recommends a more intense focus on 'how to' knowledge in the training, as opposed to a focus on 'awareness' knowledge. This will help the clinicians to work through their barriers more efficiently in the short time allotted for the training. Having a short film of an actual model clinical encounter with a provider using the innovations would be a useful tool during the training. Testimonials from clinicians that have successfully implemented the innovation may also help, as could testimonials from HIV positive patients that have positive feedback on their experiences with the innovation. An effort should also be made to recruit what Everett Rogers refers to as 'early adopters' for the trainings. Early adopters are individuals within the community, in this case HIV clinicians, that are an integrated part of the social system and have a very high level of opinion leadership (Rogers 2003). These individuals help to trigger adoption of innovations in the critical mass or majority of the target audience. In a preventive innovation such as the present one, targeting the early adopters becomes especially important because of the many barriers that are common with adoption of preventive innovations (Rogers 2003). A focus on recruitment of early adopters for the trainings will help to overcome hurdles since early adopters are much more likely than the majority to adopt a preventive innovation that has an innate degree of uncertainty with the outcome. With successful adoption taking place among the early adopters, the 126

PAGE 138

majority will have examples to follow and the adoption process will move forward more quickly. In addition to the influence of early adopters, other motivators could prove useful. The addition of incentives, or possibly a 'cue to action' could be helpful with overcoming the potentially weak motivation to perform a preventive innovation (Rogers 2003 ). Once adoption occurs, the question arises as to whether the adopted idea will then be implemented in practice. This is perhaps the most difficult step in diffusion of innovations because it requires overt behavior change in the individual (Rogers 2003). The researcher believes that implementation of this curriculum into the practice patterns of trained HIV clinicians will happen, but it will take even with the early adopters. This prediction is based on the fact that the diffusion process is usually substantially longer for preventive innovations such as the current innovation. Failure in implementation can arise if the individual tries to implement the innovation and then encounters a large barrier in the process that they are unable to resolve themselves. In order to help the implementation process, the training could include information on how to access help if questions or barriers arise for them Active information channels available to past participants could consist of trainers and experts that the clinicians could access when and if and individual barriers arise in their practice. Limitations of the Study One of the major critiques of qualitative research is the lack of structure and scientific rigor in the semi-structured interview process, thus creating bias on the part of the interviewer. In this study, probes were used occasionally to elicit 127

PAGE 139

information about structural barriers from the clinician-participants. Moreover, the preliminary survey data used to create these probes were limited. As a pilot, the course was in the process of being revised, and inevitably had 'bugs, that needed to be worked out. The participants interviewed were from a small pool of individuals that participated in the pilot trainings for the course. Many of these clinician participants were 'hand-picked' to attend the pilot and give critical feedback because they were considered leaders in the field of HIV medicine. These individuals may have given a more critical view when interviewed about barriers because they had a higher awareness of potential barriers that they had seen other clinicians have as well as their own. When interviewed these individuals were possibly more likely to discuss barriers in great detail, even barriers that they could potentially easily overcome. Many clinician participants also frequently discussed barriers in the third person, reflecting the fact that they might not have these barriers themselves, but may see these barners in others. This study was originally formulated as a mixed methodology in the hopes that the use of two types of data: qualitative and quantitative would create a more thorough analysis. The quantitative data, however, proved difficult to interpret. Basic comparisons between post-course and long -term data were difficult because the surveys were collected in different ways. The post-course data were collected separately for the three course modules, while the long-term data were collected as a single module. The preliminary survey data were useful for guiding the creation of an interview guide for the qualitative study, but provided limited opportunities for valid comparisons with the qualitative data Fortunately, a wealth of information was received from the qualitative data The researcher believes that the advantages of a qualitative approach namely 128

PAGE 140

the richness of the data possible with this approach, outweigh any potential weaknesses of this method. Future Directions and Research Needs There are more factors to be considered when assessing the barriers and challenges of health communication and risk assessment with persons living with HIV. In order to have a more picture of the barriers to health communication present in the clinical encounter with HIV positive patients, data should be obtained on the beliefs and concerns of the HIV positive patients. More information on providers' impact on HIV positive patient behavior would also be useful. As more studies are accomplished, we will obtain a more complete picture of the barriers and challenges to implementation of behavioral interventions with persons living with HIV. One of the most interesting findings is that the participants in the post course and long-term surveys did not indicate that discomfort was a major issue for them. Despite this trend, issues surrounding discomfort loom large i,n the qualitative data. Moreover, there is very little emphasis on discomfort in the existing literature on barriers. Because of the limitations of the preliminary quantitative data, it is difficult to make definitive conclusions about this finding. A more comprehensive study is necessary in order to better understand this finding. The results of this study show that there are many underlying psychological and social issues that are affecting the behavior and thus the in clinic practice patterns of the health care provider . Social norms and macro level structural issues are also affecting this practice. These issues should be 129

PAGE 141

further investigated in order to better understand not only why discomfort is a barrier for clinicians, but also why clinicians fail to report this barrier in survey format. In addition to collecting more types of data and a larger amount of data, there is also a need to analyze the data using different methods and theories. A cross-disciplinary analysis of this existing data would be useful, incorporating concepts and theories found in psychology, sociology, public health research, and anthropology. The plethora of integrative theories available in health education research, one such example being Social Cognitive Theory, would be useful in this cross-disciplinary analysis Likewise, in addition to the Diffusion of Innovations theory used in this study, other theories of organizational and behavioral change would be helpful. Ongoing evaluation of this training curriculum for clinicians who provide care to persons living with HIV and AIDS is necessary for the continued and successful capacitation of these professionals. This ongoing evaluation is useful in the guidance of needed changes and improvements to the format and content of the curriculum. 130

PAGE 142

APPENDIX A Interview Question Guide Part one: Background information about the medical provider 1. How many HIV positive patients do you see a week? 2. What is your prinCipal occupation? 3. What kind ofmedical setting do you work in? 4. Are you aware of the national guidelines for implementing STD/HIV prevention into the medical visit with HIV positive patients? 5. Have you attended training through the NNPTC on the national guidelines? Part two: Questions on barriers or challenges to implementing the guidelines 6. What do you feel are the main barriers to implementation ofbehavioral interventions related to sex and drug use with your HIV positive patients? 7. Can you share with me a time that sticks out in your mind when talking about sexual/drug practices with a HIV positive patient was a challenge? 8. Tell me about your last visit with a HN positive patient. 9. Do you feel there are any administrative barriers to change? What are these? Part three: Key informant interview guide All of these questions relate to the barriers and challenges that clinicians face in implementing the IDSA guidelines and the IDSA curriculum. 1. What do you think the main barriers are for clinicians to implementation of behavioral interventions related to sex and drug use with their HIV positive patients? 2. When clinicians mention 'time' as a barrier, what exactly do they mean? 3. What could be done to change this barrier? 4. The survey data shows that while 'time' is listed as the main barrier, administrative barriers to change' is also frequently listed as a barrier. What do you feel these 'administrative barriers to change' are? 131

PAGE 143

APPENDIXB Preliminary Quantitative Data: Participant Profile Tables Table B.l. Gender of course participants Table B _.2 of participants .,:. . .. : ,. . .... : .. ; . .. 1 " 132

PAGE 144

Table B.3. Primary professional or functional role of participants '. : Table B.4. Number of years practicing in their reported primary profession . ''

PAGE 145

Table B.S. Facility/setting 134

PAGE 146

Table B.6. HIV I AIDS Services This table summarizes the responses of providers were asked which ofthe following statements describes the way in which they most often provide services for HIV/AIDS patients. N=64 for the post-course data, and N=22 for the long-term data. Post-i l Long. course .. term I N.ot applicable/1 do not see HIV/AIDS patients . 7.8% ij 0% I R.efer HIV+ for all care . 6.3% J L ... 13 I P.r!!llary care to HIY+ J ... o/o ..... . : L._ ... I R.efer HIV-+: _\Vhe_ n th(!Y fail __ .----'0-'o/t-'o-'---'--' I .. 9%. I ()f the ease 68.8% L .. ()8}% I 3.1% L __ 4.5% 135

PAGE 147

BIBLIOGRAPHY Barfield, Thomas, ed. 2000 The Dictionary of Anthropology. Malden: Blackwell Publishers Inc. Berg, Bruce L 2001 Qualitative Research Methods for the Social Sciences. Boston: Allyn and Bacon. Bero, Lisa A 1998 Closing the gap between research and practice: an overview of systematic reviews of interventions to promote the implementation of research findings. BMJ 317:465-468. Billings, J Andrew, and John D Stoeckle 1999 The Clinical EncoUnter: A Guide to the Medical Interview and Case Presentation. St. Louis: Mosby. Bradley-Springer, Lucy, with Carol Vojir, and Peter Messeri 2003 Hard-to Reach Providers: Targeted HIV Education by the National AIDS Education and Training Centers. Journal of the Association of Nurses in AIDS Care 14(6):25-36. Bull, SS, with C Reitmeijer and JD Fortenberry 1999 Practice patterns for the elicitation of sexual history, education, and counseling among providers of STD services: results from the gonorrhea community action project (GCAP). Sex Transm Dis 26(10):584-9. 136

PAGE 148

Cabana, Michael D, with CS Rand, NR Rowe, A W Wu, MH Wilson, PC Abbound, and HR Rubin CDC CDC 1999 Why Don't Physicians Follow Clinical Practice Guidelines? A Framework for Improvement. JAMA 282(15). 2003a. Incorporating HIV Prevention into Medical Care of Persons Living with HIV. In MMWR. Pp. 1-24: CDC. 2003b HIV/AIDS Surveillance Report: HIV Infection and AIDS in the United States: CDC. Davis, Dave, with M Thomson, N Freemantle, F Wolf, P Azmanian, and A TaylorVaisey 1999 Impact ofF ormal Continuing Medical Education: Do conferences, workshops, rounds, and other traditional continuing education activities change physician behavior or health care outcomes? JAMA 282(9):867874. Davis, David A, and Anne TaylorVaisey 1997 Translating Guidelines into Practice: A systematic review of theoretic concepts, practical experience and research evidence in the adoption of clinical practice guidelines. Canadian Medical Journal Association 157(4). Devine, Sharon, with Christine Shure, Susan Dreisbach, and Kitty Corbett N.d. Denver STD/HIV Prevention and Training Center: Long-Term Course Evaluation Report for the 3-Day Intensive Clinical Course. Unpublished report, Center for Research in the Health and Behavioral Sciences, University of Colorado-Denver. 137

PAGE 149

Duffus, W A, with M Barragan, and L Metsch 2003 Effect of physician specialty on counseling practices and medical referral patterns among physicians caring for disadvantaged human immunodeficiency virus-infected populations. Clin Infect Dis 36(12): 1577-84. Elford, J, with G Bolding, M Maguire, arid L Sherr 2000 Do gay men discuss HIV risk reduction with their GP? AIDS Care 12(3):287-90. Fisher, WA 2002 The options project: clinician-initiated interVention for reduction of HIV risk behavior among HIV + patients in clinical care. HRSA Annual Meeting, Washington, D.C., 2002. Gardner, LL, with C Krawczyk, and LR Metch 2002 Physician characteristics associated with discussing HIV prevention messages in a multicenter study Paper presented at the XIV International AIDS Conference; 2002. Glynn, M., and P. Rhodes 2005 Estimated HIV Prevalence in the United States at the end of2003. Paper presented at the National HIV Prevention Conference, Atlanta, 2005. Goffman, Erving 1963 Stigma: notes on the management of spoiled identity. Englewood Cliffs: Prentice-Hall Inc Green, Lawrence W and Marshall W Kreuter 1999 Health Promotion Planning: An Educational and Ecological Approach. Boston: McGraw Hill. 138

PAGE 150

Hahn, Robert A, and Arthur Kleinman 1983 Biomedical Practice and Anthropological Theory: Frameworks and Directions. Annual Review of Anthropology 12:305-333. Hehrian, Cecil G 2001 Culture, Health, and Illness. London: Arnold. Henderson, Harold, with V. German, AT Panter, GJ Huba, and C Rohweder 1999 Systems change resulting from HIV I AIDS education and training: A cross cutting evaluation of nine innovative projects. Evaluation & the Health Professions 22(4):405-426. Huba, GJ, with AT Panter, LA Melichior, D Anderson, and J Colgrove 2000 Effects ofHIV/AIDS Education and Training on Patient Care and Provider Practices: A Cross-Cutting Evaluation. AIDS Education and Prevention 12(2):93-112. Hughes, Charles C 1978 Folk Healers and Ethnomedicine. In Health and the Human Condition: Perspectives on Medical Anthropology. M.H. Logan and J. Edward E Hunt, eds . North Scituate: Duxbury Press. Kalichman, SC, with D Rompa, M Cage, K Difonzo, and D Simpson 2001 Effectiveness of an intervention to reduce HIV transmission risks in HIV-positive people. American Journal of Preventative Medicine 21(2):84-92. Kamb, Mary L, with Martin Fishbein, John Douglas, and Fen Rhodes 1998 Efficacy ofRisk-Reduction Counseling to Prevent Human Immunodeficiency Virus and Sexually Transmitted Diseases. JAMA 280(13):1161-1167. 139

PAGE 151

Kleinman, Arthur 1981 Patients and Healers in the Context of Culture. Berkele y : University of California Press. Kleinman, Arthur 1988 The Illness Narratives: Suffering, Healing and the Human Condition United States: Basic Books. Lalonde, Bernadette, with K Uldall, GJ Huba, AT Panter, J Zalumas and LR Wolfe 2002 Impact of HIV I AIDS Education on Heal t h Care Provider Practice: Results From Nine Grantees of the Special Projects ofNational Significance Program Evaluation & the Health Professions 25(3):302320. Marks G, with JL Richardson N Crepaz, S Stoyano:ff, and J Milam 2002 Are HIV care providers talking with patients about safer sex and disclosure?: A multi-clinic assessment. AIDS 16(14):1953-7. Morin, Steve 2002 Missed Opportunities for Preventing HIV Transmission: A Study of Ryan White Clinics. San Francisco: AIDS Policy Research Cen t er/AIDS Research Institute. NNPTC/AETC 2005 Incorporating HIV Prevention into the Medical Care of Persons Living with HIV: Facilitators Guide, Course Overview. Pp. 6-7: NNPTC/AETC/CDCIHRSA 140

PAGE 152

Oldenberg, Brian, and GuyS Parcel 2002 Diffusion of Innovations. In Health Behavior and Health Education: Theory, Research, and Practice. K. Glanz, B.K. Rimer, and F.M. Lewis, eds. San Francisco: Jossey-Bass. Pappas, Gregory 1990 Some Implications for the Study of the Doctor-Patient Interaction: Power, Structure, and Agency in the Works of Howard Waitzkin and Arthur Kleinman. Social Science and Medicine 30(2): 199-204. Patton, Michael Quinn 2002 Qualitative Research and Evaluation Methods. Thousand Oaks: Sage Publications. Pendleton, David A. 1982 Communication in the Doctor's Office: A Model and a Strategy for Health Communication Research. In Straight Talk: Explorations in Provider and Patient Interaction. L.S. Pettegrew, ed. Louisville, KY: Humanainc. Pettegrew, Loyd S, ed. 1982 Straight Talk: Explorations in Provider and Patient Interaction. Louisville, KY: Hurnana Inc Preston, Deborah Bray, with A D' Augelli, CD Kassab, RE Cain, FW Schultze, and MT Starks 2004 The influence of stigma on the sexual risk behavior of rural men who have sex with men. AIDS Education and Prevention 16(4):291-303. 141

PAGE 153

Rural Center for AIDS/STD Prevention (RCAP) 2005 Stigma as a Barrier to HIV Prevention in the Rural Deep South. Fact Sheet, 16. Indiana University, Perdue University, and University of Colorado. Richardson, JL, with J Milam, and A McMutchan 2004 Effect of brief safer-sex counseling by medical providers to HIV -1 seropositive patients: a multi-clinic assessment. AIDS 18(8): 1179-86. Rogers, Everett 2003 Diffusion of Innovations. New York: Free Press. SansonFisher, Robert, with Jeremy M Grimshaw, and Martin P Eccles 2004 The science of changing providers' behaviour: the missing link in evidence-based practice. MJA 180(5):205-206. Schoenwald, Sonja K, and Kimberly Hoagwood 2001 Effectiveness, Transportability, and Dissemination oflnterventions: What matters when? Psychiatric Services 52:1190-1197. Smith, Jeff N.d. Translational Health Services Research: Pathway from Efficacy to Implementation and Theoretical Underpinnings. Unpublished final project, Department of Anthropology, University of Colorado-Denver. Stoeckle, John 1982 The Improvement of Communication between Doctor and Patient. In Straight Talk: Explorations in Provider and Patient Interaction L.S. Pettegrew, ed. Louisville, KY: Humana Inc. 142

PAGE 154

Thrun, Mark, with W Burman, D Britt and C Reitmeijer 2003 Successful Implementation of Risk-Reduction Discussions in a Primary Care Setting for those Living with HIV. Paper presented at the National HIV Prevention Conference, 2003 Umble, Karl E, and Ronald M Cervo 1996 Impact Studies in Continuing Education for Health Professions. Evaluation & the Health Professions 19(2):148-174. Waitzkin, Howard 1991 The Politics ofMedical Encounters. New Haven: Yale University Press. Wyszewianski, Leon, and Lee A. Green 2000 Strategies for Changing Clinicians' Practice Patterns The Journal of Family Practice 49(5):461-464. 143