Gendered issues in alcohol abuse and dependence among HIV-positive African Americans

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content GENDERED ISSUES
IN ALCOHOL ABUSE AND DEPENDENCE
AMONG HIV-POSITIVE AFRICAN AMERICANS
by
E. Maxwell Davis
____________________________________________________________________
A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(SOCIAL WORK)
December 2007
Copyright 2007 E. Maxwell Davis
ii
DEDICATION
I would like to dedicate this dissertation to one of my favorite people in the universe,
John Thomas Gustin, J.D., L.S.W., (1972-2007). John was a truly good and
wickedly smart guy, a loved friend, a kindred spirit, a dedicated activist and a
veteran HIV social worker. I miss you, your heart, your mind, your attitude and your
perspective on life’s mysteries, big and small. I wish that you were here to celebrate
this accomplishment with me.
iii
ACKNOWLEDGEMENTS
I would like to thank every single member of my big loud family, all of my
wonderful friends and my fantastic committee for helping me get to the completion
point of this sometimes arduous process. The last several years of my life have been
tumultuous ones and I would not have been able to mange the tumult in a way that
allowed me to continue toward a Ph.D. without the practical, emotional and
intellectual support of the people who have loved and guided me along the way.
I want to express my respect and gratitude to everyone at the Drew Center for
AIDS Research, Education and Services for doing what they do every day and for
making me feel a part of their very important work. I want to extend special thanks
to Dr. Frank Galvan at Drew for allowing me to contribute to and benefit from the
conduct of his study and for his innumerable contributions to my entire learning
process. I would also like to thank Dr. Susan Enguidanos for her ever helpful and
always humorous mentorship and crisis management efforts and last but certainly not
least, Lesley Williams for her good cheer, patience, affection and lawyerly
proofreading skills.
iv
TABLE OF CONTENTS
Dedication....................................................................................................................ii
Acknowledgements.....................................................................................................iii
List of Tables...............................................................................................................vi
List of Figures..............................................................................................................ix
Abstract.........................................................................................................................x
CHAPTER 1: The Problem and Its Underlying Framework........................................1
Background of the Problem..........................................................................................1
Purpose of the Study.....................................................................................................2
Specific Aims and Research Questions........................................................................3
Significance of the Problem..........................................................................................4
Methodology.................................................................................................................5
Definition of Terms......................................................................................................5
Organization of the Study.............................................................................................7
CHAPTER 2: Review of the Literature......................................................................10
African Americans and HIV/AIDS............................................................................12
African Americans and Alcohol Use.........................................................................14
Alcohol Use in the context of HIV/AIDS..................................................................19
Gender, Ethnicity and Health.....................................................................................23
Religiosity and Spirituality........................................................................................32
Social Support............................................................................................................37
Social Network Composition.....................................................................................44
CHAPTER 3: Research Methodology.......................................................................50
Research Hypotheses.................................................................................................50
Qualitative Research Questions.................................................................................53
Research Design.........................................................................................................53
Human Subjects Protection........................................................................................55
Sampling Procedures..................................................................................................55
Data Collection...........................................................................................................56
Instrumentation...........................................................................................................57
Data Analysis and Interpretation................................................................................63
v
CHAPTER 4: Sample Descriptions............................................................................73
Description of the Total Sample.................................................................................73
Description of the Sub Sample...................................................................................76
CHAPTER 5: Quantitative Findings..........................................................................82
Bivariate Analyses of the Total Sample.....................................................................82
Multivariate Analyses of the Moderating Influence of Gender.................................88
Bivariate Analyses of the Sub Sample.......................................................................97
Multivariate Analyses of the Mediating Influence of Social Networks...................107
Post-Hoc Power Analysis.........................................................................................119
CHAPTER 6: Qualitative Findings..........................................................................120
Participants in Open-ended Interviews.....................................................................120
Overview of Qualitative Findings............................................................................122
Factors Influencing Misuse in the Context of HIV/AIDS........................................124
The Impact of Alcohol Abuse and Dependence on HIV Self Care..........................135
Factors Critical to Recovery in the Context of HIV/AIDS.......................................145
Issues of Gender, Sexuality and Culture...................................................................156
CHAPTER 7: Discussion.........................................................................................167
Discussion of Quantitative Findings.........................................................................167
Discussion of Qualitative Findings...........................................................................173
Triangulation of Qualitative and Quantitative Findings...........................................176
Implications for Alcohol Interventions with HIV-Positive African Americans.......180
Recommendations for Future Research....................................................................183
Study Limitations.....................................................................................................184
Conclusion................................................................................................................185
References.................................................................................................................187
Appendixes...............................................................................................................207
A. SCID-1 for DSM-IV-TR Alcohol Use Disorders (Lifetime)...................207
B. HCSUS Religiosity & Spirituality Section..............................................215
C. Multidimensional Scale of Perceived Social Support..............................218
D. Social Network Questions.......................................................................219
E. Open-ended Interview Questions.............................................................223
vi
LIST OF TABLES
Table 1. Demographic Characteristics of the Total Sample and Sub Sample............74
Table 2. Religiosity, Spirituality and Social Support in Total Sample
and Sub Sample............................................................................................76
Table 3. Composition of Sub Sample Participants’ Social Networks (n = 148)........80
Table 4. Pearson’s Correlations between Key Variables in Total Sample
(N = 272)......................................................................................................82
Table 5. Bivariate Demographic and Key Variable Comparisons By Gender
(Nominal and Ordinal Data) (N = 272).......................................................84
Table 6. Bivariate Demographic and Key Variable Comparisons By Gender
(Interval Data) (N = 272).............................................................................85
Table 7. Bivariate Demographic and Key Variable Comparisons By Alcohol
Abuse or Dependence Status (Nominal & Ordinal Data) (N = 272)...........86
Table 8. Bivariate Demographic and Key Variable Comparisons By Alcohol
Abuse or Dependence Status (Interval Data) (N = 272)..............................88
Table 9. Hierarchical Logistic Regression Model for the Association of
Religiosity and Gender with Current Alcohol Abuse or Dependence
(N = 272)......................................................................................................89
Table 10. Hierarchical Logistic Regression Model for the Association of
Spirituality and Gender with Current Alcohol Abuse or Dependence
(N = 272)....................................................................................................91
Table 11. Hierarchical Logistic Regression Model for the Association of
Perceived Social Support and Gender with Current Alcohol Abuse or
Dependence (N = 272).................................................................................93
Table 12. Logistic Regression Model for the Association of Perceived Social
Support from Significant Others, Family and Friends with Current
Alcohol Abuse or Dependence (N = 272)..................................................96
Table 13. Pearson’s Correlations between Key Variables in Sub Sample
(N = 148)....................................................................................................98
vii
Table 14. Pearson’s Correlations between Social Network Variables and other
Key Variables in Sub Sample (N = 148)....................................................99
Table 15. Bivariate Demographic and Key Variable Comparisons By Gender
(Nominal and Ordinal Data) (n = 148).....................................................101
Table 16. Bivariate Demographic and Key Variable Comparisons By Gender
(Interval Data) (n = 148)..........................................................................102
Table 17. Bivariate Social Network Composition Comparisons by Gender
(n = 148)...................................................................................................103
Table 18. Bivariate Demographic and Key Variable Comparisons By Alcohol
Abuse or Dependence Status (Nominal & Ordinal Data) (n = 148)........104
Table 19. Bivariate Demographic and Key Variable Comparisons By Alcohol
Abuse or Dependence Status (Interval Data) (n = 148)...........................105
Table 20. Bivariate Social Network Composition Comparisons By Alcohol
Abuse or Dependence Status (n = 148)....................................................106
Table 21. Hierarchical Logistic Regression Model for the Association of
Religiosity and Gender with Current Alcohol Abuse or Dependence
(n = 148)...................................................................................................108
Table 22. Hierarchical Logistic Regression Model for the Association of
Spirituality and Gender with Current Alcohol Abuse or Dependence
(n = 148)...................................................................................................110
Table 23. Hierarchical Logistic Regression Model for the Association of
Perceived Social Support and Gender with Current Alcohol Abuse
or Dependence (n = 148)..........................................................................112
Table 24. Hierarchical Logistic Regression Model for the Association of
Elements of Social Network Composition with Current Alcohol Abuse
or Dependence (n = 148)..........................................................................113
Table 25. Hierarchical Logistic Regression Model for the Association of
Perceived Social Support and Number of Social Network Members
with Current Alcohol Abuse or Dependence (n = 148)............................115
viii
Table 26. Hierarchical Logistic Regression Model for the Association of
Perceived Social Support and Number of Social Network Members
who Encourage Moderation with Current Alcohol Abuse or
Dependence (n = 148)..............................................................................116
Table 27. Hierarchical Logistic Regression Model for the Association of
Perceived Social Support and Number of Social Network Members
Who Are Heavy Drinkers with Current Alcohol Abuse or
Dependence (n = 148)..............................................................................118
Table 28. Descriptive Information about Female Participants in Open-Ended
Interviews (n = 13)...................................................................................120
Table 29. Descriptive Information about Male Participants in Open-Ended
Interviews (n = 27)...................................................................................121
ix
LIST OF FIGURES
Figure 1. Gender as a Moderator of the Influence of Religiosity on Current
Alcohol Abuse and Dependence.................................................................51
Figure 2. Gender as a Moderator of the Influence of Perceived Social Support
on Current Alcohol Abuse and Dependence..............................................51
Figure 3. Social Network Composition as a Mediator of the Influence of
Religiosity and Gender on Current Alcohol Abuse and Dependence........52
Figure 4. Social Network Composition as a Mediator of the Influence of
Perceived Social Support and Gender on Current Alcohol Abuse
and Dependence..........................................................................................53
Figure 5. Alcohol Abuse and Dependence in Total Sample and Sub Sample...........76
Figure 6. Size of Sub Sample Participants’ Social Networks (n = 148)....................79
Figure 7. Sources of Perceived Social Support for Men versus Women
(N = 272)....................................................................................................95
Figure 8. Factors Influencing Alcohol Misuse and Poor HIV Self Care..................123
Figure 9. Factors Influencing Alcohol Misuse in the Context of HIV/AIDS...........125
Figure 10. Five Interrelated Domains of HIV Self Care..........................................135
Figure 11. The Impact of Alcohol Misuse on HIV Self Care...................................138
Figure 12. Factors Critical to Recovery in the Context of HIV/AIDS.....................146
x
ABSTRACT
African Americans are disproportionately impacted by HIV/AIDS and by
many psychosocial issues that complicate their experiences living with this disease.
Of these issues, alcohol abuse and dependence has been linked to accelerated HIV
disease progression, reduced efficacy of HIV medications, poor adherence to
medical regimens and risky sexual behaviors. This study uses the secondary analysis
of quantitative and qualitative data collected from 272 HIV-positive African
Americans in Los Angeles County to investiage this phenomenon. Analysis revealed
that 13.6% of participants met the criteria for current alcohol abuse or dependence.
Logistic regressions testing the influences of gender, religiosity, spirituality, social
support and social network composition on current alcohol abuse and dependence
revealed that social support was uniquely influential. Content analysis of the
narratives of forty consumers with histories of alcohol abuse or dependence
examined factors influencing both the exascerbation of and recovery from alcohol
problems in the context of HIV/AIDS and the impact of these problems on HIV self
care. These findings highlighted the influence of religiosity and sprituality but, again
emphasized the centrality of social support. The qualitative findings also provided a
unique opportunity to interpret the quantitative findings in relation to consumer
perspectives and experiences. Discussion of these findings includes
recommendations for using the influence of social support to lessen the impact of
alcohol misuse on HIV treatment and prevention efforts in this heavily impacted
community.
1
CHAPTER 1: The Problem And Its Underlying Framework
Background of the Problem
Rates of new HIV infection among all Americans peaked in the mid 1980s at
roughly 150,000 new infections per year and since the late 1990s, this rate has
stabilized at roughly 40,000 new HIV infections per year (CDC, 2007). Despite
these gains in preventing the spread of HIV, ethnic minorities in the United States
have been continuously and increasingly disproportionately impacted by HIV/AIDS.
African Americans, specifically, are more acutely impacted than members of other
ethnic groups at every stage of HIV/AIDS (CDC, 2007). African Americans are
more likely to be diagnosed with HIV infection, to be diagnosed with AIDS and to
die of complications from AIDS than members of any other ethnic group in the
United States (Smith, Brutus, Cathcart, Gathe, Johnson, Jordan, Kwaka, Nkwanyou,
Page, Scott, Vaughn, Virgil & Williamson, 2003).
In addition, African Americans living with HIV/AIDS must often cope with
their disease in the context of multiple issues that complicate their efforts to
effectively manage it (Smith et al, 2003). One of these issues is alcohol abuse and
dependence. The co-occurrence of alcohol use disorders and HIV/AIDS has been
associated with numerous medical and psychiatric complications of HIV disease
(Galvan, Burnam & Bing, 2003). These include problems with medication
adherence and less optimal medical treatment outcomes (Petry, 1999) as well as co-
morbidity with other diseases that hasten the progress of HIV disease (Conigliaro,
Gordon, McGinnis, Rabenack & Justice, 2003; Flexner, Cargill, Sinclair, Kresina &
2
Cheever, 2001). Heavy alcohol use by HIV-positive individuals has also been
associated with engagement in intravenous drug use and unprotected sex, behaviors
that carry a high risk for transmission and re-transmission of the HIV virus and other
sexually transmitted infections (Petry, 1999).
Religiosity (Bazargan, Sherkat & Bazargan, 2004), social support (Beattie &
Longbaugh, 1999) and social network composition (Bond, Kaskutas & Weisner,
2003) have all been linked to the etiology and resolution of alcohol abuse and
dependence. Among African Americans, gender has a major influence on these and
other issues related to both alcohol misuse (Caetano, Clark & Tam, 1998; Herd,
1997) and living with HIV/AIDS (CDC 2007b; CDC, 2007c). Specifically, gender is
thought to differentiate African Americans’ experiences with and uses of religiosity
(Coleman et al, 2006), social support (Hicks et al, 2005) and social network
composition (Hough et al, 2005) in the context of living with HIV/AIDS. As such, it
is critical to examine the impact of these issues on alcohol abuse and dependence in
the context of HIV/AIDS in ways that attend both to ethnic and culturally specific
patterns and realities and to the gender-related issues that are nested within them.
Purpose of the Study
HIV-positive African Americans’ abuse of and dependence on alcohol is
clearly likely to complicate both the treatment and prevention of HIV/AIDS in this
highly impacted community. Despite this, little comprehensive research has been
conducted on the incidence of, specific issues and relationships behind or treatment
strategies to address alcohol abuse and dependence in this population. The purpose
3
of this study is therefore to use theories supported by the existing literature on
African Americans, health, HIV/AIDS and alcohol misuse to (1) develop and test
quantitative models of the behavioral issues and relationships that may underlie
patterns of alcohol abuse and dependence among HIV-positive African Americans
and (2) qualitatively explore HIV-positive African American consumers’
perspectives on the reasons for, impacts of and means for resolving their problem
drinking.
Specific Aims and Research Questions
The specific aims of this investigation were to examine among HIV-positive African
Americans (1) the moderating influence of gender on the influence of religiosity and
social support on alcohol abuse and dependence, (2) the mediating influence of
social network composition on the influence of gender, religiosity and social support
on alcohol abuse and dependence and 3) how these patterns are related to HIV self
care behaviors and to potential routes for the cessation of alcohol abuse and
dependence. The quantitative research questions were as follows:
1. Is religiosity associated with a lesser likelihood of alcohol abuse or
dependence for women as compared to men?
2. Is social support associated with a lesser likelihood of alcohol abuse
or dependence for women as compared to men?
3. With greater differences in social network composition, is gender a
stronger moderator of the influence of religiosity on alcohol abuse or
dependence?
4
4. With greater differences in social network composition, is gender a
stronger moderator of the influence of social support on alcohol abuse
or dependence?
The qualitative research questions were as follows:
1. From the consumer perspective, what are the factors that influence HIV-
positive African Americans’ alcohol abuse and dependence?
2. From the consumer perspective, what is the impact of alcohol abuse and
dependence on HIV-related self care behaviors?
3. From the consumer perspective, what factors are critical to the cessation
of alcohol abuse and dependence in the context HIV infection?
Significance of the Problem
The dearth of descriptive or explanatory information about HIV-positive
African Americans’ alcohol abuse and dependence limits understanding of the
patterns and relationships that drive and define it. This issue in turn limits the
development of interventions that might be used to prevent or minimize the negative
psychosocial and health consequences of alcohol misuse in the context of HIV/AIDS
for this highly impacted community. The knowledge generated by this study will
therefore contribute to a clearer understanding of the correlates of HIV-positive
African Americans’ alcohol misuse as well its impact on HIV self care and possible
routes for its resolution. It is hoped that this information can be applied to both
chemical dependency and HIV/AIDS service settings in order to develop
5
interventions that address the needs of African Americans with both HIV/AIDS and
alcohol problems in culturally competent and effective ways.
Methodology
In order to test the proposed models, quantitative analysis was used to assess
the relationships between gender, religiosity, social support and social network
composition with regard to their impact on alcohol abuse and dependence among
HIV-positive African Americans. In order to explore consumer perspectives on
these and other issues related to the onset, impact and resolution of problem drinking
in this population, qualitative methods were used to develop a grounded theory
analysis of these issues. Finally, the separate findings from quantitative and
qualitative analyses were compared and addressed in relation to one another in order
to examine the ways in which they were both complementary and contrasting. This
triangulation of the findings enabled a more meaningful examination of all findings
in relation to the specific aims of the study.
Definition of Terms
Based on the theoretical literature reviewed, both demographic/descriptive
variables and the key variables addressed in the quantitative analysis, (gender,
religiosity, perceived social support, social network composition and alcohol abuse
and dependence), have been operationally defined in the ways most meaningful and
relevant to the analysis at hand.
Living with HIV/AIDS has been operationally defined to include participants
who are HIV-positive but have not been diagnosed with AIDS as well as those who
6
are HIV-positive and also diagnosed with AIDS, either at the time of their HIV
diagnosis or subsequent to it. Gender has been defined as the participant’s self
determination of being either male or female, necessitating the elimination of
transgender individuals from the study sample. Sexual orientation, like gender, has
been operationally defined according to participants’ self-definition, rather than
through an inventory of sexual behavior.
The social support examined for the purposes of this analysis is perceived,
rather than received social support, which allows the analysis to focus on the
participant’s perceptions rather than a number of contacts made or processes engaged
in. The measure used to assess perceived social support also allows the separate
examination of social support from family, friends and significant others, an
important factor due to relevant gender differences in this area. Similarly, in
examining religiosity, both religiosity and spirituality have been evaluated due to the
important theoretical considerations that both interrelate and differentiate these
concepts, especially in relation to African American culture, substance abuse and
gender-based differences related to these issues.
Multiple aspects of social network composition have been measured and
examined for the purposes of this analysis. The aspects of social network
composition included in the current analysis are those for which relevant theoretical
literature indicates the most direct relationship to current alcohol abuse and
dependence. These include the total number of social network members, the number
of social network members who are described by participants as heavy drinkers and
7
the number of social network members who encourage participants to moderate or
abstain from the use of alcohol.
Finally, current alcohol abuse or dependence has been defined according to
the diagnostic terms of the SCID II-R sections on alcohol abuse and alcohol
dependence, which are based on the DSM clinical diagnostic criteria for these
conditions. These criteria focus more on alcohol-related behavior patterns and on the
impact of alcohol use on the participant’s life and functioning than the
quantity/frequency measures more often used to identify heavy versus lighter
drinking. As such, the criteria for alcohol abuse and dependence used in this
analysis are likely both more behavioral and more stringent that those used in studies
that focus on heavy drinkers versus non-heavy drinkers.
Organization of the Study
Chapter 1 presents the background of the problem under study, the purpose of the
study, the specific aims and research questions to be addressed, the significance of
the problem, the methodology to be used and the definition of terms.
Chapter 2 is a review of the relevant literature. It addresses the following topics:
African Americans and HIV/AIDS; African Americans and alcohol use; alcohol use
in the context of HIV/AIDS and its biomedical and behavioral consequences; gender,
ethnicity and health in relation to alcohol misuse and HIV/AIDS; religiosity and
spirituality in relation to HIV/AIDS, African Americans, gender and alcohol; and
social support and social network composition in relation to African Americans,
HIV/AIDS and alcohol abuse and dependence.
8
Chapter 3 presents the methodology used in the study, including the research
questions and hypotheses, the research design, human subjects protections, the
sampling procedures, data collection methods, including descriptions of the
instruments used to measure variables of interest and information on their validity
and reliability, and the plan for data analysis and interpretation.
Chapter 4 presents overall descriptions of the samples used in the quantitative
analysis, including descriptive information about the total sample and sub sample.
Chapter 5 presents the quantitative findings of the study. These findings include
those related to the hypotheses regarding the moderating effect of gender on the
impact of religiosity and perceived social support on current alcohol abuse and
dependence and those related to the hypotheses regarding the mediating effects of
social network composition on the influence of gender, religiosity and perceived
social support on current alcohol abuse and dependence.
Chapter 6 presents the qualitative findings of the study from the content analysis of
open-ended interviews with study participants. These findings includes models
developed from the analysis of consumers’ perspectives on the factors that influence
HIV-positive African Americans’ alcohol abuse and dependence, the impact of
alcohol misuse on HIV-related self care behaviors and the factors critical to the
cessation of alcohol abuse and dependence in the context HIV infection.
Chapter 7 presents discussion, analysis and interpretation of the quantitative and
qualitative findings, including highlights of points of congruence and conflict
9
between the quantitative and qualitative findings and culminating in conclusions and
recommendations.
10
CHAPTER 2: Review of the Literature
Despite the increasing shift of the burden of the HIV/AIDS epidemic in the
United States to ethnic minority communities, the distribution of targeted research
funds has not shifted accordingly. A dearth of HIV/AIDS research focused on and
intervention strategies tailored to African Americans and Latino/as has hampered
efforts to address the specific needs of these communities (Trubo, 2004). The bulk
of HIV/AIDS treatment research, specifically, has been biomedical in nature and has
focused on men who identify as gay or bisexual (Jones, Beach, Forehand & Foster,
2003). This emphasis has precluded knowledge development about the experiences
and needs of the growing proportion of those living with HIV/AIDS who are female
and whose sexuality is not readily categorized according to the definitions used in
dominant U.S. culture (Jipguep, Sanders-Phillips & Cotton, 2004; Wohl, Johnson,
Lu, Jordan, Beall, Curier & Simon, 2002).
Similarly, research on alcohol and alcoholism has been slow to attend to
issues of culture and ethnicity, resulting in the failure of major alcohol research
initiatives to address the influence of cultural issues, misunderstanding of
racial/ethnic differences related to alcohol use (Caetano, Clark & Tam, 1998) and a
delay in the development of culturally specific alcohol interventions (Harper, 2001).
The impact of critical studies on applied fields such as social work, nursing, public
health and health promotion, however, has led to the re-examination of many tools
and methods used in bio-behavioral research. As such, researchers in many
disciplines have increasingly focused on developing knowledge that is both
11
culturally competent and gender specific. These efforts are critical to the
development of models for psychological functioning, identity development and
health-related behaviors that are relevant to and useful for populations previously
marginalized in health research (Cokley, 2005; Jipguep, Sanders-Phillips & Cotton,
2004; Wilson, 1986).
This literature review addresses a population, HIV-positive African
Americans, that has been historically underserved in the HIV/AIDS treatment
literature. It also explores a juxtaposition of issues, such as gender, sexuality and
alcohol abuse and dependence, that have been similarly understudied as well as
others such as religiosity, spirituality, social support and social network composition,
about which more knowledge has been developed. Its purpose therefore is both (1)
to draw connections between these issues in ways that demonstrate support for the
models tested and questions posed in this analysis and (2) to highlight gaps in current
knowledge that complicate and necessitate this inquiry.
Documentation
This comprehensive literature review was conducted using the Medline,
ProQuest, PsychInfo and PubMed databases. In addition, websites of bodies such as
the Centers for Disease Control and the National Institutes of Health were located by
web search and were used to obtain government reports and statistical compilations
via download or email request. Search terms used included African American,
Black, HIV, AIDS, alcohol, alcohol abuse, alcohol dependence, alcoholism, gender,
sexuality, sexual orientation, religiosity, religion, spirituality, social support, social
12
network as well as combinations of these terms with each other and with other
health-related terms.
African Americans and HIV/AIDS
African Americans have been disproportionately impacted by the HIV/AIDS
crisis in America almost since its inception and this disparity continues to increase.
In 1982, African Americans comprised only 12% of the United States population but
23% of new AIDS diagnoses (Harper, 2001). In 2005, African Americans comprised
13% of the U. S. population, but 49% of Americans diagnosed with HIV/AIDS that
year. This rate of AIDS diagnosis for African American adults and adolescents was
10 times that of whites and almost three times that of Latino/as (CDC, 2007b).
AIDS is the leading cause of death for African American men aged 25 to 44 and
African American women aged 25 to 34, the third leading cause of death for African
American women aged 35 to 44 and the fourth leading cause of death for African
American women aged 45 to 54 (CDC, 2007c).
In addition, African Americans face a number of disparities with regard to
HIV/AIDS treatment and survival outcomes. They are more likely than whites to be
unaware of their HIV-positive status and to receive diagnosis and treatment
information later in the course of their illness (CDC, 2007b; Smith et al, 2003).
They are more likely than whites to suffer from co-morbid conditions such as
diabetes, cardiovascular disease, fat abnormalities and nephropathy that are
considered complications of HIV drug regimens and compromise HIV treatment
outcomes (Smith et al, 2003). They are underrepresented in HIV/AIDS research and
13
clinical trials and are less likely than white patients to receive investigational
treatments when established regimens fail (Smith et al, 2003). Of all Americans
diagnosed with AIDS between 1997 and 2004, the proportion of African Americans
who have survived at nine years post diagnosis, (66%), is smaller than that of Native
Americans (67%), Latino/as (74%), whites (75%) and Asian Americans (81%)
(CDC, 2007b).
Myriad contextual issues are thought to contribute to these disparities by
complicating African Americans’ experiences living and coping with HIV/AIDS
(Smith et al, 2003). Nearly one in four African Americans lives in poverty and
African Americans with HIV/AIDS appear to be even more likely than their non-
infected counterparts to be poor (CDC, 2007b). Poverty has been found to
negatively impact stress levels, likelihood of victimization, psychological
functioning, health status and substance use and abuse (Jipguep, Sanders-Phillips &
Cotton, 2004). Poverty has also been specifically associated with low rates of HIV
knowledge, limited access to high-quality health care and inadequate housing, all
factors that present challenges to HIV-related health and survival (CDC, 2007b).
Research also indicates that substance abuse is intimately linked to HIV/AIDS risk
for African Americans, both directly as well as indirectly, through related behaviors
like high risk sex and the exchange of sex for drugs or money (Welch, 2000; Wright,
2001). High rates of casual and chronic substance abuse, especially of crack
cocaine, present numerous challenges to both HIV prevention and treatment efforts
14
targeting this community (Adimora & Schoenbach, 2002; CDC, 2007b; Welch,
2000).
Finally, as in other communities of color, many cultural stigmas surrounding
both HIV and homosexuality flourish in the African American community (Buseh,
Stevens, McManus, Addison, Morgan & Millon-Underwood, 2006). These attitudes
and beliefs have been reflected in the frequently contradictory responses to the
HIV/AIDS crisis generated by community stakeholders and have hampered
community-based efforts at HIV/AIDS prevention and education (Brooks, Etzel,
Hinojos, Henry & Perez, 2005; Buseh et al, 2006; Jemmott, Jemmott & Hutchinson,
2001). In addition, among African Americans there exists a widespread mistrust of
medical care providers and medical research that dates back to the betrayals of the
Tuskegee Syphilis Study (Jemmott, Jemmott & Hutchinson, 2001; Smith et al,
2003). This mistrust, as well as culturally specific suspicions and mythology about
the etiology of HIV/AIDS and its treatments, have further hampered HIV education,
prevention and treatment efforts introduced through traditional health care systems
(Jemmott, Jemmott & Hutchinson, 2001; Ross, Essien & Torres, 2006).
African Americans and Alcohol Use
Until the mid twentieth century, African Americans were frequently
described as having more permissive attitudes toward and more problems with
alcohol than whites, a phenomenon attributed to the effects of ‘family breakdown’
and psychological dysfunction in the African American community (Caetano, Clark
& Tam, 1998). However, because most alcohol-related research failed to address
15
issues of race and ethnicity until well into the 1980s, for many years these
conclusions could neither be supported nor refuted with empirical evidence
(Caetano, Clark & Tam, 1998; Harper, 2001). In more recent years, the body of
knowledge about alcohol use and misuse among African Americans has grown, but
much of the descriptive information developed has brought to light new
inconsistencies and contradictions that reveal the need for more thorough
investigation (Harper, 2001).
To date, the National Institute on Alcohol Abuse and Alcoholism has
disseminated ten special reports to the U.S. Congress on alcohol use and health.
These reports appear to contradict previous assumptions about alcohol use in the
African American community (Harper, 2001; Obot, 1996). They present African
Americans and whites as having both comparable rates of overall alcohol
consumption and shared patterns of use related to that consumption (Harper, 2001).
For example, NIAAA data indicates that both whites and African Americans are
more likely to drink, to be heavy drinkers and to have problems related to drinking if
they are (1) male, (2) younger, (3) single or newly divorced or (4) live in the western
United States or in large metropolitan areas (Harper, 2001).
Other more recent and culturally specific research findings, however, support
the assertion that alcohol-related beliefs and behaviors vary significantly by ethnicity
(Caetano, 1997; Caetano, Clark & Tam, 1998). Among these are the National
Alcohol Surveys, which first focused explicitly on the alcohol-related experiences of
African Americans and other ethnic minorities in 1984. This study used a national
16
probability sample to examine the prevalence and incidences of alcohol
consumption, (Caetano & Clark, 1998a), alcohol-related problems (Caetano & Clark,
1998b) and situational norms and attitudes toward drinking (Caetano & Clark,
1998c) among African Americans, whites and Hispanics. In addition, the findings
from the 1984 National Alcohol Survey have been expanded and extended through
repeated National Alcohol Surveys that provide longitudinal data on the stability of
and trends related to these findings.
Across these years of studies, differences have emerged between the drinking
attitudes and behaviors of African Americans and other ethnic groups that
complicate the information presented in the NIAAA reports. In both 1984 and 1995,
rates of heavy drinking and dependence-related problems were found to be higher
among African American men than among white men (Caetano & Clark, 1998b) and
the incidence of dependence-related problems and social consequences from
drinking were found to be higher among African American women than among
white women (Caetano, 1997). At the same points in time, however, the rates of
abstention from alcohol use were found to be higher among African American men
than among white men and among African American women than among white
women (Caetano & Clark, 1998a). Thus, these findings appear to indicate that larger
proportions of African Americans, as compared to whites, both abstain from and
experience problems with alcohol.
These surveys also reveal ethnically based differences in the stability of
alcohol-related patterns. Recent years have seen a reduction in the average per
17
capita consumption of alcohol in the United States, but longitudinal results indicate
that this trend has had a differential impact on African Americans in relation to other
ethnic groups (Caetano & Clark, 1998a). Between 1984 and 1995, the incidence of
frequent heavy drinking was found to have decreased among white men (from 20%
to 12%) but remained stable among African Americans (at 15%) (Caetano & Clark,
1998a). During the same time period, the incidence of frequent heavy drinking was
found to have decreased among white women (from 5% to 2%) but remained stable
among African American women (at 5%) (Caetano & Clark, 1998a). Finally, over
the same period of time, rates of abstention from alcohol remained stable among
whites but increased among African Americans. (Caetano & Clark, 1998c).
Researchers suggest that these trends indicate that as whites move from heavier to
lighter drinking without many adopting abstention from alcohol, African Americans
evidence a mixed trend in which light drinkers move toward abstention while heavier
drinkers continue to drink heavily (Caetano & Clark, 1998a).
This greater stability of frequent heavy drinking over time places African
Americans at increased risk for experiencing alcohol-related problems and the
consequences associated with them (Caetano, 1997). Indeed, studies have
consistently found that African Americans suffer more health and psychosocial
problems as a consequence of their alcohol use than members of other ethnic groups,
including financial problems, legal problems, health-related problems, relationship
problems with primary partners, social problems with other friends and family
members and job-related problems (Caetano & Clark, 1998a; Caetano & Clark,
18
1998b; Caetano, Clark & Tam, 1998; Conigliaro, Maisto, McNeil, Kraemer,
Conigliaro & O’Connor, 1998; Harper, 2001; Obot, 1996). Among African
American couples, unlike white and Hispanic couples, alcohol-related problems of
both male and female partners have been found to be strong predictors of intimate
partner violence within couple relationships (Cunradi et al, 1999).
As research has moved away from seeking single factor explanations for
drinking behaviors, more attention has focused on the complex interplay of cultural,
historical, psychological and social realities that impact drinking. Despite this shift,
the interpretation of descriptive information has been hampered by the fact that few
predictors of drinking among African Americans have been firmly established
(Caetano, Clark & Tam, 1998). Research has shown that African Americans, like
other ethnic minorities, are particularly subject to stressors, such as unemployment,
lack of social support, rural-urban migration and racism, that have been linked to
self-destructive behaviors like heavy drinking (Caetano & Clark, 1998a). In
addition, specific stressors related to ethnic minorities’ interactions with dominant
U.S. culture such as acculturative stress, socioeconomic stress and minority stress,
have been found to influence alcohol use patterns (Caetano, Clark & Tam, 1998).
These findings suggest that variations both between and within ethnic groups are
impacted by both environmental characteristics and individual attributes that cannot
be addressed in monolithic terms (Caetano & Clark, 1998b).
In an attempt to integrate these factors, Harper has described a framework of
five underlying factors that characterize the complicated landscape of alcohol use in
19
the African American community (Obot, 1996). The first is an overriding cultural
ambivalence toward alcohol that manifests as the observed polarized extremes in
alcohol-related behavior. The second is the disproportionately wide availability of
alcohol in African American residential communities. The third is the use of alcohol
to cope with widespread frustration related to unemployment and economic
instability. The fourth is the use of alcohol to cope with the distresses associated
with racism and discrimination and to create a common social activity with peers.
The fifth is the tendency for African American women to consume less alcohol than
African American men due to family responsibilities, religious beliefs, dislike of
drinking and less exposure to alcohol-based social situations (Obot, 1996). This
contextual framework reinforces the importance and interplay of both individual and
community attributes in relation to African Americans’ risk for problem drinking and
the health and psychosocial problems associated with it (Harper, 2001; Obot, 1996).
Alcohol Use in the context of HIV/AIDS
Evidence suggests that rates of heavy drinking are disproportionately high
among people living with HIV/AIDS (Cook, Sereika, Hunt, Woodward, Erlen &
Conigliaro, 2001; Galvan, Bing, Fleishman, London, Caetano, Burnam, Longshore,
Morton, Orlando & Shapiro, 2002; Welch, 2000). Data from the HIV Cost and
Services Utilization Study, a national probability survey of HIV-positive adults
receiving medical care in 1996, indicates that the incidence of heavy drinking among
HIV-positive individuals is at least 8%, which approaches twice the 4.5% incidence
seen in the general population, (Galvan et al, 2002). Smaller studies have found
20
incidences of problem drinking among HIV-positive individuals that range from
14% (Cook et al, 2001) to as high as 41% (Isaki & Kresina, 2000). In addition,
research indicates that rates of heavy drinking are specifically higher among HIV-
positive individuals who use cocaine or heroin and who have lower levels of
education and lower incomes (Galvan et al, 2002), all conditions associated with
ethnic minority status (Welch, 2000).
Evidence also suggests that rates of HIV infection are disproportionately high
among those who abuse alcohol (Petry, 1999). Research on those entering alcohol
treatment indicates that even when controlling for injection drug use, MSM sexual
behavior and geographic differences, rates of HIV infection among individuals who
abuse alcohol are up to five times higher than their non alcohol-abusing counterparts
(Petry, 1999). Research also suggests that HIV-positive African Americans are less
able to find help for alcohol-related problems when they ultimately seek treatment
(Welch, 2000).
These statistics are of concern because alcohol abuse and dependence are
especially detrimental to the health and well being of people living with HIV/AIDS
(Conigliaro, Gordon, McGinnis, Rabenack & Justice, 2003; Isaki & Kresina, 2000;
Petry, 1999). A review of the literature on HIV and alcohol use problems found
consistent evidence that the co-occurrence of alcohol use disorders and HIV is
associated with multiple medical and psychiatric complications of HIV disease,
immunosuppressant effects, more rapid progression to AIDS, problems with
medication adherence and less desirable treatment outcomes (Petry, 1999). The
21
author also found consistent associations between alcohol use by HIV-positive
individuals and both intravenous drug use and risky sexual behaviors, which remain
the two most common modes of HIV transmission in the United States (Petry, 1999).
Thus, alcohol abuse and dependence by HIV-positive African Americans raises
concerns about both biomedical and behavioral issues that impact HIV treatment as
well as prevention.
Biomedical Consequences
Many arguments about the detrimental effects of heavy drinking on the health
and well being of people living with HIV/AIDS were historically based on anecdotal
evidence (Isaki & Kresina, 2000). However, more recent research has significantly
strengthened the support for this association. Biomedical evidence suggests that the
mechanism for the impact of alcohol use on immune functioning and on the
progression of HIV disease may lie in the increase of HIV R5 strain infection in the
blood after alcohol exposure (Wang, Douglas, Metzger, Guo, Li, O’Brien, Song,
Davis-Vogal & Ho, 2002). Evidence also suggests that heavy drinking has both
independent and synergistic effects (in tandem with HIV) on the decline of cognitive
functioning in individuals living with HIV, including the hastened compromise of
verbal reasoning, auditory processing and reaction times (Green, Savenau &
Bornstein, 2004).
In addition to direct physiological effects, heavy drinking in the context of
HIV/AIDS yields secondary health effects in the form of frequent comorbidity with
hepatitis, anemia (Conigliaro et al, 2003) and cirrhosis of the liver (Flexner, Cargill,
22
Sinclair, Kresina & Cheever, 2001). Hepatitis contributes to the progression of liver
disease in those already at increased risk for liver toxicity due to the use of HIV
medications (Isaki & Kresina, 2000). Cirrhosis of the liver complicates antiretroviral
treatments due to its association with both impaired liver function and altered drug
metabolism (Flexner et al, 2001). Evidence also suggests that heavy drinking may
result in nutritional deficiencies that leave immune-compromised individuals more
susceptible to infection and to the exacerbation of other medical problems associated
with certain HIV medications (Galvan et al, 2002).
Behavioral Consequences
Heavy alcohol use among people living with HIV/AIDS is also associated
with behavioral issues in HIV/AIDS treatment, including a diminished likelihood of
being prescribed antiretroviral treatment (Flexner et al, 2001) and non-adherence to
HIV medication regimens (Cook et al, 2001; Flexner et al, 2001; Palepu, Horton,
Tibbetts, Meli & Samet, 2003; Samet, Horton, Meli, Freedberg & Palepu, 2004).
Heavy alcohol use is also specifically associated with higher HIV RNA levels and
lower CD4 counts in patients receiving HAART (Samet, Horton, Traphagen, Lyon &
Freedberg, 2003) and has been shown to deleteriously impact both the efficacy and
the toxicity of HIV medications (Fein, Fletcher & Di Scalfani, 1998). It is perhaps
unsurprising then that at progressive levels of severity, heavy alcohol use by HIV-
positive individuals has been independently associated with increased odds of
hospitalization over time (Palepu, Horton, Meli, Tibbetts & Samet, 2005).
23
Heavy alcohol use in the context of HIV/AIDS also poses significant issues
in relation to sexual behavior and thus the prevention of HIV transmission and re-
transmission (Conigliaro et al, 2003; Petry, 1999). Chuang, Liebschutz, Horton &
Samet (2003) reported in a study of 349 HIV-positive individuals with histories of
problematic alcohol use that the majority reported engaging in high-risk sexual
behaviors. One qualitative investigation of the role of alcohol use in the sexual
scripts of HIV-positive MSMs found that among the men interviewed, alcohol use
functioned as a social lubricant, was perceived as a prerequisite to sex, increased
sexual adventurism and assertiveness, altered partner selection and sexual behaviors,
fueled pre-existing sexual compulsivity and enabled some to overcome guilt, fear
and anxiety sufficiently enough to engage in sexual behaviors that they otherwise
considered taboo (Parsons, Vicioso, Punzalan, Halkitis, Kutnick & Velasquez, 2004).
Similarly, alcohol use has been associated with greater motivation to engage in risky
sexual behavior among non-HIV infected heterosexual women considered to be at
high risk for infection (Maisto, Carey, Carey, Gordon & Schum, 2004).
Gender, Ethnicity and Health
It is well documented that many health disparities between men and women
persist, including differences in average lifespan and in the morbidity and mortality
associated with specific mental and physical health problems (Doyal, 2006;
Gesenway, 2001; Read & Gorman, 2005). Similarly well established are the
disparities between various racial and ethnic groups with regard to the incidence of
various diseases, overall health, access to health care and health-related outcomes
24
(Read & Gorman, 2005;Taylor & Braithwaite, 2001). Research exploring the
mechanisms of gender-based differences suggests that gender is associated with the
structural and behavioral determinants of many aspects of individual health for both
men and women (Denton & Walters, 1999). Among African Americans, gender-
based differences in interpersonal relationships, support, health responsibility and
nutrition have all been found to impact health-promoting lifestyles, health status and
health-related behaviors (Johnson, 2005).
Many limitations exist in the literature on health disparities, however. First,
due to the historical reliance on the biomedical model in health research, many
health-related differences between men and women have been examined largely or
only in terms of the biological differences between the sexes (Bird & Rieker, 1999).
More recent research suggests that gender and its associated gender roles, gendered
expectations and social and individual constructions of masculinity and femininity
significantly impact health-related behaviors and outcomes and that these effects
have not been adequately recognized in health care research or practice (Bird &
Rieker, 1999; Doyal, 2006; Gesenway, 2001). Second, research has only recently
begun to examine how race and ethnicity “condition” gendered health disparities,
despite evidence that the nature and magnitude of gender differences vary
significantly by ethnicity (Read & Gorman, 2005). In order to create meaningful
models related to the health of ethnic minority men and women, therefore, it is
imperative to integrate the health-related impacts of the biological phenomena of
sex, the non-biological phenomena of gender, patterns related to culture and ethnicity
25
and the interplay between these factors (Bird & Rieker, 1999, Read & Gorman,
2005).
Gender, Ethnicity and Alcohol Misuse
Gender distinctly influences the processes and consequences of alcohol use,
including the frequency and volume of alcohol consumption, rates of alcohol-related
problems and the consequences of those problems (Green, Freeborn & Polen, 2001;
Wilsnack, Vogeltanz, Wilsnack & Harris, 2000). In general, more men than women
are alcohol dependent and experience alcohol-related problems, but women with
alcohol problems appear to be at greater risk for the adverse consequences of
problem drinking and alcohol-related disease (Walter, Gutierrez, Ramskogler,
Hertling, Dvorak & Lesch, 2003). Evidence also suggests that men and women tend
to consume alcohol under different circumstances, with different people and at
different times of day (Green, Freeborn & Polen, 2001). Women with alcohol
problems are more likely than their male counterparts to be influenced into the
initiation and maintenance of problem drinking by romantic partners and to have
greater responsibility for childcare and family responsibilities (Redgrave, Swartz &
Romanoski, 2003).
The etiology of these differences appears to involve both the direct and
indirect influences of both the physiological and social aspects of gender. For
instance, many of the factors known to independently impact alcohol consumption
patterns such as income, socioeconomic status, employment status and marital status,
are also related to gender (Green, Freeborn & Polen, 2001). Cross-cultural research
26
also suggests that some of the biologically based differences between women’s and
men’s drinking behaviors and experiences are amplified by the expectations
associated with and performance of gender roles (Wilsnack et al, 2000). In addition,
genetic research suggests that the balance of the impact of environmental versus
genetic factors on substance abuse problems varies significantly between men and
women (Jang, Livesley & Vernon, 1997).
Gender also influences many aspects of health that, in turn, influence alcohol
consumption, including self-reported chronic conditions and health status, number
and types of chronic illness and functional status (Green, Freeborn & Polen, 2001).
In addition, gender has been found to impact the ways in which various dimensions
of alcohol use are associated with specific health outcomes and with variance in
individuals’ physical functioning (Green, Perin & Polen, 2004). Gender has also
been found to differentiate the complicated relationships between alcohol
consumption and health-related beliefs and behaviors (Minugh, Rice & Young,
1998).
Research on gender and alcohol use specific to African Americans has been
limited but confirms that the incidence of problem drinking among African
Americans as well as the consequences associated with that drinking vary by gender
(Caetano, Clark & Tam, 1998; Herd, 1997). In addition, research suggests that the
relationship between gender and the effects of heavier drinking on alcohol
dependence symptoms is different among African Americans than among whites,
suggesting an interactive effect of ethnicity and gender on alcohol use that has yet to
27
be adequately explored (Herd, 1997). As such, in order to meaningfully examine the
etiology and impact of alcohol abuse and dependence among HIV-positive African
Americans, it is critical to attend specifically to issues of gender.
Gender, Ethnicity and HIV/AIDS
In the United States, HIV-positive men and women differ by gender in terms
of patterns of HIV transmission (CDC, 2007c), knowledge, attitudes and beliefs
about HIV (Whetten, Reif, Lowe & Eldred, 2003) and both the biomedical and
behavioral aspects of living with HIV/AIDS (Mrus, Williams, Tsevat, Cohn & Wu,
2005; Turner, Laine, Cosler & Hauck, 2003). In the African American community,
the HIV/AIDS epidemic has followed a distinct pattern of transmission that has
increasingly impacted African Americans differently according their gender (CDC,
2007c). In 2005, for example, the rate of AIDS diagnosis for African American men
was eight times that of white men while the rate of AIDS diagnosis for African
American women was almost 24 times that of white women (CDC, 2007b). In the
same year, African American men comprised 41% of all American men living with
HIV/AIDS while African American women comprised 64% of all women living with
HIV/AIDS (CDC, 2007b).
Underlying these discrepancies is an epidemiological trend specific to the
African American community: Among African American women, HIV transmission
has occurred primarily among women who identify as being sexually oriented to
men, through sexual contact with male partners (CDC, 2007c; Smith et al, 2003).
Among African American men, HIV transmission has occurred largely through
28
sexual contact with same sex partners, although to a significant degree among men
who do not identify as being sexually oriented to men and/or whose lives are
characterized by multiple, concurrent but mutually exclusive psychosexual
identifications and behaviors (Malebranche, 2003; Smith et al, 2003; Williams,
Wyatt, Resell, Peterson & Asuan-O’Brien, 2004; Wohl et al, 2002). This pattern is
reflected in national statistics showing the predominant route of HIV transmission
for African American men as sexual contact with other men and the predominant
route of HIV transmission for African American women as high-risk heterosexual
contact (CDC, 2007b).
This phenomenon has been met with a unique set of responses within the
African American community that are grounded in culturally specific values,
attitudes and practices around gender and gender roles (Buseh et al, 2006;
Whitehead, 1997). The construction of African American masculinity is based on
ideals of economic capacity as well as heterosexual attraction and sexual prowess
(Levant, Majors & Kelley, 1998; Whitehead, 1997). Male homosexual behavior,
whether related to desire, survival during incarceration or subsistence in the context
of substance abuse, conflicts with cultural ideals about race, gender role expectations
and definitions of masculinity and is therefore often perceived as a weakness and an
embarrassment (Brooks, Etzel, Hinojos, Henry & Perez, 2005). For many African
Americans, homosexuality is also perceived as immoral on the basis of religious
principles and conflicts with community norms around sexuality (Brooks et al, 2005;
Williams et al, 2004). Research suggests that tolerance for homosexual behavior in
29
these communities is often conditionally based on the tacit agreement to avoid
disclosure or overt display of homosexual behavior (Crawford, Allison, Zamboni &
Soto, 2002; Williams et al, 2004).
These patterns influence HIV/AIDS among African Americans in a number
of ways. First, evidence suggests that many African American men who engage in
sexual relationships with both men and women experience extreme pressure to
conceal their homosexual behavior (Brooks et al, 2005; Crawford, Allison, Zamboni
& Soto, 2002). In addition, due to its historic association with homosexuality,
HIV/AIDS itself has also been heavily stigmatized in the African American
community and many HIV-positive African American men may experience pressure
to conceal their HIV status, regardless of their sexual behavior (Beatty, Gaiter &
Wheeler, 2004; Buseh et al, 2006; Coleman & Hummel, 2005). Both of these
patterns of omission discourage the use of HIV prevention services and hinder
condom use and the disclosure of HIV status to sex partners, enabling misleading but
culturally entrenched messages about sexuality and HIV risk to persist unchallenged
(Malebranche, 2003; Williams et al, 2004; Wohl et al, 2002).
HIV-positive African American men who conceal information about their
sexual behavior and/or health status may do so in order to reinforce their masculinity
and preserve family and social supports (Williams et al, 2004). Unfortunately, these
efforts may also lead to denial and isolation that leave these men further vulnerable
to depression and to delay in addressing HIV-related needs (Coleman & Hummel,
2005). At the same time, those who are open about their sexuality and/or HIV status
30
may perceive or experience alienation from previously supportive family and
community members and religious institutions (Buseh et al, 2006; Williams et al,
2004). Due to the traditional African American model of reliance on frequent
contact with and sustained proximity to close and extended family members for
social support (Wilson, 1986), these rejections may force distinct changes in the
composition of HIV-positive African American men’s social networks that
compromise their connections to community resources and individual social
supports.
Like HIV-positive African American men, HIV-positive African American
women are subject to many forms of stigma related to race, poverty and disease
status that impact their health and well-being (Sandelowski, Lambe & Barroso,
2004). In addition, as HIV/AIDS challenges cultural ideals about gender and gender
roles for African American women, they too face stigmas within their own
community that are gender-specific (Shambley-Ebron & Boyle, 2006). Because
HIV/AIDS among African American women has been linked to crack cocaine use
and associated subsistence sex work (Wohl, Lu, Odem, Sorvillo, Pegues & Kerndt,
1998), these women may be stigmatized by their assumed sexual promiscuity
(McNair & Prather, 2004). In addition, they may labeled as weak and as failures
with regard to motherhood. All of these issues compromise their ability to fulfill the
ideals of African American womanhood and the culturally specific gender roles
associated with them (Shambley-Ebron & Boyle, 2006).
31
In fact, HIV-positive African American women are also more likely than
their male counterparts to be single parents who are unemployed and/or receive
public assistance (Wohl et al, 1998). As a result, they frequently must balance
multiple care giving responsibilities that, in the face of limited resources, lead to
their prioritization of family and children’s demands over addressing their own HIV-
related care needs (Shambley-Ebron & Boyle, 2006). Thus, while HIV-positive
African American women may be more engaged with their families than their male
counterparts, family responsibilities also may be a significant source of stress for
them (Jones, Beach, Forehand & Foster, 2003). Finally, after being diagnosed with
HIV, African American women may be able to rely on religiously based community
resources more readily than their male counterparts, both because of the strength of
their preexisting relationships with religious institutions and the less direct
association of homosexuality with their disease status (Hicks, Allen & Wright,
2005).
Based on the issues highlighted, it seems likely that a number of gendered
issues differentiate the experiences of African American men and women in relation
to both alcohol abuse and dependence and living with HIV/AIDS. In order to
understand the etiology of gender differences in alcohol abuse and dependence
among HIV-positive African Americans, it is critical to determine the specific
mechanisms through which these differences arise and manifest. This need leads the
current analysis to the consideration of the impact of gender-based differences in
32
religiosity, social support and social network composition on alcohol abuse and
dependence among HIV-positive African Americans.
Religiosity and Spirituality
In years past, there was little perceived need to distinguish the concepts of
religiosity and spirituality from one another and, as a consequence, these terms were
often used interchangeably (Miller & Thorensen, 2003; Zinnbauer et al, 1997). In
recent decades, however, disillusionment with organized religion has increased and
secular attitudes have become more prevalent in American society. As more
individuals have sought out personal as opposed to institutional ways to express their
beliefs, the meanings behind the concepts of religiosity and spirituality have shifted
and it has become necessary to differentiate their meanings more clearly (Zinnbauer
et al, 1997).
Miller and Thorensen (2003) suggest that spirituality is usually understood at
the individual level and is concerned with transcendent aspects of life that are not
perceptible to the senses, but that are used to understand the material world. They
suggest that religion, though concerned with spiritual matters, takes the form of a
social entity and is defined by the practical boundaries of that phenomenon, so that
religion is to spirituality as medicine is to health (Miller & Thorensen, 2003).
Zinnbauer et al (1997) also found that the concepts of spirituality and religiosity,
though related and modestly correlated, are distinct constructs. They found that
individuals’ definitions of spirituality most often refer to a belief in a higher power, a
feeling of connection to that higher power and the integration of core values with
33
actions and behavior, while definitions of religiosity include some similar concepts
but emphasize organized activities such as church attendance and participation in
rituals as well as adherence to organizational dogma (Zinnbauer et al, 1997).
Religiosity and HIV/AIDS
A growing body of literature supports the conclusion that religiosity and
spirituality both play significant roles in the lives of people living with HIV/AIDS
(Pargament et al, 2004). Research indicates that most HIV/AIDS patients, especially
those who are members of ethnic minority groups, subscribe to an organized religion
and use participation in that religion to cope with their disease (Cotton et al, 2006a).
Many studies have also suggested that an emphasis on spirituality, described as
having a relationship with God or a higher power, is even more common among
those living with HIV/AIDS than belonging to a specific religious denomination or
attending religious services (Pargament et al, 2004). In general, HIV-positive men
and women evidence the frequent use of religious coping in order to deal with HIV-
associated losses, overcome guilt and shame and renew a sense of purpose in life in
the context of HIV/AIDS (Pargament et al, 2004).
It is also clear that religiosity and spirituality influence the health and well
being of those living with HIV/AIDS. Specific dimensions of spirituality among
HIV-positive individuals have been associated with improved immune status, lower
anxiety and lower perceived stress (Pargament et al, 2004). In addition, longitudinal
research indicates that an increase in spirituality and/or religiosity after HIV
diagnosis is associated with significantly greater maintenance of CD4 cells and
34
significantly better control of HIV viral load over a period of four years (Ironson,
Stuetzle & Fletcher, 2006). An increase in spirituality and/or religiosity among
individuals living with HIV/AIDS has also been directly and indirectly associated
with an improved outlook on living with HIV/AIDS (Szaflarski et al, 2006) and
lower spiritual well being has been associated with significant depressive symptoms
(Yi et al, 2006). Finally, elements of religious coping used by HIV-positive
individuals have been associated with greater optimism, self-esteem and life
satisfaction as well as with less alcohol use (Cotton et al, 2006a).
Religiosity and African Americans
It is well established that religion is a powerful force in the lives of many
African Americans (Levin & Taylor, 1993). Large national surveys have
consistently found that African Americans report higher rates of adult church
attendance, regular church attendance and regular prayer than white Americans
(Galvan, Collins, Kanouse, Pantoja & Golinelli, 2007). Religion and spirituality
have been found to shape individual, family and community relationships among
African Americans (Mattis & Jagers, 2001) and religion and religious institutions
have been instrumental in African American social activism and in the provision of
social, emotional and economic support within the African American community
(Taylor, Mattis & Chatters, 1999).
A great deal of evidence also suggests that African Americans’ levels of
religious involvement and uses of religion are gender-specific. Research indicates
that although African American men are typically actively involved in religious
35
practice, African American women are significantly more involved in and committed
to it (Levin & Taylor, 1993). These differences have been found to exist across
organizational, non-organizational and subjective aspects of religious participation
such as attendance at religious services, frequency of prayer and stated importance of
religious or spiritual beliefs (Chatters, Taylor & Lincoln, 1999).
Religiosity, African Americans, HIV/AIDS and Alcohol
Not surprisingly, HIV-positive African Americans are more likely than HIV-
positive whites to describe themselves as having become increasingly spiritual
(although not necessarily more religious) after their HIV diagnosis and to believe
that their religiosity and spirituality have helped them to live longer (Cotton et al,
2006b). Research indicates that spiritual indicators of purpose and meaning, more
than indicators of religious well being, are significantly related to psychological well
being among HIV-positive African Americans (Coleman & Holzemer, 1999).
However, among heterosexual HIV-positive African Americans specifically, degree
of religious well being has also been correlated with lower rates of depression
(Coleman, 2004). Research also suggests that among HIV-positive African
American substance abusers, religion is an important source of support and recent
church attendance is associated with the receipt of HIV medical care (Latkin, Tobin
& Gilbert, 2002).
Despite these generalities, gender-based differences among African
Americans in relation to spirituality and to a greater degree religiosity appear to
persist in the context of HIV/AIDS. For instance, HIV-positive African American
36
men and women have been found to differ in the their frequency of the use of prayer
to manage specific HIV-related symptoms such as fatigue, nausea and depression
(Coleman et al, 2006). In research that attends to issues of gender, high levels of
spiritually based coping and a positive association between spirituality and the
receipt of HIV-related social support have been found among HIV-positive African
American women (Simoni, Martone & Kerwin, 2002). Research also suggests that
for these women, greater religious involvement reduces psychological distress
(Prado et al, 2004) and prayer is significantly inversely related to feelings of
depression (Morse et al, 2000).
Among HIV-positive African American men, however, research on the
impact of religiosity has been more limited and some research suggests that the role
that religiosity plays in their lives is complex and gender-specific. In ethnographic
research, both gay and non-gay identifying HIV-positive African American MSM
identified church attendance less as a resource for spiritual guidance and more as a
social opportunity to make sexual connections (Williams et al, 2004). This finding
echoes research suggesting that among African American MSMs, attendance at
traditional religious institutions is unrelated to gains in spirituality and is negatively
associated with feelings of self worth (Woodyard et al, 2000).
Consistent with decades of research on religion and alcohol consumption,
religious participation has been found to play a regulatory role that inhibits the
likelihood of alcohol consumption among urban African Americans seeking medical
care (Bazargan, Sherkat & Bazargan, 2004). Frequency of church attendance has
37
also been associated with fewer alcohol problems among young African American
adults (Bowie, Ensminger & Robertson, 2005). In addition, the importance of
religion in an individual’s life has been identified as one of the strongest predictors
of current drinking among those living with HIV/AIDS, regardless of ethnicity
(Caetano & Clark, 1998a). Among HIV-positive African American women, research
has found a negative association between level of spirituality and substance abuse
(Simoni, Martone & Kerwin, 2002) and that both prayer and attendance at church
related activities are significantly inversely related to the use of drugs and alcohol
(Morse et al, 2000). Again, among HIV-positive African American men, research on
the impact of religiosity on alcohol use has been more limited. Given the gender-
based differences related to both religiosity and alcohol abuse and dependence
among HIV-positive African Americans, though, it seems likely that the relationship
between these two issues is also influenced by gender.
Social Support
First conceptually articulated by Cassel (1974) and Caplan (1974), social
support has been described as the social relationships that communicate to an
individual that s/he is cared for, esteemed and exists within an interactive network of
supportive interpersonal exchanges. Cobb (1976) first described the beneficial
effects of social support on health and functioning. Since that time, social support
has been widely found to influence mental and physical health as well as
achievement, performance and functional outcomes in a variety of domains
(Belgrave, 1998b; Hurdle, 2001). Social support has also been conceptualized and
38
measured in a variety of ways. Various typologies of social support measurement
focus on the presence or absence of a supportive interaction (support existence), on
who provides the support in question (social support network structure) and on the
nature or content of the support provided (type or content of support). According to
the latter, social support may be personal, impersonal, formal or informal and may be
characterized by emotional, cognitive or material support (Belgrave, 1998b).
Similarly, social support can be categorized in terms of both its structural and
functional aspects. Functional aspects of social support include what is gained
through the support provided or the needs and/or purposes served, while structural
aspects of social support include the individual’s degree of integration or
embeddedness within the network providing support (Belgrave, 1998b). Finally, the
functional dimension of social support can also be categorized in terms of enacted
social support, or whether or not some action is taken to provide support, as well as
perceived social support, or whether or not the recipient perceives some form of
support to have been provided or to have been useful or valuable (Belgrave, 1998b).
One of the conceptual advantages of focusing on perceived social support is
its congruence with social support theory, which hypothesizes that social support
protects individuals against the negative impact of stressors by causing them to
appraise stressful incidences less negatively (Lakey & Cohen, 2000). Thus, this
theoretical perspective focuses on the individual’s perception of the availability
and/or quality of social support in relation to a given stressor and asserts that because
higher levels of perceived social support cause individuals to believe that help will
39
be available to them when it is needed, it enhances their capacity to cope with stress
(Pierce, Sarason & Sarason, 1995). Indeed, lower expectations about the social
support available from family and others have been associated with poorer
psychological functioning in crisis situations and higher levels of perceived social
support have been associated with more positive self description, which may in turn
promote more effective coping in the face of life stressors (Pierce, Sarason &
Sarason, 1995).
Social Support and HIV/AIDS
A strong body of literature suggests that social support is significantly related
to medical and psychosocial outcomes for people living with many acute and chronic
health conditions, including HIV/AIDS (Hurdle, 2001; Turner-Cobb et al, 2002).
Among HIV-positive individuals, greater satisfaction with social support has been
positively associated with adjustment to living with HIV/AIDS (Turner-Cobb et al,
2002) and higher levels of social support have been associated with more positive
affect and greater self-efficacy to adhere to antiretroviral treatment (Simoni, Frick &
Huang, 2006). Social support has also been found to positively influence actual
medication adherence and this influence has been found to vary by source,
evidencing distinct patterns for social support from friends, family and partners
(Power et al, 2003). Among HIV-positive women, higher levels of social support
from friends, families, partners and organizations, mediated by self esteem and
mastery, have been inversely associated with depression (Simoni, Huang, Goodry &
Montoya, 2005) and lower levels of social support have been associated with greater
40
psychological distress (Hudson, Lee, Miramontes & Portillo, 2001; Mellins, Ehrhart,
Rapkin & Havens, 2000) and more symptoms of anxiety and depression (Catz, Gore-
Felton & McClure, 2002).
Social Support and Alcohol Use
With regard to alcohol abuse and dependence, both general and alcohol-
specific social support have been shown to promote abstinence from alcohol during
and after alcohol treatment (Beattie & Longabaugh, 1999). In accordance with many
of the most common modalities for sobriety maintenance such as Alcoholics
Anonymous and other 12-step programs, social support focused on an individual’s
alcohol has been found to predict abstinence more directly and over a longer period
of time than more general forms of social support (Beattie & Longabaugh, 1999).
Research suggests that the impact of social support on alcohol use may occur
through both the direct effect of social support on drinking behaviors and through
social support’s role in reducing depression, which is in turn related to reduced
alcohol use (Peirce, Frone, Russell, Cooper & Mudar, 2000). Indeed, in the
relationship between social support, alcohol use and chronic illness, lower levels of
social support have been associated with more depressive symptoms, which are in
turn associated with increased alcohol use and alcohol-related problems, the
worsening of health problems and poorer treatment outcomes for both men and
women (Green, Freeborn & Polen, 2001).
41
Social Support, African Americans and HIV/AIDS
It is well established that African Americans utilize social support as an
important informal means to address a variety of problems, ranging from economic
difficulties and housing issues to interpersonal problems with partners and family
members (Lincoln, Chatters & Taylor, 2005). This pattern is especially evident at
the onset of serious personal problems and crises (Taylor, Hardison & Chatters,
1998). Among African Americans, social support has been positively associated
with general well being (Griffin, Amodeo, Clay, Fassler & Ellis, 2006) as well as
with health promoting behaviors (Turner-Musa & Wilson, 2006).
Both the structural and functional aspects of social support among African
Americans follow culturally distinct patterns. African Americans are more likely to
seek and receive social support from informal sources such as family and extended
family, kin networks and their religious and spiritual communities than members of
some other ethnic groups (Snowden, 1998; Taylor, Hardison & Chatters, 1998). A
number of studies confirm that among African Americans, levels of social support
are directly and indirectly linked to both religiosity and spirituality (Mattis & Jagers,
2001; Prado et al, 2004; Turner-Musa & Wilson, 2006). In addition, both the
sources of African Americans’ social support and their use of various types of social
support may vary depending on the nature of the problem for which help is sought.
African Americans have been found to be less likely than whites, for example, to
seek help from friends, family or religious leaders when dealing with mental health-
related problems (Snowden, 1998).
42
Among HIV-positive African Americans, social support has been positively
associated with mental health status (Stewart, Cianfrini & Walker, 2005) and low
levels of social support have been associated with increased risk for a range of
psychiatric disorders and associated problems (Myers & Durvasula, 1999). The
culturally specific ways in which social support for African Americans living with
HIV/AIDS may be domain or issue-specific may be an important issue, however.
Research suggests that one critical element of securing social support for HIV-
positive substance abusers may be the ability to discuss HIV/AIDS itself (Galkin &
Strauss, 2000). Indeed, disclosure of HIV status has been linked to greater frequency
of HIV-specific social support among HIV-positive African American women
(Simoni et al, 2000) and culturally specific responses appear to result in different
associations between social support and mental health outcomes for HIV-positive
African American men than for HIV-positive white men (Gant & Ostrow, 1995).
These differences may reflect the fact that some social and interpersonal
responses to HIV/AIDS within the African American community have impacted the
social support that HIV-positive African Americans perceive and experience. As a
result of these responses, both the sources and patterns of social support among HIV-
positive African Americans vary by gender (Hicks et al, 2005). HIV-positive
African American women appear to be more likely than their male counterparts to
receive social support in the context of family relationships (Owens, 2003).
However, those relationships may function as both sources of social support and
sources of stress with regard to coping with HIV/AIDS, due to issues surrounding
43
family responsibilities and the disclosure of HIV status (Owens, 2003). Greater
religiosity among HIV-positive African American women than among their male
counterparts may also boost the social support available to them (Prado et al, 2004).
Indeed, among HIV-positive African American women, social support has been
linked to reduced psychological distress (Prado et al, 2004), engagement in HIV-
related intervention efforts (Prado et al, 2002), improved medication adherence
(Edwards, 2006) and self care agency and self care practices (Hurst, Montgomery,
Davis, Killion & Baker, 2005).
Research suggests that HIV-positive African American men may be more
alienated from both religious and familial sources of social support after their HIV
diagnosis and may rely more on social support from friends, with wide variance in
the nature of the support provided (Williams et al, 2004). Despite this finding,
research also suggests that when compared to HIV-positive white men, HIV-positive
African American men report comparable levels of social support from friends and
significantly higher levels of social support from immediate family members
(Heckman et al, 2000). Other research suggests, however, that HIV-positive African
American men who have sex with men may have significantly less interaction with
established circles of social support within the (predominantly white) gay male
community that typically provide support with regard to HIV and sexuality issues
(Gant & Ostrow, 1995). Finally, research suggests that HIV-positive African
American men may be less likely than HIV-positive men in other ethnic groups to
disclose their HIV status or to discuss HIV-related concerns with lovers, close
44
friends and family members, thus limiting the HIV-specific nature of the social
support that they are able to secure via their available social networks (Mason,
Simoni, Marks, Johnson & Richardson, 1997).
Social Network Composition
The analysis of social networks is one way in which researchers explore the
contexts and processes through which social support is provided, using theories and
techniques geared toward understand social relationships and how such relationships
influence behavior (Valente, Gallagher & Mouttapa, 2004). Such analysis can focus
on either the structure or the function of social networks and has proven powerful in
understanding health-related outcomes (Hawe, Webster & Shiell, 2006). Structural
social network analysis may focus on the characteristics of social network members
and the frequency of interactions between them, while functional analysis may focus
on the types and degrees of resources and support provided within the network
(Hawe, Webster & Shiell, 2006). The most complex forms of social network
analysis involve relational datasets in which information is collected on the position
of individuals within a given network, the relationships between network members,
the degree of density in the network structure, the strength of the ties that connect
network members and the relationships between network structure and positions and
how resources are accessed within the network (Hawe, Webster & Shiell, 2006).
Social Networks and Alcohol Use
Research indicates that people who abuse alcohol and other substances
frequently belong to social networks populated by friends, family members and
45
others who also misuse substances or who tacitly approve of such behavior (Valente,
Gallagher & Mouttapa, 2004). The properties of social networks appear to be critical
in relation to alcohol abuse and dependence, in that social networks composed
largely of heavy or problem drinkers versus those composed largely of individuals
who encourage and support reduction in alcohol use exhibit predictably opposite
influences on drinking behavior (Bond, Kaskutas & Weisner, 2003). Indeed,
research has found that 90-day abstinence from alcohol use at both one and three
years post treatment completion is predicted by the percentage of social network
members who are heavy drinkers and the percentage of social network members who
encourage moderation (Bond, Kaskutas & Weisner, 2003).
The composition and functioning of social networks have long been used by
recovery programs based on the 12-step model to support and sustain sobriety for
individuals with alcohol problems (Bond, Kaskutas & Weisner, 2003). Indeed,
research has also found that 90 day abstinence from alcohol use at both one and three
years post treatment completion is predicted by involvement with Alcoholics
Anonymous during the preceding year and utilization of AA-based support for
reducing alcohol use (Bond, Kaskutas & Weisner, 2003). Similar research found
that at one year after alcohol treatment, involvement with Alcoholics Anonymous
predicted lower alcohol consumption and fewer alcohol-related problems and that
the mechanisms through which these effects manifested appear to be the 24-hour
availability, role modeling and experientially based advice on staying sober offered
by other AA members (Kaskutas, Bond & Humphries, 2002). Finally, Witbrodt and
46
Kaskutas (2005) found that for alcoholics six and twelve months out of outpatient
treatment, abstinence was predicted by the number of 12-step meetings attended and
the number of 12-step prescribed activities in which participants engaged. They
identified the two AA-related activities that appeared to be key to these findings as
having a sponsor and engaging in service work to help others in recovery.
Social Networks and HIV/AIDS
Research suggests that social isolation may predispose those living with HIV
to delays in securing appropriate medical care and to accelerated disease progression
(Hough, Magnan, Templin & Gadelrab, 2005). The composition of HIV-positive
individuals’ social networks has been identified as especially important in relation to
HIV-specific support. Research suggests that people with HIV/AIDS tend to rely
more heavily on networks of friends than on family members for such support and
that those with more diverse social networks involving more types of social
relationships may experience greater benefits in relation to morbidity and mortality
(Hough et al, 2005). However, this pattern may also represent an area of
vulnerability, especially for HIV-positive African Americans, due to the fact that
many men and women enter into living with HIV/AIDS with social networks
weakened by past behavioral transgressions around substance abuse and sexuality as
well as by HIV stigma (Hough et al, 2005). In addition, as individuals living with
HIV/AIDS age and their illness progresses, their social networks may become
depleted due to their inability to maintain reciprocal relationships, placing them at
increased risk for isolation and poorer health outcomes (Shippy & Karpiak, 2005).
47
Finally, significant gender-based differences typically exist between the
structure and support provided by men’s and women’s social networks (Green,
Freeborn & Polen, 2001). Research suggests that the social networks of HIV-
positive women are more heavily populated by family members, and especially by
children, than those of HIV-positive men (Hough et al, 2005) and that the
composition of HIV-positive women’s social networks influences the degree of
disruption that HIV/AIDS causes in their lives (Ciambrone, 2002). Indeed, the
proportion of family members and children in HIV-positive women’s social
networks has been found to negatively impact their perceptions of the availability,
frequency, helpfulness and desirability of the social support that they receive through
their social networks (Hough et al, 2005).
Social Networks, African Americans and HIV/AIDS
The social networks of African Americans have historically been centered in
family, extended family and kin relationships (Belgrave, 1998b; Taylor, Hardison &
Chatters, 1996) as well as religious affiliations and associations (Mattis & Jagers,
2001). However, social and interpersonal responses to HIV/AIDS within the African
American community have impacted the social networks through which HIV-
positive African American men and women are likely to receive support (Hicks et al,
2005; Hough et al, 2005) and substance abuse may further complicate the functional
aspect of these networks. In one study of African American injection drug users,
those who were HIV-positive had larger social networks that included more females
and more kin, as well as more sources of emotional and instrumental support, than
48
whose who were not HIV-positive. Despite appearing to have mobilized these
resources, however, those living with HIV continued to rely on those social network
members who were ongoing substance abusers (Knowlton, Hua & Latkin, 2004).
In a study of medical care utilization by HIV-positive African American
injection drug users, support from kin networks was not associated with the use of
medical services (Knowlton, Hua & Latkin, 2005). However, having more female
social network members and more sources of emotional, financial and instrumental
support was associated with access to medical care providers; having more female
social network members and more sources of emotional support was associated with
the use of outpatient medical services; and having more active substance abusers in
one’s social network was associated with the use of emergency departments for care
(Knowlton, Hua & Latkin, 2005).
In addition to changes in social network composition associated with
HIV/AIDS and substance abuse, the composition and functioning of HIV-positive
African Americans’ social networks appear to vary significantly by gender (Hough et
al, 2005). These differences are likely to be influential in terms of outcomes related
to alcohol abuse and dependence, an area in which social networks are seen to exert
significant influence (Valente, Gallagher & Mouttapa, 2004). As such, among
African Americans living with HIV/AIDS, gender based differences in social
network composition are theorized to form the mechanism through which both
religiosity and social support differentially influence alcohol abuse and dependence.
49
Conclusion
The evidence is strong that African Americans experience disproportionately
high rates of HIV infection as well as problems with alcohol abuse and dependence,
and multiple other contextual issues that complicate both of these phenomena. This
evidence thus suggests the likelihood of a problematic and potentially dangerous
combination of these issues within this community. In addition, the existence of
gender-based differences in relation to HIV/AIDS, alcohol abuse and dependence,
religiosity, social support and social network composition within the community of
African Americans living with HIV/AIDS calls for analysis that attends to issues of
gender and the mechanisms though which these differences arise. Thus, the current
analysis tests models of the associations between gender, religiosity, social support,
social network composition and alcohol abuse and dependence in an effort to
develop a more clearly conceived, gender and culturally specific model for the
structure and impact of these associations in the lives of HIV-positive African
Americans.
50
CHAPTER 3: Research Methodology
This chapter includes the research questions, the hypotheses, and a
description of the research methodology. The last includes descriptions of the
research design, human subjects protections, sampling procedures, instrumentation,
and procedures for data collection, analysis and interpretation.
Research Hypotheses
The primary purpose of this investigation is to better understand the issues
and processes underlying alcohol abuse and dependence among HIV-positive
African Americans and to explicate the differential ways that these influences
function for men versus women. In addition, this study aims to use consumer
narratives to suggest ways in which these patterns may be related to men’s and
women’s HIV-related self care behaviors and to potential avenues for reducing the
impact of alcohol abuse and dependence on those behaviors. To this end, the first
segment of this analysis used quantitative data from a sample of HIV-positive
African Americans to address the following hypotheses, which are based on the
research questions and illustrated by Figures 1 through 4, which outline each set of
hypothesized variable relationships.
Hypothesis 1: Gender moderates the influence of religiosity on current alcohol
abuse and dependence such that greater religiosity will be associated
with a lesser likelihood of current alcohol abuse or dependence for
women as compared to men.
51
Figure 1. Gender as a Moderator of the Influence of Religiosity on Current Alcohol
Abuse and Dependence
Hypothesis 2: Gender moderates the influence of perceived social support on current
alcohol abuse and dependence such that greater perceived social
support will be associated with a lesser likelihood of current alcohol
abuse or dependence for women as compared to men.
Figure 2. Gender as a Moderator of the Influence of Perceived Social Support on
Current Alcohol Abuse and Dependence
Religiosity
Gender
Religiosity
X
Gender
Current
Alcohol Abuse
or
Dependence
Perceived
Social Support
Gender
Perceived
Social Support
X
Gender
Current
Alcohol Abuse
or
Dependence
52
Hypothesis 3: Social network composition mediates the influence of religiosity and
gender on current alcohol abuse and dependence such that with
greater differences in social network composition, gender will be a
stronger moderator of the influence of religiosity on current alcohol
abuse or dependence.
Figure 3. Social Network Composition as a Mediator of the Influence of Religiosity
and Gender on Current Alcohol Abuse and Dependence
Hypothesis 4: Social network composition mediates the influence of perceived
social support and gender on current alcohol abuse and dependence
such that with greater differences in social network composition,
gender will be a stronger moderator of the influence of perceived
social support on current alcohol abuse or dependence.
Religiosity
Religiosity
X
Gender
Current
Alcohol Abuse
or
Dependence
Social
Network
Composition
Gender
53
Figure 4. Social Network Composition as a Mediator of the Influence of Perceived
Social Support and Gender on Current Alcohol Abuse and Dependence
Qualitative Research Questions
In addition, qualitative analysis of data from a sub-sample of HIV-positive African
Americans with histories of alcohol abuse and/or dependence in the context of
HIV/AIDS was used to explore three additional research questions:
1. From the consumer perspective, what are the factors that influence HIV-
positive African Americans’ alcohol abuse and dependence?
2. From the consumer perspective, what is the impact of alcohol abuse and
dependence on HIV-related self care behaviors?
3. From the consumer perspective, what factors are critical to the cessation
of alcohol abuse and dependence in the context HIV infection?
Research Design
This study consists of the secondary analysis of data collected during a multi-
phase NIGMS and UARP-funded study of alcohol use among HIV-positive ethnic
Perceived
Social Support
Perceived
Social Support
X
Gender
Current
Alcohol
Abuse or
Dependence
Social
Network
Composition
Gender
54
minorities (NIGMS dates: 8-1-03 to 7-31-06, UARP dates: 8-1-06 to 3-31-07). The
specific goals of the parent study were (1) to establish more comprehensive
information about alcohol consumption by HIV-positive people by utilizing a
clinical diagnostic measure of actual alcohol abuse and (2) to establish more
comprehensive information about alcohol consumption by HIV-positive people by
including sufficient numbers of HIV-positive Latinos and African Americans, testing
the hypothesis that HIV-positive African Americans would be more likely to report
alcohol abuse than HIV-positive Latinos.
The parent study employed both the sequential and parallel use of mixed
quantitative and qualitative methods. The first phase of the study involved a series
of nine focus groups conducted to identify issues and domains relevant to the
triangulation of HIV/AIDS, alcohol use and culture within the African American and
Latino communities. Content analysis of these focus group conversations was used
to identify significant themes around alcohol use in the context of HIV in the African
American and Latino communities. The identified themes were ultimately
incorporated into the second phase of the study as quantitative variables. The second
phase of the parent study was cross-sectional and involved the simultaneous
administration of a questionnaire including both standardized measures and semi-
structured clinical inventory questions focused on the quantitative variables and
open-ended interviews conducted with a subset of the larger sample.
55
Human Subjects Protection
The parent study was conducted under the supervision of the Institutional
Review Board of The Charles R. Drew University of Medicine and Science
(approval dates: 8-1-03 to 12-13-07). An application for exemption was submitted to
the University of Southern California Institutional Review Board for review on
December 1, 2006 and approved on January 19, 2007.
Sampling Procedures
Sampling relevant to this analysis occurred during the second phase of the
parent study. For this phase of the study, participants were recruited through flyers,
online notices and presentations at HIV/AIDS clinics, HIV/AIDS service
organizations and other social service agencies in Los Angeles County. Study
personnel trained as clinical social workers screened potential participants who
responded to recruitment materials over the telephone in order to determine their
appropriateness vis a vis the study’s inclusion criteria. In order to be included in the
study, potential participants were required to be over the age of eighteen, African
American, HIV-positive and non-psychotic. The study sample was recruited through
clinics and social service organizations that provide HIV-positive individuals with
services ranging from HIV medical care and psychosocial support to assistance with
transportation, housing and food. As such, the sample consists largely of individuals
with limited financial resources who rely on the assistance of social service providers
to meet basic as well as HIV-related needs.
56
The final purposive sample for the second phase of the parent study consisted
of 283 HIV-positive African Americans who were interviewed between June 2005
and March 2007. Of the 283 participants in the parent study, 157 provided data on
the size, composition and nature of their social networks and 43 participated in open-
ended interviews. Due to the elimination of transgender individuals from the current
analysis, the study sample for this investigation includes 272 HIV-positive African
American men and women, of whom 148 provided social network data and 40
participated in open-ended interviews.
Data Collection
The same clinically trained study personnel who conducted the initial
telephone screenings interviewed participants selected for inclusion in the study.
Interviews were conducted in private study rooms at the research offices of the Drew
Center for AIDS Research, Education and Services, the JWCH Offices at the
Weingart Center and the WE CAN House.
Based on study aims and on the content analysis of focus group data, all
participants in the second phase of the study responded to clinical interview
questions about demographic facts, past and current alcohol abuse, past and current
alcohol dependence and past and current depression. All study participants also
completed standardized instruments used to measure HIV stigma, religiosity,
familialism and multidimensional perceived social support. After a measure
assessing social network composition was added to the study questionnaire,
subsequent participants completed it as well. For these portions of the interviews,
57
study personnel presented interview questions to participants verbally and recorded
their verbal responses on paper questionnaires.
In addition, a purposive sub sample of study participants was recruited from
the larger sample to respond concurrently to open-ended interviews about the causes
and consequences of alcohol abuse and dependence in the context of HIV/AIDS and
on the relationship between alcohol abuse and dependence and HIV self care. These
individuals were selected for participation in open-ended interviews based on
personal histories of alcohol abuse or dependence since their HIV diagnoses, their
assessed ability to provide coherent and reflexive narratives about their experiences
and observations and their willingness to do so in the interview setting. For this
portion of the interviews, study personnel posed interview questions to study
participants verbally and used a tabletop microphone to record their verbal responses
on audiocassette.
Instrumentation
This analysis utilized the quantitative data on current alcohol abuse and
dependence, religiosity, perceived social support and social network composition
collected during the second phase of the parent study.
Alcohol Abuse and Dependence
Current and lifetime alcohol abuse and current and lifetime alcohol
dependence were assessed using the Alcohol Use Screening section of the research
version of the Structured Clinical Interview for DSM IV TR Axis I Disorders,
attached as Appendix A. Because the SCID-I is not a fully structured interview tool
58
and requires that the interviewer (a trained mental health professional) make clinical
judgments during administration, its reliability is a function of the conditions under
which it is administered (First, Gibbon, Spitzer & Williams, 2002). However, tests
of the inter-rater reliability of isolated sections of the SCID-I have yielded kappa
scores ranging from .70 to 1.0 within diagnoses and sites, which is comparable to
other well-established mental health diagnostic instruments such as the NIMH
Diagnostic Interview Schedule (First, Gibbon, Spitzer & Williams, 2002).
Other studies testing the test-retest reliability of the Alcohol Misuse sections
of the SCID-I in isolation (as it is utilized for this study) have established coefficient
alphas of .77 for those sections at seven and ten day test-retest intervals (Zanarini et
al, 2000). Again, because of the nature of the SCID-I interview, its procedural
validity is difficult to establish in the same manner used for fully structured
measures. However, its wide and respected use within psychiatric and psychological
research and therapeutic communities and its use as the “gold standard” in the
assessment of other instruments vouches for its utility in determining diagnostic
distinctions in useful and tenable ways (Shear et al, 2000; Steiner, Tebes, Sledge &
Walker, 1995).
Religiosity and Spirituality
Religiosity and spirituality were measured using the Religiosity and
Spirituality subsection of the HIV Cost and Services Utilization Study survey
instrument, (http://www.rand.org/health/projects/chipts/projects/proj002.html),
which is attached as Appendix B. This instrument was adapted from the Religion
59
and Spirituality section of the MIDI (The Midlife Development Inventory)
developed by the MacArthur Foundation Research Network on Successful Midlife
Development for the National Survey of Midlife Development in the United States
(http://midus.wisc.edu/midus1/). The adapted scale includes 12 items that were used
to measure religiosity and spirituality separately in the HCSUS, using a national
probability sample of 2,864 adults under treatment for HIV infection. Thus, this
measure was specifically adapted to be used with a diverse population of HIV-
positive individuals, including HIV-positive African Americans as comprise the
current study sample (Lorenz et al, 1999, Lorenz et al, 2005).
The opening questions ask the participant to identify his or her personal
religious preference from a list of options that includes 43 denominational
affiliations as well as the categories (agnostic or atheist), (no religious preference)
and (other) as well as whether or not s/he considers him or herself a Christian.
The religiosity section of the measure consists of five questions that ask for
self-reports about degree of religiosity, the importance of religion in the participant’s
life, degree of identification with the participant’s religious group, degree of
preference for being with people of the same religion and frequency of attendance at
services. For the first four questions, possible responses range from 1 (very) to 4
(not at all), with an option of (does not apply) for the question about identification
with religious group. For the fifth question, the possible responses range from 1
(more than once per week) to 5 (never). Factor analysis of these five items yielded a
single factor, with each item having a factor loading of .57 or higher. To create the
60
religiosity scale scores as used, each of the five items was reverse scored so that
higher scores indicated greater levels of religiosity. Cronbach’s alpha for the
religiosity scale was .71.
The spirituality section of the measure consists of four questions that ask for
self-reports about degree of spirituality, the importance of spirituality in the
participant’s life, the degree to which spiritual practices are used to deal with life
stressors and the degree to which spiritual beliefs are used to guide decisions. For
the first two questions, possible responses range from 1 (very) to 4 (not at all). For
the third and fourth questions, possible responses range from 1 (often) to 4 (never).
Factor analysis of these four items yielded a single factor, with each item having a
factor loading of .70 or higher. To create the spirituality scale scores as used, each of
the four items was reverse-scored so that higher scores indicated greater levels of
spirituality. Cronbach’s alpha for the spirituality scale was .68.
Perceived Social Support
Perceived social support was measured using the Multidimensional Scale of
Perceived Social Support (Zimet, Powell, Farley, Werkman & Berkoff, 1990), which
is attached as Appendix C. This scale consists of twelve items, of which four
questions each measure perceived social support from friends, family members and a
significant other, respectively. For each item, participants are asked to describe their
degree of agreement with a positive statement about social support provided by
family, friends or a significant other using a seven point scale that ranges from (very
strongly disagree) to (very strongly agree). The range of possible scores for each
61
subscale is from 4 to 28, with a higher score reflecting a higher level of perceived
social support.
Past psychometric assessment of this scale has yielded evidence of three
distinct subscales and of good construct validity via good internal reliability, based
on Cronbach’s coefficient alpha values ranging from .81 to .98 for the subscales and
total scale and test-retest values ranging from .72 to .85 (Zimet et al, 1990). This
scale has also been found to be valid and reliable when used with a variety of
different populations, including culturally diverse adolescents and clinical as well as
non-clinical populations (Zimet et al, 1990). Factor analysis of the scale’s twelve
items confirmed the existence of three distinct factors, with each item having a factor
loading of .84 or higher with its identified factor and .78 or higher with the full scale.
Cronbach’s alpha was .93 for the perceived social support from family subscale, .92
for the perceived social support from friends subscale and .96 for the perceived
social support from significant other subscale. Cronbach’s alpha for the full scale
was .91.
Social Network Composition
Social network composition was assessed using a series of questions
(attached as Appendix D) that were adapted from the personal network surveying
procedure developed by McCallister and Fischer (1983) and modified based on the
work of Bond, Kaskutas and Weisner

(2003). During this procedure, research
participants were first asked to list all friends, family members, spouses, lovers and
partners whom they see or with whom they talk on the phone at least once within a
62
typical two-week period and consider part of their social circle. The interviewer
recorded up to ten of these “regular contacts” as named and described by the
respondent. The respondent was then asked to provide a variety of information
about each social network member identified, including age, gender and ethnicity,
the nature of their relationship, including its origin, duration and intensity, their
general and immediate proximity to the participant, whether or not that person was a
heavy or problem drinker and whether or not that person encouraged moderation or
abstinence with regard to the participant’s use of alcohol.
Based on a review of the literature on social networks in relation to substance
abuse and recovery (Bond, Kaskutas & Weisner, 2003; Kaskutas, Bond &
Humphries, 2002; Valente, Gallagher & Mouttapa, 2004), three aspects of social
network composition were attended to in the multivariate analysis. These included
the mean number of members in each participant’s social network, the mean number
of social network members who encouraged moderation in or abstinence from
alcohol use and the mean number of social network members who were identified by
the participant as being heavy or problem drinkers.
Alcohol Abuse and Dependence and HIV Self Care
For the purpose of the open-ended interviews the principal research team
developed an interview guide, attached as Appendix E. This guide includes a series
of open-ended questions about participants’ HIV-related self care behaviors, the
barriers they perceive to these self care behaviors, the impact of alcohol use on their
HIV-related self care and their perceptions of what factors are critical to the
63
prevention or minimization the impact of alcohol use on HIV-related self care. Each
interview lasted between 25 and 45 minutes and included the opportunity for the
participant to add any information or observations they considered relevant but that
had not been explored in the questions posed.
Data Analysis and Interpretation
This analysis involved the examination of both the quantitative and
qualitative information described as well as the comparison and integration of the
models provided by these two sources. For the purposes of the quantitative analysis,
responses to all standardized measures were entered into an SPSS database for
statistical analysis. Initially, descriptive statistics, including frequencies, measures of
central tendency and measures of dispersion for demographic and key variables were
calculated to provide complete pictures of the total and sub samples.
Next, bivariate correlations between all key variables were assessed in order
to examine the dataset for multicollinearity. In order to ensure comparability of the
results from the total and sub sample, bivariate comparisons of the sub sample
against those not included in the sub sample were run on all demographic and key
variables. In addition, bivariate analyses were conducted for both samples on
demographic and key variables in order to compare men against women and those
with current alcohol abuse or dependence problems against those without such
problems. This analysis allowed variables not included in the proposed models to be
controlled for in the multivariate analysis. For all bivariate analyses, χ
2
tests using
Pearson’s χ
2
as the test of significance were used to compare groups on nominal
64
level variables, Chi square tests using Kendall’s Tau c as the test of significance
were used to compare groups on low level ordinal level variables and T-tests were
used to compare groups on interval level variables. A significance level of 0.05 was
used for all analyses.
Multivariate analyses were conducted in order to address the research
questions and hypotheses. For these analyses, a series of logistic regression models
were used to test the proposed models of the moderating effects of gender on the
influence of religiosity and perceived social support on alcohol abuse and
dependence, both with and without the mediating effects of social network
composition. Logistic regressions were utilized because of the nature of the outcome
variable under study: When using the SCID-I to assess current alcohol abuse and
dependence, the outcome variable produced is dichotomous (coded 0 for the absence
and 1 for the presence of current alcohol abuse or alcohol dependence). Thus,
although other variables included in the multivariate analysis are measured at the
interval level or can be dichotomized through dummy coding, the outcome variable
of current alcohol abuse or dependence is dichotomous and cannot be meaningfully
formulated at any other level of measurement. Because of this fact, logistic
regressions are the most appropriate means of analysis for this data (Orme &
Buehler, 2001).
In order to test Hypothesis 1, that religiosity is associated with a lesser
likelihood of alcohol abuse and/or dependence for women as compared to men, a
hierarchical logistic regression was conducted in three distinct steps: The main effect
65
of religiosity was entered first, the main effect of gender was entered second and the
interaction term (religiosity by gender) was entered third, with alcohol abuse or
dependence as the outcome variable (Baron & Kenny, 1986; Kenny, 2004). Due to
concerns about addressing both of the interrelated concepts of religiosity and
spirituality, another hierarchical logistic regression was conducted in relation to
Hypothesis 1, attending to spirituality. This regression was also conducted in three
distinct steps: The main effect of spirituality was entered first, the main effect of
gender was entered second and the interaction term (spirituality by gender) was
entered third, with alcohol abuse or dependence as the outcome variable (Baron &
Kenny, 1986; Kenny, 2004).
In order to test Hypothesis 2, that perceived social support is associated with
a lesser likelihood of alcohol abuse and/or dependence for women as compared to
men, another hierarchical logistic regression was conducted in three distinct steps:
The main effect of perceived social support was entered first, the main effect of
gender was entered second, and the interaction term (perceived social support by
gender) was entered third, with alcohol abuse or dependence as the outcome variable
(Baron & Kenny, 1986; Kenny, 2004). In addition, subscale analyses including T-
tests and an additional logistic regression were conducted vis a vis Hypothesis 2, in
order to determine the significance of individual social support subscales with regard
to gender-based differences in social support and the influence of perceived social
support on current alcohol abuse and dependence.
66
In order to test Hypothesis 3 and Hypothesis 4, a four-step process based on
the work of Baron & Kenny (1986), Kenny (2006) and Muller, Judd & Yzerbyt
(2005) was followed to evaluate the proposed mediated moderation relationships.
The first of the four steps is to establish the direct effects of the mediated
moderation, or that there is significant relationship between the predictor, the
moderator, their interaction term and the outcome variable. The second step is to
establish the indirect effects of the mediated moderation, or that a significant
relationship exists between the predictor, the moderator, their interaction term and
the outcome variable.
The third step in this process is to establish that when all variables of interest
are entered into the regression model, there is a significant relationship between the
proposed mediator and the outcome variable. This step is necessary because it is not
sufficient to correlate the mediator only with the outcome variable, as they may be
correlated simply because they are both caused by the initial predictor variables.
Thus, the initial variables must be controlled in establishing the effect of the
mediator on the outcome variable (Kenny, 2006). After the introduction of the
proposed mediator, assessment of the reduction in the p-value of the predictor and
the moderator is used to establish the presence of a mediation effect. Using the
recommendation of Kenny (2006), the mediation effect was considered to be present
only with a reduction of at least 50% in the p-values of the predictor and moderator
Furthermore, only when the stated goals for all of the first three steps have been met
do the data support the stated hypothesis.
67
The fourth and final step in this process is to use the regression conducted in
step 3 to determine whether or not the proposed mediator completely or partially
mediates the influence of the predictor and moderator on the outcome variable. In
this equation, unless the effect of the predictor and the moderator on the outcome
variable is zero when controlling for the mediator, the mediator is considered a
partial mediator (Kenny, 2006).
In order to test Hypothesis 3, that with greater differences in social network
composition, gender will be a stronger moderator of the influence of religiosity on
current alcohol abuse and dependence, the first step of the four step process was
addressed through two hierarchical logistic regressions conducted using the sub
sample of participants about whom social network data was available. In the first,
the main effect of religiosity was entered into the regression first, the main effect of
gender was entered second and the interaction term (religiosity by gender) was
entered third, with alcohol abuse or dependence as the outcome variable (Baron &
Kenny, 1986; Muller, Judd & Yzerbyt, 2005). Again due to concerns about
addressing both of the interrelated concepts of religiosity and spirituality, a second
hierarchical logistic regression was conducted using the spirituality variable. This
regression was also conducted in three distinct steps: The main effect of spirituality
was entered first, the main effect of gender was entered second and the interaction
term (spirituality by gender) was entered third, with alcohol abuse or dependence as
the outcome variable (Baron & Kenny, 1986; Kenny, 2004). If these regressions
failed to establish significant relationships between religiosity, spirituality, gender or
68
their interaction terms and current alcohol abuse or dependence, the latter three steps
in the four-step process cannot be conducted in relation to Hypothesis 3 and the
analyses stopped here (Baron & Kenny, 1986; Muller, Judd & Yzerbyt, 2005).
In order to test Hypothesis 4, that with greater differences in social network
composition, gender will be a stronger moderator of the influence of perceived social
support on current alcohol abuse and dependence, the first step of the four step
process was addressed through a hierarchical logistic regression using the sub sample
of participants about whom social network data was available. This regression was
conducted in three distinct steps: The main effect of perceived social support was
entered first, the main effect of gender was entered second, and the interaction term
(perceived social support by gender) was entered third, with alcohol abuse or
dependence as the outcome variable (Baron & Kenny, 1986; Muller, Judd &
Yzerbyt, 2005). Again, if these regressions failed to establish significant
relationships between perceived social support, gender or their interaction terms and
current alcohol abuse or dependence, the latter three steps in the four-step process
can not be conducted in relation to Hypothesis 4 and the analyses stopped here
(Baron & Kenny, 1986; Muller, Judd & Yzerbyt, 2005).
To further explicate the influence of social network composition, analyses of
the direct effects of three aspects of social network composition on current alcohol
abuse and dependence were conducted. In these regressions, the three aspects of
social network composition were entered in separate blocks; number of social
network members in Block 1, number of social network members encouraging
69
moderation or abstinence in Block 2 and number of social network members who
drink heavily in Block 3.
In addition, in order to test whether or not the targeted elements of social
network composition might have a moderating influence on the impact of perceived
social support on current alcohol abuse or dependence, another series of hierarchical
logistic regressions were conducted addressing three aspects of social network
composition (number of social network members, number of social network
members encouraging moderation in or abstinence from alcohol use, and number of
social network members who drink heavily) as moderators. Each regression was
conducted in three distinct steps: The main effect of perceived social support was
entered first, the main effect of the specified aspect of social network composition
was entered second and the interaction term (social support by social network
composition) was entered third, with alcohol abuse or dependence as the outcome
variable (Baron & Kenny, 1986; Kenny, 2004).
The final segment of the analysis focused on the qualitative information
collected though 40 open-ended interviews. This analysis was used to illustrate and
supplement the quantitative models with information from the narratives of HIV-
positive African Americans about (1) the factors influencing their alcohol abuse
and/or dependence in the context of HIV infection, (2) the relationship between their
alcohol abuse and/or dependence and HIV-related self care behaviors and (3) the
factors critical to their cessation of alcohol abuse and/or dependence in the context
70
HIV infection. This analysis was also used to examine points of congruence and
conflict between the models supported by these two distinct sources of data.
Analysis of the qualitative data was conducted by using the content analysis
features of the QSR Nvivo 7 qualitative analysis software package to examine
transcripts of the 40 open-ended interviews. This analysis employed the analytic
methodology of “coding, consensus, co-occurrence, and comparison” outlined by
Willms et al. (1992) and was rooted in grounded theory, or theory derived from data
and then illustrated by characteristic examples of data (Glaser & Strauss, 1967). As
such, the analysis of the interview transcripts was conducted in several steps.
First, the interview transcripts were imported into Nvivo project file. After
thorough review, the empirical material contained in the transcripts was coded at a
very general level, in order to condense the data into more manageable units for
detailed analysis. Segments of the interview transcripts, ranging from single phrases
to several grouped paragraphs, were then assigned general codes based on their
content. These open codes were based on either a priori themes, typically derived
from the research questions or from questions in the interview guide, as well as
emergent themes that developed during the course of the analysis. In some
instances, the same text segment was assigned multiple or overlapping codes.
The final list of codes consisted of a numbered list of themes, issues,
behavioral accounts and opinions that related to both the process and content of these
phenomena. Some codes focused on the research questions under investigation; (1)
the factors influencing alcohol abuse and/or dependence in the context of HIV
71
infection, (2) the relationship between alcohol abuse and/or dependence and HIV self
care and (3) the factors critical to the cessation of alcohol abuse and/or dependence
in the context HIV infection. Other axial codes related to the variable relationships
explored in the quantitative models and to comparisons with those models in order to
illustrate and supplement those models with the thick description provided by the
qualitative information. Still other codes were used to organize information
according to the researcher’s etic perspective, or outside interpretation of
participants’ statements, language, beliefs and intentions.
Based on these codes, the software program QSR Nvivo 7 (Fraser, 2000) was
used to generate a series of categories arranged in a treelike structure connecting
transcript segments, grouped into separate categories or “nodes.” These nodes and
trees were used to examine the association between different a priori and emergent
categories and to identify the existence of new, previously unrecognized categories.
The number of times that categories co-occurred, either as duplicate codes assigned
to the same text segments or as codes assigned to adjacent text segments in the same
discussion, were recorded. Specific examples of co-occurrence were illustrated with
transcript texts in the analysis. Finally, through the process of constantly comparing
these categories with each other, the different categories were further condensed into
broad themes.
Once the transcripts were coded, the content analysis of data resumed by
comparing and contrasting data within and across codes. A thematic grid that
identifies common and distinct themes, based on the format described by Miller and
72
Crabtree (1992), was produced, using representative passages from the transcripts to
illustrate each theme and to compare the presence and absence of themes within
different groups of participant narratives. Finally, conceptual models were
developed from the analysis of the interview data and were compared to the
statistical models. Both points of congruence and contradictions between the models
from the two data sources were analyzed in order to highlight meaningful
information related to the content and processes addressed in the research questions.
73
CHAPTER 4: Sample Descriptions
Description of Total Sample
The total sample used for the initial portion of these analyses consisted of 272
HIV-positive African American men and women living in or around Los Angeles
County. As presented in Table 1, the majority of the sample was male (73.2%; N =
199) and participants’ mean age was 44.35 years (SD = 7.75). Roughly one third of
these participants (34.6%; N = 94) described themselves as either gay or same
gender loving and an additional 17.3% (N = 47) described themselves as bisexual.
Almost ninety percent of participants (N = 238) described themselves as Christians,
with the most common denominational affiliations being Baptist (N = 98) and Born-
again Christian or Pentecostal (N = 31).
As Table 1 also shows, 44.9% of participants in the total sample (N = 122)
reported completing high school or its equivalent as their highest level of educational
attainment. The sample was heavily skewed toward low-income individuals, as
83.8% (N = 228) reported an annual income of $15,000 or less. Individuals in
unstable living circumstances were also heavily represented, as 33.5% (N = 91) of
the total sample described their living situation as homeless, temporary or
transitional housing. Finally, the average amount of time that total sample
participants had been living with HIV was 10.86 years (SD = 6.55).
74
Table 1. Demographic Characteristics of the Total Sample and Sub Sample
Total Sample Sub Sample
Variable N = 272 % n = 148 %
Gender
Male
Female
199
73
73.2
26.8
109
39
73.6
26.4
Age
20-34
35-44
45-54
55 or older
22
116
108
26
8.1
42.6
39.7
9.6
14
64
58
12
9.5
43.2
39.2
8.1
Sexual Orientation
Heterosexual
Gay/Lesbian
Bisexual
131
94
47
48.2
34.6
17.3
73
50
25
49.3
33.8
16.9
Religious Affiliation
Christian
Non-Christian
No Religious Affiliation
238
12
22
87.4
4.5
8.1
130
8
10
87.8
5.4
6.8
Level of Education
Less than High School
High School or Equivalent
More than High School
64
122
86
23.5
44.9
31.6
40
63
45
27.0
42.6
30.4
Income
$0 - $5,000 per year
$5,001 - $10,000 per year
$10,001 - $15,000 per year
$15,001 or more per year
91
91
46
44
33.5
33.5
16.9
16.2
49
47
26
26
33.1
31.8
17.6
17.7
Living Situation
Homeless
Temporary/Transitional
Living Alone
Living with Family
Living with Roommate/s
14
77
94
72
15
5.1
28.3
34.6
26.5
5.5
9
47
50
36
6
6.1
31.8
33.8
24.3
4.1
Years Since HIV Diagnosis
Less than 5 Years
5 to10 Years
11 to 15 Years
16 Years or More
Not reported
54
66
73
72
7
20.4
24.9
27.5
27.2
2.6
32
36
46
34
0
21.6
24.3
31.1
23.0
0.0
75
As Table 2 demonstrates, the mean score on the religiosity measure for the
total sample was 16.51 (SD = 3.60) out of a possible range of 5 to 22, indicating a
moderately high mean level of self-described religiosity in the total sample. This
trend is reflected in the fact that 44.9% of participants (N = 122) described
themselves as “somewhat religious” while 60.3% (N = 164) described religion as
“very important” in their lives. The mean score on the spirituality measure was
14.26 (SD = 2.00) out of a possible range of 4 to 16, indicating a high mean level of
self-described spirituality in the total sample. This trend is reflected in the fact that
66.9% of participants (N = 182) described themselves as “very spiritual” and 87.1%
(N = 237) described spirituality as “very important” in their lives.
Table 2 also shows that the mean score on the measure of perceived social
support for the total sample was 57.65 (SD = 16.98) out of a possible range of 12 to
84. This mean indicates an average degree of perceived social support for this
sample that is less than one standard deviation above the theoretical midpoint of the
scale. This average is also somewhat lower than the range of means between 66.96
and 72.12 found in four samples used to validate the measure (Zimet, Dahlem, Zimet
& Farley, 1988; Zimet et al, 1990).
As presented in Figure 5, 40.4% (N = 110) of all participants reported
engaging in the heaviest drinking of their lives since receiving their HIV diagnosis.
While 75.4% (N = 205) of total sample participants were determined to have met the
criteria for alcohol abuse or dependence at some time in their lives, 13.6% (N = 37)
met the criteria for current alcohol abuse or alcohol dependence when interviewed.
76
Table 2. Religiosity, Spirituality & Social Support in Total Sample & Sub Sample
Full Sample
(N=272)
Sub Sample
(n=148)
Variable Mean (SD) Mean (SD)
Religiosity 16.51 (3.60) 16.69 (3.52)
Spirituality 14.26 (2.00) 14.25 (2.06)
Perceived
Social Support
57.65 (16.98) 58.53 (16.85)
Figure 5. Alcohol Abuse and Dependence in Total Sample and Sub Sample
40.4%
75.4%
13.6%
37.8%
81.1%
14.9%
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
HIV Alcohol Use Occurred
Since HIV Diagnosis
Lifetime History of Alcohol
Abuse or Dependence
Currently Alcohol Abusing or
Alcohol Dependent
Total Sample (N = 272)
Sub Sample (N = 148)
Description of Sub Sample
The sub sample used to test the mediating influence of social network
composition on the moderating effects of gender on the influence of spirituality,
religiosity and perceived social support on current alcohol abuse and dependence
consisted of 148 participants about whom data on social network composition was
77
available. Analyses indicated that with regard to demographic characteristics and
key variables in the multivariate analysis (gender, spirituality, religiosity, perceived
social support and current alcohol abuse or dependence), the sub sample is
representative of the total sample (all p values > .05). However, bivariate analysis
did reveal that participants in the sub sample differed significantly from participants
not included in the sub sample with regard to two variables related to alcohol abuse
and dependence histories. Participants in the sub sample were more likely to have a
lifetime history of alcohol abuse or dependence (χ
2
= 5.71, df = 1, p = .02) and less
likely to have engaged in the heaviest drinking of their lives since receiving their
HIV diagnoses (χ
2
= 7.62, df = 2, p = .02).
As shown in Table 1, the majority of the sub sample (73.6%; n = 109) was
male and the mean age among sub sample participants was 44.03 years (SD = 7.66).
Roughly one third of these participants (33.8%; n = 50) described themselves as gay
or same gender loving and an additional 16.9% (n = 25) described themselves as
bisexual. Almost ninety percent of sub sample participants (n = 130) described
themselves as Christians, with the most common denominational affiliations being
Baptist (N = 60) and Born-again Christian or Pentecostal (n = 16).
Table 1 also shows that 42.6% of the sub sample participants (n = 63)
reported high school as their highest level of educational attainment. The sub sample
was also heavily skewed toward low-income, unstably housed individuals, with
82.4% (n = 122) reporting an annual income of $15,000 or less and 37.9% (n = 56)
describing their living situation as homeless, temporary or transitional housing. The
78
average amount of time that sub sample participants had been living with HIV was
10.53 years (SD = 6.47).
Table 2 shows that the mean score on the religiosity measure for the sub
sample was 16.69 (SD = 3.52) out of a possible range of 5 to 22, indicating a
moderately high mean level of self-described religiosity in the total sample. This
trend is reflected in the fact that 39.9% of participants (n = 59) described themselves
as “somewhat religious” while 59.5% (n = 88) described religion as “very important”
in their lives. The mean score on the spirituality measure was 14.25 (SD = 2.06) out
of a possible range of 4 to 16, indicating a high mean level of self-described
spirituality in the total sample. This trend is reflected in the fact that 66.9% of
participants (n = 99) described themselves as “very spiritual” and 83.8% (n = 124)
described spirituality as “very important” in their lives.
Table 2 also shows that the mean score on the measure of perceived social
support for the sub sample was 58.53 (SD = 16.85) out of a possible range of 12 to
84. This mean indicates an average degree of perceived social support for this
sample that is less than one standard deviation above the theoretical midpoint of the
scale. This average is also somewhat lower than the range of means from 66.96 to
72.12 found in four samples used to validate the measure (Zimet, Dahlem, Zimet &
Farley, 1988; Zimet et al, 1990).
Figure 5 demonstrates that 37.8% (n = 56) of sub sample participants
reported engaging in the heaviest drinking of their lives since receiving their HIV
diagnosis. While 81.1% (n = 120) of sub sample participants were determined to
79
have met the criteria for alcohol abuse or dependence at some time in their lives,
14.9% (n = 22) met the criteria for current alcohol abuse or alcohol dependence at
the time of being interviewed.
With regard to the aspects of social network composition utilized in these
analyses, Figure 6 demonstrates that participants reported an average social network
size of 4.78 members (SD = 3.03).
Figure 6. Size of Sub Sample Participants’ Social Networks (n = 148)
7
13
19
20 20
18
9
7
11
4
20
0
5
10
15
20
25
0 1 2 3 4 5 6 7 8 9 10
Number of Social Network Members (0 to 10) Reported by Sub Sample Participants
Mean = 4.78 (SD = 3.03)
N
Table 3 shows that these social networks included an average of 3.34 (SD =
2.79) social network members who encouraged moderation with or abstinence from
alcohol and .60 (SD = 1.03) whom the participant described as heavy drinkers. They
also included an average of 4.08 (SD = 2.65) social network members to whom the
80
participant felt especially close, 2.26 (SD = 2.04) family members, 3.91 (SD = 2.91)
individuals who lived in the same city as the participant and 1.64 (SD = 1.99) who
lived within five minutes of the participant’s residence. Of note is that although
16.2% (n = 24) of participants reported having no social network members who
encouraged moderation with or abstinence from alcohol, 64.9% (n = 96) reported
having no social network members who were heavy drinkers. In addition, 6.8% (n =
10) of participants indicated that they did not feel especially close to anyone in their
social network.
Table 3. Composition of Sub Sample Participants’ Social Networks (n = 148)
Sub Sample (n = 148)
Variable Mean (SD) Range
Number of Social Network Members
Who Encourage Moderation/Abstinence
3.34 (2.79) 0 - 10
Number of Social Network Members
Who Are Heavy Drinkers
.60 (1.03) 0 - 5
Number of Social Network Members
To Whom Participant Feels Close
4.08 (2.65) 0 - 10
Number of Social Network Members
Who are Family Members
2.26 (2.04) 0 - 10
Number of Social Network Members
Who Live in Same City As Participant
3.91 (2.91) 0 - 10
Number of Social Network Members
Who Live Within 5 Minutes of
Participant
1.64 (1.99) 0 - 9
In summary, both the total and sub samples appear to be representative of the
population of HIV-positive African American men and women served by social
service and public health agencies in and around Los Angeles County in a variety of
ways (Wohl et al, 1998; Wohl et al, 2002). Both samples include more men than
81
women and the majority of the men describe themselves as gay or bisexual while the
majority of the women describe themselves as heterosexual. The majority of
participants in both samples are between the ages of 35 and 54 and the distribution of
length of time since diagnosis is such that those recently diagnosed as well as those
who have been living with HIV/AIDS for many years are represented. The sample is
predominantly comprised of very low-income individuals, of whom only a small
proportion reported graduating from a four-year college. A higher proportion of
sample participants than would be expected in the general population reported being
currently homeless or residing in temporary or transitional housing. Finally, the
assessed rates of historic and current alcohol abuse and dependence are significantly
higher in this sample than would be expected in the general population (Caetano &
Clark, 1998b). These trends point to the potential significance of alcohol abuse and
dependence and related contextual issues in the lives of the population represented
by these study participants.
82
CHAPTER 5: Quantitative Findings
Bivariate Analyses of the Total Sample
Initially, total sample correlations between key variables in the multivariate
analysis were tested using Pearson’s correlations. Results of this analysis are shown
in Table 4.
Table 4. Pearson’s Correlations between Key Variables in Total Sample (N = 272)
Gender Alcohol
Abuse or
Dependence
Religiosity Spirituality Perceived
Social
Support
Gender
Pearson Correlation
Sig. (2-tailed)
1.0
-
Alcohol Abuse
or Dependence
Pearson Correlation
Sig. (2-tailed)
- .12
.05
1.0
-
Religiosity
Pearson Correlation
Sig. (2-tailed)
.14
.02
- .10
.12
1.0
-
Spirituality
Pearson Correlation
Sig. (2-tailed)
.08
.17
- .04
.50
.35
< .01
1.0
-
Perceived
Social Support
Pearson Correlation
Sig. (2-tailed)
.13
.04
- .22
< .01
.25
< .01
.22
< .01
1.0
-
These findings revealed some significant correlations that support the
research hypotheses: Current alcohol abuse or dependence was significantly
negatively correlated with female gender (p = .05) and with perceived social support
(p < .01); female gender was positively correlated with religiosity (p = .02) and with
perceived social support (p = .04). In addition, however, religiosity and spirituality
were found to be significantly positively correlated (p < .01) and perceived social
83
support was significantly positively correlated with both religiosity (p < .01) and
spirituality (p < .01). Although these findings raise concerns about multicollinearity,
these concerns are mitigated by the fact that none of these three variables will be
entered into multivariate models with one another.
1
Comparisons Based on Gender
Bivariate analyses were conducted in order to determine whether or not
significant differences existed between men and women in the total sample with
regard to demographic variables as well as variables addressed in the multivariate
analyses. As outlined, the total sample was comprised of 73.2% men (N = 199) and
26.8% women (N = 73). As Table 5 demonstrates, results of the bivariate analyses
revealed no significant differences between these men and women with regard to
their type of living situation or level of income. However, significant differences did
emerge between the genders with regard to sexual orientation (χ
2
= 71.2, df = 2, p <
.01), highlighted by the fact that 90.4% (N = 66) of women identified themselves as
exclusively heterosexual whereas 67.3% (N = 134) of men identified themselves as
either gay or bisexual.
Table 5 also shows that bivariate analyses revealed no significant differences
between men and women in the total sample with regard to the proportion of
individuals who reported engaging in their heaviest drinking since being diagnosed
with HIV. However, men were significantly more likely than women to be identified

1
In order to examine this issue, each set of regressions in the multivariate analysis was initially run
while controlling for the other two variables. None of the regression output changed significantly
with the inclusion of these controls, thus regression results are presented without these controls.
84
as having a lifetime history of alcohol abuse or dependence (χ
2
= 4.97, df = 1, p =
.03) and to be identified as currently abusing or dependent on alcohol (χ
2
= 3.87, df =
1, p = .05).
Table 5. Bivariate Demographic and Key Variable Comparisons By Gender
(Nominal and Ordinal Data) (N = 272)
Men
(N = 199)
Women
(N = 73)
Variable N (%) N (%) χ
2
(df) p
Sexual Orientation
Heterosexual
Gay/Lesbian
Bisexual
65 (32.7%)
91 (45.7%)
43 (21.6%)
66 (90.4%)
3 (4.1%)
4 (5.5%)
71.7 (2) <.01
Income*
$0 - $5,000 per year
$5,001 - $10,000 per year
$10,001 - $15,000 per year
$15,001 or more per year
59 (29.6%)
73 (36.7%)
36 (18.1%)
31 (15.6%)
32 (43.8%)
18 (24.7%)
10 (13.7%)
13 (17.8%)
-.08 (.06) .22
Living Situation
Homeless
Temporary/Transitional
Living Alone
Living with Family Members
Living with Roommate/s
12 (6.0%)
59 (29.6%)
69 (34.7%)
47 (23.6%)
12 (6.0%)
2 (2.7%)
18 (24.7%)
25 (34.2%)
25 (34.2%)
3 (4.1%)
4.23 (4) .38
Heaviest Drinking Since HIV Dx
Yes
No
N/A (No History of Alcohol Use)
87 (43.7%)
109 (54.8%)
3 (1.5%)
23 (31.5%)
48 (65.8%)
2 (2.7%)
3.67 (2) .17
Hx of Alcohol Abuse/Dependence
Yes
No
157 (78.9%)
42 (21.1%)
48 (65.8%)
25 (34.2%)
4.97 (1) .03
Currently Abusing/Dependent
Yes
No
32 (16.1%)
167 (83.9%)
5 (6.8%)
68 (93.2%)
3.87 (1) .05
* Significance calculated using Kendall’s Tau C with Standard Error
Finally, as demonstrated in Table 6, results of these bivariate analyses also
revealed no significant differences between men and women with regard to their
mean age, years of education or years since HIV diagnosis. With regard to variables
85
included in the multivariate models, mean levels of spirituality were found to be
comparable among men and women, but women reported significantly higher levels
of both religiosity (t
(270)
= 2.80, p = .01) and perceived social support (t
(270)
= 2.12, p
= .04) than men.
Table 6. Bivariate Demographic and Key Variable Comparisons By Gender
(Interval Data) (N = 272)
Men
(N = 199)
Women
(N = 73)
Variable Mean (SD) Mean (SD) t
(df)
p
Age 44.19 (7.18) 44.81 (9.18) .52
(270)
.60
Years of Education 13.10 (2.36) 13.52 (2.51) -1.44
(270)
.15
Years Since HIV Diagnosis 11.33 (6.53) 9.69 (3.64) -1.82
(263)
.07
Religiosity 16.20 (3.84) 17.36 (2.67) 2.80
(270)
.01
Spirituality 14.16 (2.08) 14.53 (1.73) 1.37
(270)
.17
Perceived Social Support 56.33 (16.1) 61.23 (16.53) 2.12
(270)
.04
Comparisons Based on Alcohol Abuse/Dependence Status
Bivariate analyses were also conducted in order to determine whether or not
significant differences existed between participants with and without current alcohol
abuse or dependence problems in terms of demographic variables and variables
addressed in the multivariate analyses. Only 13.6% of the total sample (N = 37) was
identified as currently abusing or dependent on alcohol.
86
Table 7. Bivariate Demographic and Key Variable Comparisons By
Alcohol Abuse or Dependence Status (Nominal & Ordinal Data) (N = 272)
* Significance calculated using Kendall’s Tau C with Standard Error
Table 7 shows that between participants with and without current alcohol
problems, significant differences emerged with regard to gender (χ
2
= 3.87, df = 1, p
= .05) and sexual orientation (χ
2
= 9.71, df = 2, p = .01). Men comprised 71.1% (N
= 167) of those without current alcohol problems, but 86.5% (N = 32) of those with
such problems. In addition, bisexual individuals comprised 17.3% (N = 47) of the
Alcohol
Abusing /
Dependent
(N = 37)
Non Alcohol
Abusing /
Dependent
(N = 235)
Variable N (%) N (%) χ
2
(df) p
Gender
Male
Female
32 (86.5%)
5 (13.5%)
167 (71.1%)
68 (28.9%)
3.87 (1) .05
Sexual Orientation
Heterosexual
Gay/Lesbian
Bisexual
15 (40.5%)
9 (24.3%)
13 (35.1%)
116 (49.4%)
85 (36.1%)
34 (14.5%)
9.71 (2) .01
Income*
$0 - $5,000 per year
$5,001 - $10,000 per year
$10,001 - $15,000 per year
$15,001 or more per year
14 (37.8%)
11 (29.7%)
6 (16.2%)
6 (16.2%)
77 (32.8%)
80 (34.0%)
40 (17.0%)
38 (16.2%)
-.02 (.05) .71
Living Situation
Homeless
Temporary/Transitional
Living Alone
Living with Family Members
Living with Roommate/s
5 (13.5%)
11 (29.7%)
15 (40.5%)
5 (13.5%)
1 (2.7%)
9 (3.8%)
66 (28.1%)
79 (33.6%)
67 (28.5%)
14 (6.0%)
9.63 (4) .05
Heaviest Drinking Since HIV Dx
Yes
No
N/A (No History of Alcohol Use)
17 (45.9%)
20 (55.1%)
0 (0.0%)
93 (39.6%)
138 (58.7%)
4 (1.7%)
.80 (2) .67
Hx of Alcohol Abuse/Dependence
Yes
No
37 (100.0%)
0 (0.0%)
168 (71.5%)
67 (28.5%)
13.1 (1) <.01
87
total sample, but 35.1% (N = 13) of those with current alcohol abuse or dependence
problems. In addition, participants with current alcohol problems were more likely
than those without such problems to be homeless or live alone but less likely to live
with family members or roommates (χ
2
= 9.63, df = 4, p = .05).
Bivariate analyses revealed no significant differences between these groups
with regard to level of income. In addition, no significant differences were found
between those with and without current alcohol abuse or dependence problems in
terms of the proportion who reported engaging in their heaviest drinking since being
diagnosed with HIV. Those with current alcohol problems, however, were
predictably significantly more likely to have a history of past alcohol abuse or
dependence (χ
2
= 13.1, df = 1, p < .01), despite the fact that fully 71.5% (N = 168) of
those without current problems also had a history of past alcohol abuse or
dependence.
Finally, as shown in Table 8, bivariate analyses revealed no significant
differences between those participants with and without current alcohol abuse or
dependence problems with regard to mean age, years of education or years since
HIV diagnosis. With regard to variables included in the multivariate analyses, no
significant differences were found between those with and without current alcohol
abuse or dependence problems with regard to mean levels of spirituality or
religiosity. However, significant differences were found in mean levels of perceived
social support, which were significantly higher for those without current alcohol
abuse or dependence problems (t
(270)
= -3.77, p < .01).
88
Table 8. Bivariate Demographic and Key Variable Comparisons
By Alcohol Abuse or Dependence Status (Interval Data) (N = 272)
Alcohol
Abusing/Dependent
(N = 37)
Non Alcohol
Abusing/Dependent
(N = 235)
Variable Mean (SD) Mean (SD) t
(df)
p
Age 44.86 (7.58) 44.27 (7.79) .43
(270)
.67
Years of Education 13.78 (2.71) 13.88 (2.36) -.23
(270)
.82
Years Since
HIV Diagnosis
9. 58 (6.98) 11.10 (6.43) -1.30
(263)
.19
Religiosity 15.62 (3.50) 16.65 (3.60) -1.62
(270)
.10
Spirituality 14.05 (2.03) 14. 29 (1.99) -.68
(270)
.50
Perceived
Social Support
48.11 (17.15) 59.15 (16.49) -3.77
(270)
<.01
Based on the results of these bivariate analyses, it was determined that
significant differences existed with regard to sexual orientation between both men
and women and between those with and without current alcohol abuse and
dependence problems. To control for this potential confounder, sexual orientation, a
categorical variable, was controlled for using dummy codes in all multivariate
models related to the moderating influence of gender on the predictors of current
alcohol abuse or dependence.
Multivariate Analyses of the Moderating Influence of Gender
Gender, Religiosity and Alcohol Abuse and Dependence
In order to address Hypothesis I, a series of hierarchical logistic regressions
were used to test whether or not gender significantly moderates the impact of
religiosity on current alcohol abuse and dependence. Religiosity, gender and their
89
interaction term were entered into the model in blocks, controlling for sexual
orientation at each step. Results of these logistic regression analyses, illustrated in
Table 9, show that the overall model was significant (χ
2
= 14.77, df = 5, p = .01) and
the Hosmer and Lemeshow Goodness-of-Fit statistic was not significant (χ
2
= 2.52, df
= 8, p = .96), indicating that the model is a good fit to these data. Religiosity was not
significantly associated with current alcohol abuse or dependence in any block of the
model and gender did not appear to moderate this association.
Table 9. Hierarchical Logistic Regression Model for the Association of Religiosity
and Gender with Current Alcohol Abuse or Dependence (N = 272)
Block Individual Predictor Variables
χ
2
(df)
p-value OR 95% CI p
Block 1
Religiosity
10.39 (3) .02
.93 (.85 - 1.03) .15
Block 2
Gender (Female)
12.18 (3) .01
.36 (.12 – 1.06) .06
Block 3
Religiosity
Gender (Female)
13.82 (4) .01
.94
.38
(.85 – 1.03)
(.13 – 1.12)
.20
.08
Block 4
Religiosity
Gender (Female)
Religiosity X Gender
14.77 (5) .01
.95
7.04
.84
(.86 – 1.05)
(.02 – 2654)
(.58 – 1.21)
.34
.52
.34
(Note: All values are reported controlling for sexual orientation dummy variables:
homosexual (1 = Yes, 0 = No) and bisexual (1 = Yes, 0 = No); heterosexual is reference)
In the first block of the model, with religiosity as the lone predictor variable,
despite model significance (χ
2
= 10.39, df = 3, p = .02) there was no significant
decrease in the odds of alcohol abuse or dependence based on degree on religiosity
90
(p = .15). In Block 2, with gender as the lone predictor variable, there was again
model significance (χ
2
= 12.18, df = 3, p = .01) but only a trend effect for gender (p
= .06) in that females were 64% less likely to be currently alcohol abusing or
dependent than males (OR = .36, 95% CI = .12 – 1.06). When gender was entered
into the model with religiosity in Block 3, the model again remained significant (χ
2
= 13.82, df = 4, p = .01) but gender failed to significantly moderate the still
insignificant association of religiosity with alcohol abuse or dependence (p = .20).
Finally, in Block 4, the model was again significant but the interaction term was not
(p = .34), indicating that the main effects of gender and religiosity are independent.
Greater religiosity, therefore, does not appear to be significantly associated
with decreased odds of current alcohol abuse or dependence after controlling for
gender (p = .20) and gender, also insignificantly associated with current alcohol
abuse or dependence (p = .08), does not appear to significantly moderate this
association at its entry into the model or in the full model. The significance of the
model at each step despite these results appears to stem from the significant
association between sexual orientation and current alcohol abuse or dependence, a
confounder statistically controlled for in these models. As such, it appears that
Hypothesis I must be rejected. However, due to the concerns regarding the
conceptual and operational interrelation of religiosity and spirituality with regard to
their influence on alcohol use as well as their degree of correlation within the dataset,
further analysis was conducted with regard to this hypothesis.
91
Gender, Spirituality and Alcohol Abuse and Dependence
In order to further test Hypothesis 1, a second series of hierarchical logistic
regressions were used to test whether or not gender significantly moderates the
impact of spirituality on current alcohol abuse and dependence. Spirituality, gender
and their interaction term were entered into the model in blocks, controlling for
sexual orientation at each step. Results of these logistic regression analyses are
illustrated in Table 10.
Table 10. Hierarchical Logistic Regression Model for the Association of Spirituality
and Gender with Current Alcohol Abuse or Dependence (N = 272)
Block Individual Predictor Variables
χ
2
(df)
p-value OR 95% CI p
Block 1
Spirituality
9.13 (3) .03
.93 (.78 - 1.09) .36
Block 2
Gender (Female)
12.18 (3) .01
.36 (.12 – 1.06) .06
Block 3
Spirituality
Gender (Female)
12.65 (4) .01
.94
.38
(.80 – 1.12)
(.13 – 1.11)
.49
.08
Block 4
Spirituality
Gender (Female)
Spirituality x Gender
12.29 (5) .01
.91
.00
1.56
(.76 – 1.08)
(.00 – 91.34)
(.71 – 3.47)
.29
.22
.27
(Note: All values are reported controlling for sexual orientation dummy variables:
homosexual (1 = Yes, 0 = No) and bisexual (1 = Yes, 0 = No); heterosexual is reference)
Table 10 shows that the overall model was again significant (χ
2
= 12.29, df =
5, p = .01) and the Hosmer and Lemeshow Goodness-of-Fit statistic was not
significant (χ
2
= 5.10, df = 8, p = .75), indicating that the model is a good fit to these
92
data. However, like religiosity, spirituality was not significantly associated with
current alcohol abuse or dependence in any block of the model and gender did not
appear to moderate this association.
In the first block of the model, with spirituality as the lone predictor variable,
despite model significance (χ
2
= 9.13, df = 3, p = .03) there was no significant
decrease in the odds of alcohol abuse or dependence based on degree of spirituality
(p = .36). In Block 2, with gender as the lone predictor variable, there was again
model significance (χ
2
= 12.18, df = 3, p = .01) but only a trend effect (p = .06) in
that females were 64% less likely to be currently alcohol abusing or dependent than
males (OR = .36, 95% CI = .12 – 1.03). When gender was entered into the model
with spirituality in Block 3, the model again remained significant (χ
2
= 12.65, df = 4,
p = .01) but gender failed to significantly moderate the still insignificant association
of spirituality with alcohol abuse or dependence (p = .49). Finally, in Block 4, the
model was again significant but the interaction term was not (p = .27), indicating that
the main effects of gender and spirituality are independent.
Greater spirituality, therefore, also does not appear to be significantly
associated with decreased odds of current alcohol abuse or dependence after
controlling for gender (p = .49) and gender, also insignificantly associated with
current alcohol abuse or dependence (p = .08), does not appear to significantly
moderate this association either at its entry into the model or in the full model. The
significance of the model at each step despite these results appears to stem from the
93
significant association between sexual orientation and current alcohol abuse or
dependence. As such, Hypothesis 1 must be rejected.
Gender, Social Support and Alcohol Abuse and Dependence
In order to test Hypothesis 2, another series of hierarchical logistic
regressions, shown in Table 11, were used to test whether or not gender significantly
moderates the impact of perceived social support on alcohol abuse and dependence.
Social support, gender and their interaction term were entered into the model in
blocks, controlling for sexual orientation at every step.
Table 11. Hierarchical Logistic Regression Model for the Association of Perceived
Social Support and Gender with Current Alcohol Abuse or Dependence
(N = 272)
Block Individual Predictor Variables
χ
2
(df)
p-value OR 95% CI p-value
Block 1
Social Support
20.65 (3) <.01
.96 (.95 - .98) <.01
Block 2
Gender (Female)
12.18 (3) .01
.36 (.12 – 1.06) .06
Block 3
Social Support
Gender (Female)
23.21 (4) <.01
.97
.43
(.95 – .99)
(.14 – 1.27)
<.01
.13
Block 4
Social Support
Gender (Female)
Soc Sup X Gender
23.72 (5) <.01
.97
.29
1.01
(.94 – .99)
(.01 – 7.22)
(.95 – 1.07)
<.01
.45
.81
(Note: All values are reported controlling for sexual orientation dummy variables:
homosexual (1 = Yes, 0 = No) and bisexual (1 = Yes, 0 = No); heterosexual is reference)
The results of these logistic regression analyses, illustrated in Table 11, show
that the overall model was significant (χ
2
= 23.72, df = 5, p < .01) and the Hosmer
94
and Lemeshow Goodness-of-Fit statistic was not significant (χ
2
= 6.35, df = 8, p =
.61), indicating that the model is a good fit to these data. However, the significance
of the model appears to rely entirely on the influence of perceived social support,
which is significant at (p < .01) alone, and remains significant with the introduction
of gender into the model.
In the first block of the model, both the model and the main effect of social
support alone were significant (p < .01); for each one standard deviation increase in
degree of social support, the odds of alcohol abuse or dependence decreased by 46%.
Once again, with gender as the lone predictor variable in Block 2, there was a trend
effect (p = .06) in that females were 64% less likely to be currently alcohol abusing
or dependent than males (OR = .36, 95% CI = .12 – 1.06). However, when entered
into the model with social support in Block 3, gender did not significantly moderate
the association of social support with alcohol abuse or dependence, resulting in the
odds of alcohol abuse or dependence in relation to social support remaining virtually
unchanged. Finally, in Block 4, the interaction term was not significant (p = .81),
indicating that the main effects of social support and gender are independent. Hence,
it appears that increased perceived social support is significantly associated with
decreased odds of current alcohol abuse and dependence (p < .01) but gender neither
significantly associated with current alcohol abuse or dependence (p = .13) nor
significantly moderates the influence of perceived social support within the model.
As such, Hypothesis 2 must also be rejected.
95
In an effort to explicate the results leading to the rejection of Hypothesis 2,
secondary analyses related to the subscales that comprise the perceived social
support scale were conducted. T-tests were conducted to examine which of the three
subscales within the full scale (perceived social support from significant others,
perceived social support from family and perceived social support from friends)
contributed to differences between men and women in total perceived social support
(t
(270)
= -2.12, p = .04). As illustrated in Figure 7, results of this analysis revealed
that the only significant gender-based differences in the perceived social support
subscale scores lay in perceived social support from family, for which women’s
scores were significantly higher than men’s (t
(270)
= -2.15, p = .03). There were no
significant gender-based differences in scores for perceived social support from
significant others (t
(270)
= -1.43, p = .16) or friends (t
(270)
= -1.36, p = .18).
Figure 7. Sources of Perceived Social Support for Men versus Women (N = 272)
Mean + Standard Error for Perceived Social Support Subscale Scores
16
17
18
19
20
21
22
Significant Other (p = .16) Family (p = .03) Friends (p = .18)
Men (N=199)
Women (N=73)
96
Finally, another series of logistic regressions were conducted in order to
determine which of the perceived social support subscales contributed to the
significant association between total perceived social support and current alcohol
abuse or dependence. Subscale scores for perceived social support from significant
others, family and friends were each regressed on current alcohol abuse and
dependence in separate blocks before all three were entered in a final regression in a
single block. As illustrated in Table 12, this analysis revealed that social support
from significant others (p = .03) and social support from friends (p = .02) are both
significantly associated with current alcohol abuse or dependence in the full model,
while social support from family is not (p = .65).
Table 12. Logistic Regression Model for the Association of Perceived Social
Support from Significant Others, Family and Friends with Current Alcohol
Abuse or Dependence (N = 272)
Block Individual Predictor Variables
χ
2
(df)
p-value OR 95% CI p-value
Block 1
S. O. Support
10.10 (1) <.01
.94 (.90 - .98) <.01
Block 2
Family Support
3.29 (1) .07
.96 (.92 – 1.00) .07
Block 3
Friend Support
11.54 (1) <.01
.91 (.87 – .96) <.01
Block 4
S. O. Support
Family Support
Friend Support
16.42 (3) <.01
.95
1.01
.93
(.91 – 1.00)
(.96 – 1.07)
(.87 – .99)
.03
.65
.02
(Note: All values are reported controlling for sexual orientation dummy variables:
homosexual (1 = Yes, 0 = No) and bisexual (1 = Yes, 0 = No); heterosexual is reference)
97
Because social support from family is the only component of perceived social
support that varies significantly by gender, the results of this logistic regression may
provide some explanation for why gender does not appear to significantly moderate
the association between total perceived social support and current alcohol abuse or
dependence, despite bivariate analysis indicating that there are significant differences
between men and women both in levels of perceived social support (p = .04) and
rates of current alcohol abuse or dependence (χ
2
= 3.87, df = 1, p = .05).
Bivariate Analyses of the Sub Sample
Hypotheses 3 and 4 relate to the mediating influence of social network
composition on the moderating effects of gender. As such, the models used to test
these hypotheses include variables related to social network composition and these
analyses were conducted on data from the sub sample of 148 male and female
participants about whom social network information was collected. Based on the
theory supporting the models being tested, identified differences between those
participants included in the sub sample and those not (with regard to lifetime history
of alcohol abuse or dependence and heaviest drinking since HIV diagnosis) were not
expected to impact the multivariate analysis. However, variable correlations and
bivariate analyses based on gender and on alcohol abuse or dependence status were
conducted on the sub sample in order to evaluate the impact of reduced sample size
on the associations found in the full sample and to examine differences related to
social network composition.
98
Two sets of correlations between key variables in the sub sample were tested
using Pearson’s correlations. Results shown in Table 13 revealed that the power of
the significant correlations found in the total sample was reduced.
Table 13. Pearson’s Correlations between Key Variables in Sub Sample (N = 148)
Gender Alcohol
Abuse or
Dependence
Religiosity Spirituality Perceived
Social
Support
Gender
Pearson Correlation
Sig. (2-tailed)
1.0
-
Alcohol Abuse
or Dependence
Pearson Correlation
Sig. (2-tailed)
- .08
.35
1.0
-
Religiosity
Pearson Correlation
Sig. (2-tailed)
.03
.75
- .10
.21
1.0
-
Spirituality
Pearson Correlation
Sig. (2-tailed)
.06
.46
.01
.96
.45
> .01
1.0
-
Perceived
Social Support
Pearson Correlation
Sig. (2-tailed)
- .01
.95
- .17
.04
.20
.01
.26
> .01
1.0
-
With regard to correlations supporting the research hypotheses, only the negative
correlation between current alcohol abuse or dependence and perceived social
support persisted (p = .04). However, religiosity and spirituality remained highly
positively correlated (p < .01) and perceived social support remained positively
correlated with both religiosity (p = .01) and spirituality (p < .01). Again, although
these findings raise concerns about multicollinearity, these concerns are mitigated by
99
the fact that none of these three variables will be entered into multivariate models
with one another.
2
Table 14 shows that there were no significant correlations between any of the
key social network variables and gender or current alcohol abuse and dependence.
Table 14. Pearson’s Correlations between Social Network Variables
and Other Key Variables in Sub Sample (N = 148)
Relig. Spirit. Perceived
Soc Sup
No. SN
Members
No. SN
Encourage
Moderation
No. SN
Heavy
Drinkers
Religiosity
Pearson Cor.
Sig. (2-tailed)
1.0
-
Spirituality
Pearson Cor.
Sig. (2-tailed)
.45
> .01
1.0
-
Perceived
Soc Support
Pearson Cor.
Sig. (2-tailed)
.20
.01
.26
.01
1.0
-
No. of SN
Members
Pearson Cor.
Sig. (2-tailed)
.13
.11
.23
> .01
.46
> .01
1.0
-
No. SN Who
Encourage
Moderation
Pearson Cor.
Sig. (2-tailed)
.17
.04
.25
> .01
.41
> .01
.78
> .01
1.0
-
No. SN Who
Drink Heavily
Pearson Cor.
Sig. (2-tailed)
.06
.50
.10
.22
.21
.01
.47
> .01
.21
> .01
1.0
-

2
In order to examine this issue, each set of regressions in the multivariate analysis was initially run
while controlling for the other two variables. None of the regression output changed significantly
with the inclusion of these controls, thus regression results are presented without them.
100
However, results presented in Table 14 do show that total number of social
network members was significantly positively correlated with spirituality (p < .01)
and perceived social support (p < .01) as well as with the number of social network
members encouraging moderation in alcohol use (p < .01) and the number of social
network members considered heavy drinkers by the participant (p < .01). The
number of social network members who encourage moderation in alcohol use was
significantly positively correlated with religiosity (p = .04), spirituality (p < .01),
perceived social support (p < .01) and number of social network members considered
heavy drinkers (p = .01). The number of social network members whom participants
considered heavy drinkers was also significantly positively correlated with perceived
social support (p = .01). Because these findings do suggest multicollinearity among
the social network variables, they must be examined within separate models.
Comparisons Based on Gender
Bivariate analyses were conducted in order to determine whether or not
significant differences existed between men and women in the sub sample with
regard to demographic variables as well as variables addressed in the multivariate
analyses. As outlined, the sub sample was comprised of 73.6% men (n = 109) and
26.4% women (n = 39).
101
Table 15. Bivariate Demographic and Key Variable Comparisons By Gender
(Nominal and Ordinal Data) (n = 148)
Men
(n = 109)
Women
(n = 39)
Variable n (%) n (%) χ
2
(df) p
Sexual Orientation
Heterosexual
Gay/Lesbian
Bisexual
40 (36.7%)
48 (44.0%)
21 (19.3%)
33 (84.6%)
2 (5.1%)
4 (10.3%)
27.62 (2) <.01
Income*
$0 - $5,000 per year
$5,001 - $10,000 per year
$10,001 - $15,000 per year
$15,001 or more per year
32 (29.4%)
39 (35.8%)
19 (17.4%)
19 (17.4%)
17 (43.6%)
8 (20.5%)
7 (17.9%)
7 (17.9%)
-.07 (.09) .43
Living Situation
Homeless
Temporary/Transitional
Living Alone
Living with Family Members
Living with Roommate/s
8 (7.3%)
37 (33.9%)
37 (33.9%)
23 (21.1%)
4 (3.7%)
1 (2.6%)
10 (25.6%)
13 (33.3%)
13 (33.3%)
2 (5.1%)
3.62 (4) .46
Heaviest Drinking Since HIV Dx
Yes
No
N/A (No History of Alcohol Use)
65 (59.6%)
44 (40.4%)
0 (0.0%)
27 (69.2%)
12 (30.8%)
0 (0.0%)
1.13 (1) .29
Hx of Alcohol Abuse/Dependence
Yes
No
91 (83.5%)
18 (16.5%)
29 (74.4%)
10 (25.6%)
1.56 (1) .21
Currently Abusing/Dependent
Yes
No
18 (16.5%)
91 (83.5%)
4 (10.3%)
35 (89.7%)
.89 (1) .35
* Significance calculated using Kendall’s Tau C with Standard Error
As shown in Table 15, the results of the bivariate analyses revealed no
significant differences between these men and women with regard to type of living
situation or level of income but, as in the total sample, significant differences did
emerge with regard to sexual orientation (χ
2
= 27.62, df = 2, p < .01), highlighted by
the fact that 84.6% (n = 33) of women identified themselves as exclusively
heterosexual while 63.3% (n = 69) of men identified themselves as either gay or
102
bisexual. No significant differences were found between men and women in the sub
sample with regard to engaging in their heaviest drinking since being diagnosed with
HIV or, unlike in the total sample, with regard to lifetime or current alcohol abuse or
dependence.
Table 16 demonstrates that there were also no significant differences between
men and women in the sub sample with regard to their mean age, years of education
or years since HIV diagnosis. In addition, unlike in the total sample, no significant
differences were found between men and women in the sub sample with regard to
mean levels of spirituality, religiosity or perceived social support.
Table 16. Bivariate Demographic and Key Variable Comparisons By Gender
(Interval Data) (n = 148)
Men
(n = 109)
Women
(n = 39)
Variable Mean (SD) Mean (SD) t
(df)
p
Age 43.89 (7.02) 44.41 (9.31) .32
(54)
.75
Years of Education 13.88 (2.55) 13.47 (2.74) -.84
(146)
.40
Years Since HIV Diagnosis 10.62 (6.49) 10.28 (6.49) -2.82
(146)
.78
Religiosity 16.63 (3.79) 16.85 (2.69) .38
(94)
.71
Spirituality 14.17 (2.17) 14.46 (1.73) .75
(146)
.46
Perceived Social Support 58.58 (16.9) 58.38 (16.92) -.06
(146)
.95
Finally, as shown in Table 17, two significant differences were identified
between men and women in the sub sample with regard to social network
composition. Bivariate analysis revealed that women reported a significantly higher
103
mean number of family members whom they considered part of their social networks
(t
(146)
= -2.22, p = .03) while men reported a significantly higher mean number of
social network members who lived within five minutes of their residences (t
(99.35)
=
2.59, p = .01).
Table 17. Bivariate Social Network Composition Comparisons by Gender (n = 148)
Men
(n = 109)
Women
(n = 39)
Variable Mean (SD) Mean (SD) t
(df)
p
Social Network Members
(Total Number)
4.72 (3.07) 4.95 (2.95) -.40
(146)
.69
Social Network Members
Who Encourage Moderation
3.28 (2.76) 3.51 (2.91) -.44
(146)
.66
Social Network Members
Who Are Heavy Drinkers
.60 (1.02) .62 (1.07) -.10
(146)
.92
Social Network Members
To Whom Feels Close
3.94 (2.65) 4.46 (2.64) -1.05
(146)
.30
Social Network Members
Who Are Family Members
2.04 (2.02) 2.87 (1.99) -2.22
(146)
.03
Social Network Members
Who Live in Same City
3.89 (2.98) 3.97 (2.76) 1.16
(146)
.88
Social Network Members
Who Live Within 5 Minutes
1.94 (2.54) 1.05 (1.43) 2.67
(118)
.01
Comparisons Based on Alcohol Abuse/Dependence Status
Bivariate analyses were also conducted in order to determine whether or not
significant differences existed between sub sample participants with and without
current alcohol abuse or dependence problems in terms of demographic variables and
variables addressed in the multivariate analyses. Only 14.9% of the sub sample (n =
22) was identified as currently abusing or dependent on alcohol.
104
Table 18. Bivariate Demographic and Key Variable Comparisons By
Alcohol Abuse or Dependence Status (Nominal & Ordinal Data) (n = 148)
* Significance calculated using Kendall’s Tau C with Standard Error
Table 18 demonstrates that bivariate analyses revealed no significant
differences between these groups with regard to living situation, level of income or,
unlike the total sample, gender or sexual orientation, although bisexual participants
once again comprised only 16.9% (n = 25) of the sub sample but 27.3% (n = 6) of
Alcohol
Abusing /
Dependent
(n = 22)
Non Alcohol
Abusing /
Dependent
(n = 126)
Variable n (%) n (%) χ
2
(df) p
Gender
Male
Female
18 (81.8%)
4 (18.2%)
91 (72.2%)
35 (27.8%)
.89 (1) .35
Sexual Orientation
Heterosexual
Gay/Lesbian
Bisexual
10 (45.5%)
6 (27.3%)
6 (27.3%)
63 (50.0%)
44 (34.9%)
19 (15.1%)
2.05 (2) .36
Income*
$0 - $5,000 per year
$5,001 - $10,000 per year
$10,001 - $15,000 per year
$15,001 or more per year
8 (36.4%)
7 (31.8%)
3 (13.6%)
4 (18.2%)
41 (32.5%)
40 (31.7%)
23 (18.3%)
22 (17.5%)
-.02 (.07) .75
Living Situation
Homeless
Temporary/Transitional
Living Alone
Living with Family Members
Living with Roommate/s
3 (13.6%)
6 (27.3%)
11 (50.0%)
2 (9.1%)
0 (0.0%)
6 (4.8%)
41 (32.5%)
39 (31.0%)
34 (27.0%)
6 (4.8%)
8.11 (4) .09
Heaviest Drinking Since HIV Dx
Yes
No
N/A (No History of Alcohol Use)
13 (59.1%)
9 (40.9%)
0 (0.0%)
79 (62.7%)
47 (37.3%)
0 (0.0%)
.10 (1) .75
Hx of Alcohol Abuse/Dependence
Yes
No
22 (100.0%)
0 (0.0%)
98 (77.8%)
28 (22.2%)
6.03 (1) .01
105
those with current alcohol abuse or dependence problems. As in the total sample, no
significant differences were found between sub sample participants with and without
current alcohol abuse or dependence problems with regard to the proportion who
reported engaging in their heaviest drinking since being diagnosed with HIV. Those
with current alcohol problems, however, were again significantly more likely to have
a history of past alcohol abuse or dependence (χ
2
= 6.03, df = 1, p = .01), despite the
fact that fully 77.8% (n = 98) of those without current problems also had a history of
past alcohol abuse or dependence.
As shown in Table 19, bivariate analyses also revealed no significant
differences between sub sample participants with and without current alcohol
problems in terms of mean age, years of education or years since HIV diagnosis.
Table 19. Bivariate Demographic and Key Variable Comparisons
By Alcohol Abuse or Dependence Status (Interval Data) (n = 148)
Alcohol
Abusing/Dependent
(n = 22)
Non Alcohol
Abusing/Dependent
(n = 126)
Variable Mean (SD) Mean (SD) t
(df)
p
Age 44.32 (7.82) 43.98 (7.66) .19
(146)
.85
Years of Education 13.45 (2.99) 13.83 (2.53) -.62
(146)
.53
Years Since
HIV Diagnosis
9. 14 (6.79) 10.78 (6.41) -1.10
(146)
.27
Religiosity 15.82 (3.39) 16.84 (3.54) -1.26
(146)
.21
Spirituality 14.27 (1.91) 14. 25 (2.10) .06
(146)
.96
Perceived
Social Support
51.73 (18.58) 59.71 (16.31) -2.08
(146)
.04
106
As in the total sample, with regard to variables included in the multivariate analysis,
no significant differences were found with regard to mean levels of spirituality or
religiosity. However, significant differences were again found in mean levels of
perceived social support, which were higher for those without current alcohol abuse
or dependence problems (t
(146)
= -2.08, p = .04).
Finally, as shown in Table 20, no significant differences were identified
between sub sample participants with and without current alcohol abuse or
dependence problems with regard to any aspect of social network composition.
Table 20. Bivariate Social Network Composition Comparisons
by Alcohol Abuse/Dependence Status (n = 148)
Alcohol
Abusing/Dependent
(n = 22)
Non Alcohol
Abusing/Dependent
(n = 126)
Variable Mean (SD) Mean (SD) t
(df)
p
SN Members
(Total Number)
5.18 (3.47) 4.71 (2.95) .67
(146)
.51
SN Members
Who Encourage
Moderation
3.23 (2.29) 3.37 (2.88) -.21
(146)
.83
SN Members Who
Drink Heavily
.82 (1.14) .56 (1.01) 1.07
(146
)
.29
SN Members To
Whom Feels Close
4.27 (2.91) 4.05 (2.61) .37
(146)
.71
SN Members Who
Are Also Family
2.55 (2.87) 2.21 (1.87) .53
(24)
.60
SN Members Who
Live In Same City
4.18 (3.51) 3.87 (2.81) .47
(146)
.64
SN Members Who
Live Within 5 Min
1.55 (1.92) 1.74 (2.40) -.36
(146)
.72
Based on the results of these bivariate analyses, it was determined that significant
differences existed with regard to sexual orientation between men and women. As
107
such, sexual orientation, a categorical variable, was controlled for using dummy
codes at every step in all multivariate models testing the mediating influence of
social network composition on the moderating effects of gender on predictors of
current alcohol abuse and dependence.
Multivariate Analyses of the Mediating Influence of Social Network
Social Network, Gender, Religiosity and Alcohol Abuse and Dependence
In accordance with the multi-step method outlined by Baron and Kenny
(1986), in order to test Hypothesis 3, another series of hierarchical logistic
regressions were conducted in order to test whether or not gender significantly
moderates the impact of religiosity on current alcohol abuse and dependence among
sub sample participants. Religiosity, gender and their interaction term were entered
into the model in blocks, controlling for sexual orientation at each step.
Results of these logistic regression analyses, illustrated in Table 21, show that
the overall model was not significant (χ
2
= 4.35, df = 5, p = .50). In addition,
religiosity was not significantly associated with current alcohol abuse or dependence
in any block of the model and gender did significantly moderate this association. In
the first block of the model, with religiosity as the lone predictor variable, there was
no significant decrease in the odds of alcohol abuse or dependence based on degree
on religiosity (p = .23). In Block 2, with gender as the lone predictor variable, there
was also no significant effect for gender (p = .32). When gender was entered into the
model with religiosity in Block 3, gender failed to significantly moderate the still
insignificant association of religiosity with alcohol abuse or dependence (p = .24).
108
Finally, in Block 4, the interaction term was not significant (p = .78), indicating that
the main effects of gender and religiosity are independent.
Table 21. Hierarchical Logistic Regression Model for the Association of Religiosity
and Gender with Current Alcohol Abuse or Dependence (n = 148)
Block Individual Predictor Variables
χ
2
(df)
p-value OR 95% CI p
Block 1
Religiosity
3.26 (3) .35
.93 (.82 - 1.05) .23
Block 2
Gender (Female)
2.91 (3) .41
.54 (.16 – 1.84) .32
Block 3
Religiosity
Gender (Female)
4.28 (4) .37
.93
.54
(.82 – 1.05)
(.16 – 1.87)
.24
.33
Block 4
Religiosity
Gender (Female)
Religiosity X Gender
4.35 (5) .50
.95
7.04
.84
(.82 – 1.07)
(.00 – 1118)
(.63 – 1.42)
.31
.93
.78
(Note: All values are reported controlling for sexual orientation dummy variables:
homosexual (1 = Yes, 0 = No) and bisexual (1 = Yes, 0 = No); heterosexual is reference)
In the sub sample, therefore, greater religiosity is not significantly associated with
decreased odds of current alcohol abuse or dependence after controlling for gender
(p = .24) and gender, also insignificantly associated with current alcohol abuse or
dependence (p = .33), does not significantly moderate this association at its entry into
the model or in the full model.
Because no direct effects of religiosity or gender on alcohol abuse and
dependence can be established, there is no moderation present to be mediated by
social network composition and subsequent steps in the method proposed by Baron
109
& Kenny (1986) cannot logically be completed. As such, it appears that hypothesis
3 must also be rejected. However, due to concerns regarding the conceptual and
operational interrelation of religiosity and spirituality in terms of their influence on
alcohol use and the degree of correlation between these two variables in the dataset,
further analysis was conducted with regard to this hypothesis.
Social Network, Gender, Spirituality and Alcohol Abuse and Dependence
In order to further test Hypothesis 3, a second series of hierarchical logistic
regressions were used to test whether or not gender significantly moderates the
impact of spirituality on current alcohol abuse and dependence in the sub sample.
Spirituality, gender and their interaction term were entered into the model in blocks,
controlling for sexual orientation at each step. Results of these logistic regression
analyses, illustrated in Table 22, show that the overall model was not significant (χ
2
= 3.52, df = 5, p = .62). In addition, spirituality was not significantly associated with
current alcohol abuse or dependence in any block of the model and gender did
significantly moderate this association.
In the first block of the model, with spirituality as the lone predictor variable,
there was no significant decrease in the odds of alcohol abuse or dependence based
on degree on spirituality (p = .90). In Block 2, with gender as the lone predictor
variable, there was also no significant effect for gender (p = .32). When gender was
entered into the model with religiosity in Block 3, gender failed to significantly
moderate the still insignificant association of spirituality with alcohol abuse or
110
dependence (p = .96). Finally, in Block 4, the interaction term was not significant (p
= .48), indicating that the main effects of gender and spirituality are independent.
Table 22. Hierarchical Logistic Regression Model for the Association of Spirituality
and Gender with Current Alcohol Abuse or Dependence (n = 148)
Block Individual Predictor Variables
χ
2
(df)
p-value OR 95% CI p
Block 1
Spirituality
1.88 (3) .60
.99 (.79 - 1.24) .90
Block 2
Gender (Female)
2.91 (3) .41
.54 (.16 – 1.84) .32
Block 3
Spirituality
Gender (Female)
2.91 (4) .57
1.00
.54
(.79 – 1.25)
(.16 – 1.85)
.96
.33
Block 4
Spirituality
Gender (Female)
Spirituality x Gender
3.52 (5) .62
.96
.01
1.36
(.76 – 1.22)
(.00 – 1692)
(.59 – 3.13)
.75
.43
.48
(Note: All values are reported controlling for sexual orientation dummy variables:
homosexual (1 = Yes, 0 = No) and bisexual (1 = Yes, 0 = No); heterosexual is reference)
In the sub sample, therefore, greater spirituality is not significantly associated with
decreased odds of current alcohol abuse or dependence after controlling for gender
(p = .96) and gender, also insignificantly associated with current alcohol abuse or
dependence (p = .33), does not significantly moderate this association at its entry into
the model or in the full model. Because no direct effects of religiosity, spirituality or
gender on alcohol abuse and dependence can be established using the sub sample,
there is no moderation present to be mediated by social network composition and
subsequent steps in the method proposed by Baron & Kenny (1986) cannot logically
be completed. As such, hypothesis 3 must be rejected.
111
Social Network, Gender, Social Support and Alcohol Abuse and Dependence
In order to test Hypothesis 4, another series of hierarchical logistic
regressions were used to test whether or not gender significantly moderates the
impact of perceived social support on alcohol abuse and dependence in the sub
sample. Social support, gender and their interaction term were entered into the
model in blocks, controlling for sexual orientation at every step. As the results of
these logistic regression analyses illustrated in Table 23 show, the overall model was
not significant (χ
2
= 7.39, df = 5, p = .19). Although perceived social support is
significantly associated with current alcohol abuse and dependence alone, after the
introduction of gender into the model and after the introduction of the interaction
term for social support and gender into the model, gender does not appear to
moderate this association.
In the first block of the model, with perceived social support as the lone
predictor variable, the main effect of social support alone was significant (p = .04);
for each one standard deviation increase in degree of social support, the odds of
alcohol abuse or dependence decreased by 37%. Once again, in Block 2, with
gender as the lone predictor variable, there was no significant effect for gender (p =
.32). When entered into the model with perceived social support in Block 3, gender
did not significantly moderate the association of social support with alcohol abuse or
dependence, resulting in the odds of alcohol abuse or dependence in relation to social
support remaining unchanged. Finally, in Block 4, the interaction term was not
112
significant (p = .53), indicating that the main effects of perceived social support and
gender are independent.
Table 23. Hierarchical Logistic Regression Model for the Association of Perceived
Social Support and Gender with Current Alcohol Abuse or Dependence
(n = 148)
Block Individual Predictor Variables
χ
2
(df)
p-value OR 95% CI p-value
Block 1
Social Support
5.92 (3) .12
.97 (.95 - .99) .04
Block 2
Gender (Female)
2.91 (3) .41
.54 (.16 – 1.84) .32
Block 3
Social Support
Gender (Female)
6.98 (4) .14
.97
.53
(.95 – .99)
(.15 – 1.86)
.04
.32
Block 4
Social Support
Gender (Female)
Soc Sup X Gender
7.39 (5) .19
.97
.17
1.02
(.94 – .99)
(.00 – 8.04)
(.96 – 1.09)
.04
.37
.53
(Note: All values are reported controlling for sexual orientation dummy variables:
homosexual (1 = Yes, 0 = No) and bisexual (1 = Yes, 0 = No); heterosexual is reference)
Hence, it appears that increased perceived social support is significantly
associated with decreased odds of current alcohol abuse and dependence (p = .04)
but that gender is neither significantly associated with current alcohol abuse or
dependence (p = .32) nor significantly moderates the influence of perceived social
support within the model. Despite the significant direct effects of social support on
alcohol abuse and dependence, because there is no moderating effect of gender
present to be mediated by social network composition, subsequent steps in the
method proposed by Baron & Kenny (1986) cannot logically be completed. As such,
113
hypothesis 4 must also be rejected. In addition, due to the lack of significant gender-
based differences in total perceived social support and current alcohol abuse or
dependence in the sub sample, additional analysis of the perceived social support
subscales would not be meaningful.
However, in an effort to test whether or not the targeted elements of social
network composition might significantly directly influence current alcohol abuse and
dependence in the sub sample, another series of logistic regressions was conducted.
In these regressions, three aspects of social network composition were entered into
the separate model blocks; number of social network members in Block 1, number of
social network members encouraging moderation or abstinence in Block 2 and
number of social network members who drink heavily in Block 3.
Table 24. Hierarchical Logistic Regression Model for the Association of Elements
of Social Network Composition with Current Alcohol Abuse/Dependence
(n = 148)
Block Individual Predictor Variables
χ
2
(df)
p-value OR 95% CI p-value
Block 1
Number of Social
Network Members
.44 (1) .51
1.05 (.91 - 1.22) .50
Block 2
No. of Social Network
Members Who
Encourage Moderation
.05 (1) .83
.98 (.83 – 1.16) .83
Block 3
No. of Social Network
Members Who Are
Heavy Drinkers
1.05 (1) .31
1.24 (.84 – 1.83) .29
114
The results of these logistic regression analyses, illustrated in Table 24, show
that none of the models tested are significant. In the first block of the model, with
number of social network members as the lone predictor variable, there was no
significant effect on the odds of current alcohol abuse or dependence (p = .50). In
Block 2, with number of social network members who encourage moderation as the
lone predictor variable, there was also no significant effect (p = 83). Finally, in
Block 3, with number of social network members who drink heavily as the lone
predictor, there was again no significant effect (p = 29). Hence, none of these three
aspects of social network composition appear to be independently associated with
current alcohol abuse and dependence.
Finally, in order to test whether or not the targeted elements of social network
composition might have a moderating influence on the impact of perceived social
support on current alcohol abuse or dependence in the sub sample, another series of
hierarchical logistic regressions was conducted. In the first model of social network
as a moderator of social support, social support, number of social network members
and their interaction term were entered into the model in blocks. The results of these
logistic regression analyses, illustrated in Table 25, show that the overall model was
not significant.
In the first block of the model, however, both the model (χ
2
= 3.99, df = 1, p
= .05) and the main effect of social support (p = .04) were significant; for each one
standard deviation increase in degree of social support, the odds of alcohol abuse or
dependence decreased by 35%. In Block 2, when number of social network
115
members was entered, both the model (χ
2
= 7.32, df = 2, p = .03) and social support
(p = .01) remained significant, but number of social network members did not
significantly moderate the association of social support with alcohol abuse or
dependence, resulting in that odds ratio remaining virtually unchanged. After
controlling for number of social network members, the odds of alcohol abuse or
dependence decreased by 49% for each standard deviation increase in degree of
social support. Finally, in Block 3, the interaction term was not significant (p = .88),
indicating that the main effects of social support and number of social network
members are independent.
Table 25. Hierarchical Logistic Regression Model for the Association of
Perceived Social Support and Number of Social Network Members
with Current Alcohol Abuse or Dependence (n = 148)
Block Individual Predictor Variables
χ
2
(df)
p-value OR 95% CI p-value
Block 1
Social Support
3.99 (1) .05
.97 (.95 - .99) .04
Block 2
Social Support
No. in Social Network
7.32 (2) .03
.96
1.17
(.93 – .99)
(.99 – 1.39)
.01
.07
Block 3
Social Support
No. in Social Network
Soc Sup X No. in SN
7.35 (3) .06
.96
1.13
1.00
(.91 – 1.00)
(.70 – 1.83)
(.99 – 1.01)
.07
.62
.88
Hence, it appears that although social support is significantly associated with current
alcohol abuse or dependence, number of social network members is neither
116
significantly associated with current alcohol abuse or dependence (p = .07) nor
significantly moderates the influence of perceived social support within the model.
In the second model of social network as a moderator of social support, social
support, number of social network members who encourage moderation and their
interaction term were entered into the model in blocks. The results of these logistic
regression analyses, illustrated in Table 26, show that the overall model was not
significant.
Table 26. Hierarchical Logistic Regression Model for the Association of Perceived
Social Support and Number of Social Network Members who Encourage
Moderation with Current Alcohol Abuse or Dependence (n = 148)
Block Individual Predictor Variables
χ
2
(df)
p-value OR 95% CI p-value
Block 1
Social Support
3.99 (1) .05
.97 (.95 - .99) .04
Block 2
Social Support
No. in Soc Network
Encourage Moderation
4.51 (2) .12
.97
1.07
(.94 - .99)
(.89 – 1.29)
.04
.47
Block 3
Social Support
No. in Soc Network
Encourage Moderation
Soc Sup X No. Who
Encourage Moderation
5.29 (3) .15
.98
1.52
.99
(.94 – 1.02)
(.68 – 3.43)
(.98 – 1.01)
.36
.31
.39
In the first block of the model, however, both the model (χ
2
= 3.99, df = 1, p
= .05) and the main effect of social support (p = .04) were again significant; for each
one standard deviation increase in degree of social support, the odds of alcohol abuse
117
or dependence decreased by 35%. In Block 2, when number of social network
members who encourage moderation was entered, the model became insignificant
but social support remained significant, (p = .04). However, number of social
network members who encourage moderation did not significantly moderate the
association of social support with alcohol abuse or dependence, resulting in that odds
ratio remaining unchanged.
After controlling for number of social network members who encourage
moderation, the odds of alcohol abuse or dependence decreased by 41% for each
standard deviation increase in degree of social support. Finally, in Block 3, the
interaction term was not significant (p = .39), indicating that the main effects of
social support and number of social network members are independent. Hence, it
appears that the number of social network members who encourage moderation is
neither significantly associated with current alcohol abuse or dependence (p = .47)
nor significantly moderates the influence of perceived social support within the
model.
In the third model of social network as a moderator of social support, social
support, number of social network members who are heavy drinkers and their
interaction term were entered into the model in blocks. The results of these logistic
regression analyses, illustrated in Table 27, show that the overall model was not
significant.
118
Table 27. Hierarchical Logistic Regression Model for the Association of Perceived
Social Support and Number of Social Network Members Who Are Heavy
Drinkers with Current Alcohol Abuse or Dependence (n = 148)
Block Individual Predictor Variables
χ
2
(df)
p-value OR 95% CI p-value
Block 1
Social Support
3.99 (1) .05
.97 (.95 - .99) .04
Block 2
Social Support
No. in Social Network
Who Drink Heavily
6.47 (2) .04
.97
1.42
(.94 - .99)
(.94 – 2.14)
.02
.10
Block 3
Social Support
No. in Social Network
Who Drink Heavily
Soc Sup X No. in SN
Who Drink Heavily
6.72 (3) .08
.97
.81
1.01
(.94 - .99)
(.09 – 7.68)
(.98 – 1.04)
.02
.85
.62
In the first block of the model, however, both the model (χ
2
= 3.99, df = 1, p
= .05) and the main effect of social support (p = .04) were again significant; for each
one standard deviation increase in degree of social support, the odds of alcohol abuse
or dependence decreased by 35%. In Block 2, when number of social network
members who drink heavily was entered, both the model (χ
2
= 6.47, df = 2, p = .04)
and social support (p = .02) remained significant, but number of social network
members did not significantly moderate the association of social support with
alcohol abuse or dependence, resulting in the odds ratio remaining unchanged. After
controlling for number of social network members who drink heavily, the odds of
alcohol abuse or dependence decreased by 42% for each standard deviation increase
119
in degree of social support. Finally, in Block 3, the interaction term was not
significant (p = .62), indicating that the main effects of social support and number of
social network members are independent. Hence, it appears that the number of
social network members who drink heavily is neither significantly associated with
current alcohol abuse or dependence (p = .10) nor significantly moderates the
influence of perceived social support within the model.
Post-Hoc Power Analysis
In order to explore whether the lack of significant findings was due to low
power, post-hoc power calculations were performed. The variance accounted for in
the various models ranged from 5% to 8%, which is very low, indicating that a large
sample would be required to significantly detect existing differences. When the total
sample is used (n = 272), the power for the models is greater than 80%. Thus, non-
significant models using the total sample can be considered indicative of non-
significant relationships in this sample. However, for the sub sample, the power is
insufficient for the models, ranging from 54% to 64%. Power to detect specific
effects is even more limited and thus it is possible that some of the non-significance
found in models using the sub sample may be due to insufficient sample size.
120
CHAPTER 6: Qualitative Findings
Participants in Open-ended Interviews
Open-ended interviews were conducted with forty participants who were
assessed as having had alcohol abuse or dependence problems since being diagnosed
with HIV. The mean age of interviewees was 44.65 years (SD = 7.14) and the
average amount of time they had been living with HIV was 10.54 years (SD = 6.83).
72.5% percent (n = 29) of the individuals interviewed were in recovery and had been
sober for an average of 5.38 years (SD = 4.48). 27.5 % (n = 11) of the interviewees
were active drinkers when interviewed. Table 28 describes the 32.5% (n = 13) of
interviewees who were female, of whom twelve were heterosexual and one was
lesbian. Table 29 describes the 67.5% (n = 27) of interviewees who were male, of
whom nineteen were gay, four were bisexual and four were heterosexual.
Table 28. Descriptive Information about Female Participants in Open-Ended
Interviews (n = 13)
ID Sexuality Age Yrs. HIV+ Alcohol Use Status
1 Heterosexual 50 7 Sober for 5 years.
14 Heterosexual 38 4 Sober for 3 years.
15 Heterosexual 55 1 Sober for 6 months.
16 Heterosexual 43 1 Struggling with sobriety since HIV diagnosis.
17 Lesbian 52 10 Sober on & off for 7 years, most recently for 2 yrs.
2years
20 Heterosexual 42 13 Sober on & off since diagnosis, recent drinking.
24 Heterosexual 41 8 Struggling with sobriety, sober for 1 week.
29 Heterosexual 49 17 Sober for 1 year.
30 Heterosexual 46 17 Sober for 1 year.
36 Heterosexual 42 14 Sober for 11 months.
38 Heterosexual 65 11 Sober for 6 years.
39 Heterosexual 63 2 Drinking heavily on and off since HIV diagnosis.
40 Heterosexual 35 5 Relapsed after diagnosis, now sober over 3 years.
121
Table 29. Descriptive Information about Male Participants in Open-Ended
Interviews (n = 27)
ID Sexuality Age Yrs. HIV+ Alcohol Use Status
2 Heterosexual 50 10 Sober > 5 years, drank heavily for 8 mos. post Dx.
3 Gay 40 5 Sober for 3 years.
4 Gay 39 6 Currently drinking.
5 Gay 45 12 Sober for 4 years.
6 Heterosexual 45 13 Sober 12 years, drank for 6 months after diagnosis.
7 Gay 40 20 Sober 10 years, drank for 10 years after diagnosis.
8 Gay 39 13 Sober for 6 years.
9 Gay 37 10 Sober for 8 years.
10 Gay 45 18 Currently drinking.
11 Gay 39 16 Sober for 15 years.
12 Gay 45 12 Sober for 9 years.
13 Heterosexual 35 1 Has been drinking heavily for the last 10 months.
18 Bisexual 54 14 Was sober 6 years, relapsed, now sober 30 days.
19 Gay 49 23 Currently drinking, has been on & off many years.
21 Bisexual 49 4 Sober for 2 years.
22 Heterosexual 36 3 Drinking moderately for the last 2 years.
23 Gay 52 19 Sober for 13 months and on & off for 15 years
25 Bisexual 44 1 Currently drinking.
26 Gay 44 13 Sober for 10 years.
27 Gay 45 23 Episodic drinking over years since HIV diagnosis.
28 Gay 39 23 Sober for 14 years.
31 Gay 42 9 Sober 6 years but has relapsed twice since Dx.
32 Gay 48 19 Sober 10 years, drank heavily for 1 year after Dx.
33 Gay 38 16 Sober for 12 years.
34 Bisexual 44 5 Has been binge drinking since HIV diagnosis.
35 Gay 50 5 Sober for 4 years, drank heavily after diagnosis.
37 Gay 32 2 Sober 16 months, drank heavily after diagnosis.
122
Overview of Qualitative Findings
Analysis of participants’ narratives in the open-ended interviews led to the
identification of emerging patterns that illustrate the processes and experiences
involved in both the genesis and resolution of alcohol abuse and dependence in the
context of HIV/AIDS for these participants. Based on these patterns, a very general
conceptual model for these processes was developed that created some degree of
tension with the assumption, inherent to the research questions, of a direct and
simple relationship between alcohol misuse and HIV self care. Thus, an
understanding of the elements of this conceptual model may be useful for the
comprehension and contextualization of the findings as they are presented.
The three research questions seek to identify the factors that influence alcohol
abuse and dependence in the context of HIV/AIDS, the impact of alcohol problems
on HIV self care and the factors critical to recovery in the context of HIV/AIDS.
These questions presuppose fairly direct cause and effect relationships wherein
outside factors influence participants’ alcohol use, which in turn influences
participants’ self care behaviors. The patterns that emerge from the analysis of
participants’ narratives, however, instead suggest that both alcohol abuse and
dependence and the failure to address HIV care needs are set into motion by the
interaction of the same set of contextual factors. The proposed model is illustrated in
Figure 8.
123
Figure 8. Factors Influencing Alcohol Misuse and Poor HIV Self Care
As shown in Figure 8, the first element of the overarching conceptual model
suggested by participant’s narratives is a backdrop of historical events such as abuse
suffered during childhood and victimization experienced as an adult. The next
element is a critical point (or series of critical points) at which the deficits created by
these historical losses and traumas compromise the individual’s ability to cope with
contemporary life problems such as financial hardship, housing instability,
relationship problems and educational and vocational frustrations. In this
conceptualization, the increased stress and depression caused by these experiences
then influence the individual toward both alcohol abuse and dependence and the
failure to address HIV care needs.
Three other general themes that emerged from participants’ narratives
underscore this model and should be considered in relation to the findings as
presented. First, reflecting the mechanisms outlined in the proposed conceptual
model, participant narratives suggest that both abstinence from alcohol misuse and
appropriate HIV self care are made possible through interventions that mitigate the
effects of the contextual factors identified in Figure 8. Participants articulate that
Depression
and
Stress
History of
Loss and
Trauma
Alcohol
Abuse
and
Dependence
Contemporary
Life
Problems
Poor
HIV
Self Care
124
effectively addressing both alcohol abuse and dependence and HIV care issues
requires addressing and learning to cope with underlying stress and depression,
which requires both addressing deficits created by underlying historical issues and
learning how to manage challenging contemporary circumstances.
Another theme is that the relationship between depression and alcohol misuse
and HIV self care is cyclical in nature, meaning that while depression influences one
toward alcohol misuse and poor coping with HIV-related needs, alcohol misuse and
the failure to address HIV care needs also strengthen feelings of hopelessness that
reinforce and intensify depression. Finally, a third theme is the deeply intertwined
nature of substance abuse and HIV/AIDS in the African American community,
articulated as the belief that the proposed model reflects only one aspect of a system
in which the same contextual factors predispose individuals to both alcohol problems
and HIV infection.
Factors Influencing Alcohol Misuse in the Context of HIV/AIDS
Analysis of the interview transcripts revealed a number of common themes in
participants’ assessments of the factors that propelled their alcohol abuse and
dependence in the context of HIV/AIDS. As described, these factors were often
articulated as the driving forces behind both alcohol misuse and the failure to deal
proactively with HIV-related needs. The affective experiences that participants
described most often were depression, denial and feeling chronically overwhelmed
by life problems. Participants often linked these experiences to issues such as lack of
love or respect for the self, lack of self-awareness of knowledge of the self and the
125
failure to address longstanding “traumas and tragedies.” Many participants also
described their alcohol misuse and poor functioning as being propelled by feelings of
social isolation. They attributed this isolation to the failure of intimate relationships
thwarted by histories of familial or personal addiction and dysfunction, to the
inability to seek HIV-specific support because of the need to hide being HIV positive
and to outright rejection and abandonment by family members, friends and lovers
after disclosure of HIV status. These themes are illustrated in Figure 9.
Figure 9. Factors Influencing Alcohol Misuse in the Context of HIV/AIDS
126
As shown in Figure 9, two patterns emerged with regard to the relationship
between HIV diagnosis and the onset of alcohol abuse or dependence: For some
participants, their initial HIV diagnosis and the fear, anger and depression associated
with it were clearly linked to an escalation in alcohol use that ultimately led to
problems. Participant 37 recalled that at the time of his diagnosis, “I was angry. I
was upset. I was scared. I had a lot of shame, a lot of guilt. And I wanted to numb
it by drinking or partying with my friends...basically not feel what I was feeling. It
didn’t work, though...it intensified.” Participant 14 commented that when she was
diagnosed with HIV “I didn't know about support groups. I didn't know about none
of that. I just thought it was a death sentence, and I was just going to drink until I
was dead, to be honest about it.” For some, like Participant 26, an escalation in
drinking after HIV diagnosis represented an effort to escape the reality of living with
HIV permanently:
I think I drank more because I just didn’t accept myself, I didn’t
accept my situation, and I didn’t really have any hope for the
future. I didn’t think I would be as old as I am today back then. I
didn’t think I would live this long, and, you know, so to me it’s
kind of like I was kind of copping out without copping out. I don’t
believe in suicide, but I believed in, if I was drunk and something
happened, then to me it wasn’t suicide.
For other participants, problems with alcohol misuse were well established
before they were diagnosed with HIV. Participant 32 noted that “people drink for all
sorts of reasons. You drink because you’re an alcoholic. And so things could be
good or things could be bad, and...anywhere in between, you know. You actually
have a physical condition that just craves the alcohol.” For some of these
127
individuals, like Participant 33, alcohol problems created a context in which HIV
seemed like just one more consequence of an addicted lifestyle: “Before I found out I
was HIV positive I was drinking and using...that was a part of my life, drinking and
smoking weed and doing drugs. That had become so normal that it really didn’t
matter. The HIV really didn’t matter, you know?” Some participants attributed their
pre-existing alcoholism to being “an addict by nature,” noting that they were unable
to explain why alcohol remained so appealing to them in the face of the problems it
had caused in their lives. Others attributed their pre-existing alcohol problems to
long-standing patterns that they traced back to the pain of childhood abuse and/or to
growing up in family environments in which addictive behavior was normative.
Related to these patterns, the depression that many participants identified as a
principal factor in their alcohol misuse seemed to fall into two categories. The first
was a persistent underlying depression related to deep, unresolved emotional traumas
and losses. Many participants linked childhood issues such as abuse, neglect and
early parental rejection or abandonment to this depression and subsequent alcohol
misuse in adulthood. Participant 30 observed, “I mean grief or loss, or rejection, or
abandonment pops up and immediately, you know, sometimes I get depressed...I just
want to say forget it and get a drink, call it a day.” Participants also related deep
feelings of depression to traumas that occurred during their adult lives, including
unresolved grief over the loss of friends, partners and children to HIV and rejection
by family members, partners and friends because of their HIV status and/or
sexuality. Participant 23 observed of the cumulative effects of these experiences that
128
“obviously, the times that I was sober, there was a sense of I did care about myself as
a person. The times where I was not were times where I was really lost.”
Participants also described a quiet sorrow associated with the reality of living
with HIV over time as something that contributed to persistent feelings of
depression. Participant 4 commented that “I think being HIV positive...just brings
depression on you, and actually you're dying, your body's dying, and you're just
taking medication to make yourself better. So you're living with this terminal
illness.” Other participants, even as they articulated the relationship between
depression and alcohol misuse, also acknowledged its futility. As participant 16
noted, “I think a lot of people drink because of pain, memories, thoughts. No matter
what - how much you drink or whatever - guess what? When it's over, you've got
another memory working for you, and this one you may not remember, but
everybody else will.”
Participants described the other “type” of depression leading to alcohol
misuse and poor functioning as a response to more transient although cumulatively
oppressive concrete issues and struggles in their day-to-day lives. These issues
included problems related to finances, like joblessness, struggling to survive on a
fixed or unstable income and going hungry or resorting to charity because of an
inability to meet basic needs. Participant 24 reflected that “you know, when you first
find out you have HIV, what is the first thing you think? You’re going to die. I’m
through with that...it’s the joblessness that’s more hurting me. That puts me in a
state of depression.” Participant 4 also commented that:
129
Money is a major issue. I mean, I can't do anything. I can't go
nowhere. I can't see my family... I'm in a hole. I mean, I'm just
trying to get out of this hole. Like it's hard, so it'll be like, when
you feel like giving up, I guess, the drinking just like numbs it.
Participant 39’s observation reflects that both the concrete aspects of being unable to
make ends meet and her shame over this reality contribute to her depression: “I
mean then I, I’m hungry...So that makes it hard, you know, having no funds and no -
the attitude of not doing for yourself, you know? That makes it kind of hard.”
Many participants also related bouts of depression to problems with housing,
such as living in uncomfortable and undesirable quarters, repeated evictions, periodic
homelessness that required imposing on friends and family for temporary sleeping
quarters and living in shelters or on the streets. Participant 4 described his frustration
with “getting evicted because you don't have 100 bucks and you can't find anybody
that'll loan you $100. I mean, that's kind of -- you know. I don't want to be in that
position. So, that causes a lot of drinking and stuff like that.” He goes on to note:
I need stability...housing. I mean, like, there's so much stuff that I
got to get for me to feel relaxed...If I'm not relaxed, I mean,
drinking just takes the edge off. You know, I got the housing
[problems], HIV, I got a son, I got all of this stuff and like the
drinking, just like, okay, forget it. Just having a drink just deters
all of the thoughts and like just push them in the background.
Participant 25 also reflected that it was difficult for him to build a secure foundation
for good mental health, sobriety and self care when his basic needs like housing were
not adequately met:
My first thing would be comfortability, and what I mean by that is
a roof over your head and having a roof as comfortable for you as
you can make it. I mean that takes a lot of depression away right
then...I know once I get home I would be, you know, I could do
130
this in here or whatever...But it’s kind of hard living on the streets
and living in a shelter. You’re dealing with the HIV, you’re
dealing with your living arrangements, your financial situation.
Participants also described the chronic and worsening health problems
associated with HIV illness and the practical demands of managing HIV care as
factors that contributed to their depression and subsequent alcohol misuse over time.
They described their physical limitations, HIV medication side effects like
neuropathy and chronic diarrhea and periodic acute illnesses and hospitalizations as
debilitating. They also described demanding and complicated medical care regimens
as well as interactions with bureaucracies that required “running around” in order to
secure benefits and care for themselves as frustrating and discouraging. Participant
16 described how even taking care of her HIV illness could bring up complicated,
intense feelings that might lead to drinking:
Every day for a year I was taking 13 pills in the morning, so every
day I took those pills it reminded me I have HIV, which it reminds
me of who, what, when, how and everything. You know? So of
course I'm going to feel guilty and angry. And then I just tell
myself it's my fault. You know, and then that makes me even
madder, you know?
Finally, even while describing the influence of depression on their alcohol misuse,
some participants reminded the interviewer of their awareness of the self-defeating
nature of this response cycle. Participant 32 noted that “if you’re depressed and you
drink it’s like a cycle; you drink more because you’re depressed. And then you drink
because you can’t stop drinking. So once you start you can’t stop.”
Many participants identified a feeling of social isolation as an important
factor influencing their alcohol misuse in the context of living with HIV/AIDS. For
131
some, this isolation reflected a sense of displacement created by their diagnosis and a
belief that “people don’t understand” them anymore. Participant 27 explained that
outside of settings where he knows that those around him are also HIV positive, this
feeling of alienation is even reinforced by casual and ostensibly friendly interactions:
[If] I’m just around other people, we’re just talking in general
and...someone asks me a question and I tell them about it, and we
get to talking about it...they’ve not experienced or had anything
happen in their life that I have...Most people are like, “Well, I
don’t know how you made it through that” and “Ooh, I couldn’t
go through that” and stuff like that. That’s when I want to take a
drink, because right at that moment I feel different. I feel totally
different than anybody and everybody else. I feel like I don’t fit in
and I don’t belong, and those things are ostracizing me right at that
moment. In order to not be feeling that stuff mentally a drink
sounds real good.
Participant 39 admitted that since her HIV diagnosis, she had withdrawn from casual
social interactions even though she knew this behavior was not healthy for her and
potentially led to more drinking. She explained that “before I found out I was HIV, I
could talk to people, and didn’t have any problem. But, you know, the stigma has
been put on, you know, you’re this and you’re that. And so I just withdraw from,
you know, trying to associate with people.”
For many participants, the sense of isolation that influenced their alcohol
misuse appeared to reflect a dearth of positive social support from family and friends
during critical periods of their lives. Participant 5 ventured that “if people...friends
and family and stuff would've been more supportive in situations or problems that I
was going through at the time, maybe I would have drunk a lot less.” This lack of
support from intimate friends and family seemed to stem from several distinct
132
underlying patterns, the first of which was characterized by relationships having
been cut off in the wake of participants’ addiction-driven behaviors. Participant 19
offered an example of this phenomenon: “I've done messed up so much, and your
family gets tired of it. They get so sick and tired of it that they don't have faith and
they don't believe in you, and even when you be doing good, there's always this
motive that they think you're doing, you know, and so that keeps me down.”
The second emergent pattern underlying participants’ limited social support
was characterized by participants’ failure to disclose their HIV status or seek HIV-
specific support from (non-HIV positive) friends and family. This pattern was often
attributed to shame and/or the expectation that disclosure would lead to rejection and
condemnation. For some, it involved lying about the nature of their illness and the
details of their lives and for others it involved cutting off all contact in order to
conceal evidence of their failing health. Reflecting on how his failure to be open
with his family about his HIV illness had propelled his isolation and subsequent
drinking, Participant 11 commented that
If I would've came to them, opened up more, I think I wouldn't
have went that route. Honestly. And they got mad about that.
They knew that. Yeah. It's like, "You know, we talk about
everything in here, you know, everything and everything," which
is true, you know, and they just didn't -- they couldn't understand
me. I couldn't understand myself at the time.
For many participants, however, limited social support appeared to be an
even more direct consequence of the stigma associated with HIV. This pattern was
characterized by participants’ disclosures of their HIV status and requests for support
being met by measured warmth or outright rejection. Some participants’ families
133
responded to their disclosure of being HIV positive by refusing to openly discuss or
acknowledge HIV. Participant 20 described this response from her family as less
than reassuring: “Seem like everybody don't want to talk about it, you know...They're
there, you know, for support, but they don't want to -- it's just awful news, I guess, to
them. How do I -- you know, if they don't talk to me, I don't know how they feel.”
In other participants’ families, specific family members offered support while others
could or would not. Participant 35 described this pattern:
The sisters they readily acknowledge it. They ask me how my
health is and everything. But my older brother, he never mentions
it. I mean I’m sure he knows. My two kids, they know. They
don’t mention it, though...their only question to me is how is your
health? And, and for a while I was in the hospital all the time, but
now that I’m not in and out of the hospital all the time, they, they
don’t even mention it. They might to my sisters but they never say
it to me.
Other participants’ opportunities for social support were limited by their
families offering to support behind closed doors but asking them to conceal their
HIV status from others. Participant 11 described this response as leading to further
feelings of isolation:
We sat at the dining room table, all of us, you know, and they said,
"Well, we ain't going to tell nobody else in the family. This is just
between us. We here for you, you know." But, you know, I think
that made it worse, I just felt embarrassed. And I think it had a lot
to do with me doing all the drinking and stuff.
Some participants’ sources of potential social support were depleted through
outright rejections they experienced in response to disclosing their HIV status and
seeking support. Participant 26 related that “my best friend for like 17 years, we
used to do everything together, movies and shopping and trips, and after disclosing
134
my illness just wanted nothing to do with me. He just sent me a real nasty letter, we
were never friends, and don’t contact him anymore.”
Finally, the influence of limited social support on some participants’ alcohol
misuse appeared to stem from their engagement in intimate relationships and social
networks in which there was no support for abstaining from alcohol misuse. Many,
like participant 20, directly related their continued drinking to “being around other
addicts who are drinking, you know, instead of going to a support -- you know, being
around people who want to help me love myself, so I can deal with these issues.”
Participant 23 described entering into a romantic relationship with an active
alcoholic as the impetus for his own problems with alcohol misuse:
The reason that I actually picked up the drink...I was going through
a period of depression and...I became romantically involved with
this person and I fell in love...I knew the moment I saw him, that I
was going to be in trouble. I knew, you know? And I was just
feeling at my worst then, you know? And he drank. And so in
order to, I guess, belong or to feel belonging to something, I just
gave in, you know? I get -- I just got so tired of just feeling so
alone, you know?
For a variety of reasons, many participants seemed to believe that ‘healthy’
relationships were not a readily available option for those living with HIV/AIDS.
Some, like participant 19, acknowledged that this perception propelled them toward
relationships that did not offer positive social support or real intimacy:
If I meet somebody, I have to tell them that I'm HIV, whether it's
intimate or just friends, because when I don't tell them...I run into
somebody and they know. And then I get afraid that they're going
to go and tell this person because we know each other and they
know that person. Then that could put me in danger of getting beat
up, possibly killed. So hanging out with lowlifes, people that are
135
just not going to ask and they're not going to care, is a whole lot
easier from having healthy relationships.
The Impact of Alcohol Abuse and Dependence on HIV Self Care
Participants were asked to describe all of the practices and activities they
engaged in regularly in order to address their HIV-related needs. From content
analysis of the range of responses provided, five interrelated domains for HIV self
care emerged. These five domains included physical care, emotional and
psychological care, spiritual care, recovery as self care and continuing education
about HIV/AIDS and are modeled in Figure 10.
Figure 10. Five Interrelated Domains of HIV Self Care
The domain of physical self care encompassed attending medical
appointments regularly, communicating or “partnering” with the medical care team
and following their instructions, having lab work done regularly and on time, taking
136
all medications as prescribed, learning about medications and “partnering” with
pharmacists in order to maximize medication efficacy, responding proactively to
minor infections and illnesses and attending to dental care needs. Physical self care
also included developing healthy personal habits like eating a balanced and nutritious
diet, taking vitamins, drinking more water, abstaining from recreational drugs and
alcohol, exercising and staying active, getting enough rest, avoiding stress,
maintaining good personal hygiene and practicing safer sex.
The domain of emotional and psychological self care encompassed formal
activities like attending support groups, participating in therapy, volunteering as a
peer educator or speaker and participating in activism as well as activities like
playing sports, making art, meditating, taking care of pets and taking bubble baths.
This domain also encompassed personal habits like minimizing stress, cultivating
supportive relationships with others who are HIV positive, sharing with and learning
from them, seeking out positive, uplifting people, staying in touch with family and
friends, maintaining mutually nurturing relationships with partners, finding people to
talk with when problems arise, attempting to help others but also setting boundaries
and knowing when to stop helping and making space for inspirational experiences.
Finally, this domain also encompassed attitudinal habits like trying to cultivate an
open mind and positive attitude, trying not to dwell in anger or anxiety, focusing on
creating a future to look forward to and being grateful for life’s gifts.
The domain of spiritual self care encompassed activities like attending
church, praying, meditation and yoga, reading the Bible, listening to gospel music,
137
listening to religious tapes and communicating with others about religion. This
domain also encompassed believing in God, putting faith in God, turning troubles
over to God and doing things “right.” Finally, it included accepting life on life’s
terms, staying in touch with oneself, forgiving oneself and treating oneself well,
accepting others as they are and trying to stay spiritually grounded by staying
humble and trying to feel connected to the larger universe. Recovery as self care
encompassed abstaining from drugs and alcohol, participating in alcohol treatment,
learning about alcoholism, learning to open up and seek help from others as needed,
attending Alcoholics Anonymous meetings and engaging in the 12 step program by
connecting with a sponsor, working the 12 steps and being of service to others.
Continuing education about HIV/AIDS encompassed making an effort to learn more
about HIV/AIDS and stay up to date with knowledge development about HIV illness
and treatments as well as learning about medications being taken and new
medications being developed.
Participants were also asked to describe what kinds of things interfered with
their HIV self care. Content analysis of these responses focused on the impact of
alcohol abuse and dependence on HIV self care. As most participants had gone
through periods of heavy drinking as well as periods of sobriety since being
diagnosed, many were readily able to chronicle the ways in which their self care
changed during periods of heavy drinking. Participant 36 offered a useful summary
of these experiences:
When I’m drinking I just focus on drinking. And, you know what?
I do a lot of not loving myself, you know, and I let a lot of things
138
go like that I wouldn’t do it when I was sober. You know, like
take, having responsibilities, like paying bills and taking my
medicine. I don’t take it at all when I’m not sober. Going to
doctors’ appointments, I don’t do that at all. I isolate. No support
system. Stop attending my groups. It’s different.
Analysis revealed that participants described alcohol misuse as having both
short and long term impacts on their HIV self care. Figure 11 presents a model of
the ways in which participants’ alcohol misuse impacted their HIV self care.
Figure 11. The Impact of Alcohol Misuse on HIV Self Care
Many participants reported that initially, problematic drinking contributed to a
general failure to acknowledge or respond to their HIV diagnosis in a timely or
139
proactive fashion. They also noted that in the longer term, their drinking often
resulted in chronic or periodic neglect of their medical, emotional/psychological and
spiritual care needs in specific and predictable ways.
As shown in Figure 11, participants’ descriptions of their failure to respond to
their initial HIV diagnosis because of alcohol misuse outlined two separate but
related phenomena. The first involved denying and using alcohol to maintain denial
about being HIV positive. Participants described this pattern as using alcohol to put
HIV “on the back burner” or to keep themselves “in a fog” in order to maintain
denial about the reality of their diagnosis. For some, like participant 3, this occurred
in the context of ongoing drinking: “All it did was help to block it out of my...me
even thinking of it, more or less. So I would keep myself fully drunk, intoxicated...to
avoid having to even think of it, or think about it.” For others, like participant 40,
this pattern involved a relapse: “When I was diagnosed with HIV, I had used alcohol
before but I had stopped for like four years. But because I was diagnosed with HIV I
went back, because I wasn’t into an acceptance. And so instead of being out of
denial, I was in denial.” For many participants, like participant 2, failure to respond
to their diagnosis also involved postponing the establishment of an appropriate HIV
care regimen: “I didn't seek medical attention. I didn't seek what medication I
needed to take. I didn't try to educate myself about it. I just like thought it was a
death sentence and just accepted it. I didn't try to do something about it. I didn't
fight back.” For others, like participant 29, this pattern also involved failing to
disclose their diagnosis to others for prolonged periods of time:
140
I was so hooked up in...my alcoholism, it didn’t bother me for a
while, because I didn’t care, you know...it took from, from ’89 till
’98 for to really -- I didn’t ever tell my family or nobody between
them times, because I felt that my family I’m real close to, but I
felt they’re going to reject me, you know?
Participants’ chronic or periodic misuse of alcohol also propelled the neglect
of their physical self care in more long-term ways, even after acceptance of their
diagnosis and initial establishment of care. Two of the ways that participants most
commonly described themselves as neglecting their physical care were failing to
adhere to their medication regimens and failing to engage with their medical care
teams. The latter often involved not making or keeping appointments and not being
honest with care providers about alcohol use or lapses in self care. As participant 18
described, “when I was drinking...I would miss doctor's appointments, postpone
them, put them off, lie to my doctors, wouldn't take my meds. I'd look at my pillbox
at the end of the week. It would be full compartments where I'd just miss doses.”
Participants often attributed these behaviors to periods of intensive alcohol use being
characterized by depression, low self esteem, lack of investment in the idea of taking
care of oneself and “not thinking clearly.” Participant 14 reflected on a period of
heavy drinking in his life:
I wasn't taking medicine. Drinking was my medicine...I was on
some regimen...I don't even know the name of the pills now, but
there were a lot of them, and they were big, and I didn't take them.
I would drink, and...I just thought, "I'm going to die with this
disease. Let me die."
Others, like participant 9, recalled being too ashamed of themselves during periods
of heavy drinking to seek medical care: “I still had conscience. I was still conscious
141
of...how I looked and how I appeared to people, and me being drunk and trying to go
to the VA Medical Center, I couldn't see it. I couldn't do it.”
In addition to neglecting their medical care, many participants also described
slipping into unhealthy personal habits during periods of heavy drinking. This
pattern often included disruption in basic functions like eating regularly, nutritiously
or at all, drinking water, getting adequate rest, exercising and attending to personal
hygiene. Participant 10 described these periods as times when he would “stay up all
night. I stay up for days. And then when I do rest, it be like a crash. And then
sometime it won't even be really long before somebody was knocking on my door
waking me up again.” Participant 9 noted that during these periods “I wouldn't dress
the same. I wouldn't clean myself. The hygiene was terrible...I just wouldn't brush
my teeth or nothing. And just, the only thing that was on my mind was that next
drink.”
In addition, many participants described periods of heavy drinking as times
when they became “reckless” about sex, more promiscuous and much more likely to
engage in unsafe sex. They described drinking as helping them to “forget” about
HIV and muting their concerns about exposing others to the virus. They explained
that while drinking, such behavior could be “blamed on the alcohol, ” which both
lowered their inhibitions and made them less likely to “go through the whole
rigmarole” of breaching the topic of HIV with a potential sex partner and taking
physical precautions. Participant 21 recalled that when drinking heavily, his attitude
about safe sex was altered:
142
It doesn't even pop into your head. You just say, ‘Okay, well, I
shouldn't be doing this.’ You know what I'm saying? Or maybe I
should allow this person to have a choice in the matter before I
decide I want to sleep with them, disclose that, you know. You're
more promiscuous. The behavior just change.
Analysis also revealed that during periods of heavy drinking, participants
often neglected what they identified as their emotional and spiritual HIV care needs.
Many participants described this neglect as driven by a sense of hopelessness and by
the habit of postponing addressing their needs such that they failed to “take control
of life and health.” As participant 19 described, when drinking heavily, “I pick what
I want to deal with and don't, and the things that I deal with are the easy things, and
the things that I don't want to deal with, I cover them up.” Participant 18 explained
that “when it [drinking] gets to be every day, your body just gets tired, and your
mind just gets tired, and you're depressed.” Participants described periods of heavy
drinking as times when they not only tended to become increasingly depressed but
also were less likely to address symptoms of depression even upon recognizing them,
creating a “vicious cycle” that became difficult to escape. Participant 7 noted that
It [drinking] impairs the mind...it kind of puts depression in your
mind, because you think about "Okay, I got this thing that I could
die at any time," then you're drinking, and then that makes you
more and more depressed about, you know, your status, and how
people perceive you.
Many participants also described withdrawing from the religious and spiritual
practices that might have otherwise sustained them during periods of alcohol misuse,
noting that religious and spiritual matters were “not on the agenda” or “just totally
ignored.” Participant 24 recalled that “when I started that [drinking], I stopped going
143
to church as much. I just had problems. You hear the Lord talking to you, but you
don’t hear that. I mean you don’t want to hear it. You’re in denial. Alcohol, that’s
all I got to say.”
Finally, participants also described neglecting their emotional and spiritual
needs during periods of heavy drinking by disengaging from both formal and
informal sources of support. This pattern included withdrawing from participation in
supportive services like HIV support groups, recovery support groups, Alcoholics
Anonymous meetings and individual therapy. Participant 15 recalled about her own
disengagement that “I couldn’t fool nobody with alcohol on my breath, you know, so
I didn’t go.” Other participants described isolating themselves from family, friends
and informal networks of HIV positive and recovery supports. They described
themselves as “recluses” who were typically alone with “all of the windows closed,”
“off in the corner somewhere drinking” during these times in their lives. They
described cutting off contact with family and friends for days or weeks at a time, not
wanting to talk to anyone and functioning just enough to “take care of the house and
pay the bills.” Participant 30 recalled both the isolation and the shame involved in
living this way:
I’d feel uncomfortable and...disappointed, not only so much in
myself but in what people think, you know, and then all these old
crazy feelings, like well, I’m going to avoid that because I’m
not...and then I avoid the phone calls, that they call. And I don’t
return the calls, you know, but that’s how my disease is, you
know? And I mean that’s just how alcoholism is, you know?
Participant 14 noted that “my mom would come and check on me, and I would have
to hide my liquor and tell her I'm not drinking. But, I mean, a person know, they
144
could smell it on your breath. So you're really lying to yourself.” Others, like
participant 40, noted that withdrawing from loved ones also meant failing those who
trusted or depended on them:
After I started drinking, I stopped taking care of my kids, my
husband had to get outside help. As soon as I started drinking, it
was just me and the alcohol. I wasn’t taking care of myself
because I didn’t do nothing, I laid in the bed all day. I rationalized
that as resting but in reality, now I can’t rationalize that.
Finally, many participants seemed profoundly aware of the lasting
consequences of their alcohol-driven failures in self care, even as they continued to
struggle with sobriety. Participant 27 recounted “my drinking affected...my living
with HIV because I wasn’t taking my medication when I was drinking...So I became
resistant to a lot of meds. My meds have been changed quite frequently because
when I drink I don’t take them.” Others, like participant 8, reflected that it was only
as they began to address their alcoholism in the context of HIV/AIDS that they
developed an awareness of the interrelated nature of physical, emotional and other
forms of self care:
I didn't realize that I needed to take care of the emotional side, you
know what I'm saying? I got to the point I was taking -- trying to
take care of the physical and ignoring the emotional, and that was
where the depression was coming in at, and I wasn't understanding,
which was turning around and affecting my physical.
Other participants articulated that neglecting their physical and psychosocial HIV
care needs had contributed to ongoing problems with both alcohol and HIV/AIDS.
145
Participant 17 commented on the effects of alcohol use on the progression of his HIV
disease and, conversely, the effects of progressed HIV disease on his ability to
address his alcoholism:
You don't take care of yourself physically or you don't take your
medicines, you don't see the doctor. The HIV progresses. The
immune system gets weaker. The alcohol more impairs the
immune system. And you subject yourself to a lot of opportunistic
or germs and infections that'll take care -- take advantage of a
suppressed immune system. So when you get into treatment, you
have some physical problems, and it takes time to deal with the
physical problems before you can even get to the addiction
problems.
Factors Critical to Recovery in the Context of HIV/AIDS
Participants identified a number of factors that they viewed as critical to their
recovery from alcohol abuse and dependence in the context of HIV/AIDS and to
effective HIV self care. These processes, modeled in Figure 12, largely represented
the inverse of those described as leading to the misuse of alcohol and to poor HIV
self care.
Participants described many of the processes that helped them to recover
from alcohol abuse and dependence as having been accomplished in the context of
their involvement with various kinds of supportive services, including 12-step and
other recovery support programs, alcohol treatment programs and multidisciplinary
HIV care systems. However, they almost uniformly described the most critical
element of recovery as something individually and internally determined. They
described making the decision to recover and committing to doing so as something
that had to happen and come from inside of them. As participant 15 explained, “it
146
was just my will to live. Everything that everybody was telling me was useful, but
the bottom line, it was up to me.”
Figure 12. Factors Critical to Recovery in the Context of HIV/AIDS
Participants also consistently reported that in order to enact their decision to
get sober, they had to develop enough “self love” to create the determination to
“make it.” Participant 33 commented that “I think you have to realize that you’re
worth saving...you have to care enough about your own life to say ‘Okay I don’t
want to live like this anymore’” and noted that “caring about myself and caring
147
about my own life has kept me sober, you know, seeing how precious my life is, you
know, seeing how good life is, makes you want to live life, you know?” Many
participants also described being “ready” as a requisite for reaching this threshold
and asserted that no matter what resources were available to them while they were in
the throes of addiction, nothing could be of any help to them until they were ready to
recover. Many also described the process of “hitting bottom” as a part of this
process and as essential to their subsequent determination to stay sober. Participant
40 asserted that “I think people have to reach their own bottom to get to it, to either
die, you know, to drown or swim.” Participant 32 commented about the process of
recovering in the context of HIV/AIDS that:
HIV is one of the things that brought me to a bottom. Very
quickly. I mean there were a lot of things that I drank over -- I
really drank over the shame of being gay and self-hatred. And
then once I had to deal with the HIV [card.] I had -- it pushed, it
really put in perspective, what are you really ashamed about, you
know?
Participants also described the process of making a successful commitment to
recovery as requiring taking responsibility for themselves and their actions,
developing a sense of purpose about their lives and establishing concrete personal
goals. Participant 5 described his determination to maintain his sobriety in the
context of coming to terms with past mistakes:
I want to do the dang thing. I done messed up so much in the past,
I'm tired of messing up. Yeah. I just want to do something right,
accomplish something, do something right for once...I'm not
getting no younger...So I got to have something, some kind of
accomplishment that, you know, I can put on my belt or something
before like God come and lift me up out of here.
148
Participant 40 reflected that her desire for her life to leave a positive footprint
propelled her to address both her alcohol misuse and her HIV care needs so that she
could be functional and well enough focus on the things that were important to her:
I love my life. I love my family. I believe that we only have one
closing statement, you know, we’re born, we live, we die, so I
don’t want to die with a negative, I don’t want to leave negative
things lingering on my kids’ heads or on my sisters’ and brothers’
heads. I want to be something, a positive asset. And so that’s why
I want to live life to the fullest and every day I expand myself.
Although participants described all of these internal processes so critical to
their recovery as self-determined, many also directly related their determination to
the cultivation of a personal relationship with God. They explained that it was this
sense of connection to their own spirituality or to a “higher power” that enabled them
to love, respect and forgive themselves and to seek out and appreciate the positive
aspects of their lives even in the context of HIV/AIDS. Participant 31 noted that his
spirituality was what helped him pull out of repeated relapses because “it helps me
realize that obviously I have a plan and a purpose on this earth.” Other participants
described their spirituality as providing inspiration, guidance to make better choices
and the ability to accept help from others.
Many participants reflected that although they had grown up with religious
practice, the recovery involved a process of spiritual transformation or renewal for
them. Participant 33 recalled that “I really didn’t become really spiritual until I got
into recovery...But before then...I was brought up in the church, so there was some
religious undertones that were still with me even though I was drinking and using,
you know?” Finally, while some participants described their relationship with God
149
as essential to maintaining sobriety, asserting that they “couldn’t do it alone,” others,
like participant 27, described their spiritual awakening in the context of recovery as a
more contemplative process:
I’m restructuring and rebuilding my faith, you know, through
spirituality. Not through church and religious based things...it’s
about my conscious contact and my understanding of who God is,
for me...and that’s how I’m learning how to redirect all that stuff. I
don’t always immediately think to go there. But when I do, I go
there and I go there deeply, and I go there with some longevity...I
find somewhere or someplace to just be with me and God, you
know, and tell him what’s going on with him, and talk to him like
I’m talking to you, you know, and that helps me, it helps me get
out of the depression, it helps me get out of the obsession of
wanting to take a drink and it, it helps me to do the right thing.
Another process that many participants described as critical to recovery from
alcohol abuse or dependence in the context of HIV/AIDS was that of addressing the
issues “at the root” of their alcohol misuse. They described this as a process of self-
examination that involved developing greater self-awareness and self-knowledge,
reinterpreting past experiences and learning new ways to deal with problems. Many,
like participant 12, attributed the necessity of this process to the fact that alcohol
misuse was only the symptom of deeper problems:
Alcohol and drugs are a minute problem in regards to your
addiction. Usually, it’s something around you or something that's
happened to you in life that allows you to medicate to make you
forget...And most addicts, they forget about these things and they
start using to hide, to run. And that's where I think the problem
lies...certain things that we’re brought up in growing up that we
just don't know how to deal with, so we use those as outlets.
Feeling inadequate. Not completing school, for me. There are so
many things that just -- I just want to forget about.
150
Participant 36 noted that learning new skills to deal with old pain had helped her to
avoid problem drinking:
A lot of the drinking and the emotional problems was based on my
childhood. So when I go into therapy, we work on that, you know,
anger problems and things like that. And they give me a lot of
tools where I can use, so I won’t have to go to the liquor store. I
have another alternative to go and seek help before I act, do
something negative that might hurt, make the situation worse.
Similarly, participant 17 described sobriety as “accepting some truths about yourself
and learning to fashion a life around those truths, you know, and keep them in the
forefront...you know, you think about those things before you make any decisions
about anything, in order to maintain the lifestyle that you have, because you know
that with one drink it'll all change.” Finally, others, like participant 12, described
dealing with underlying emotional issues as laying the groundwork both for his
recovery and for his ability to cope with HIV/AIDS:
Well, when you're in the process - and when we say the process,
that's the process of recovery - it allows you to step out of yourself
and look at some of the things that made you or allowed you to get
into your addiction. And once you have accepted the things or the
issues that you allowed to get you into your addiction, then you
can work on them. And then you'd feel better about yourself. And
then you'd want to take care of yourself. You'd want to enjoy the
rest of your life. You'd want to be productive in life. It kind of --
it pretty much brings you back into the -- a full circle in regards to
having the will to live. Life becomes so meaningful. You have a
purpose now. And your primary purpose isn't just that drug or that
drink now. It's all about you and keeping the focus on you.
Many participants noted that the process of self-examination they went
through in the recovery process included incorporating HIV/AIDS into their life
narratives in more positive ways. Some described this experience as making a
151
decision to live with HIV rather than die from it. Participant 37 described this
process:
The shift in thinking, of looking at it as not like a death sentence
but like an opportunity to, for me to appreciate every second that I
have here, every day that I have here. Realizing that there’s like a
lot of help that I can give just by being positive to the community,
and not as it being a death sentence or shame, but like an
opportunity for me to tell my story. Just to live, to be an example
that you don’t have to like use drugs and all this other stuff just
because you’re positive, but you can live a healthy life, you know?
And be positive.
For others, it was important to take stock of the positive things that HIV had brought
into their lives. Participant 22, for example, came to see HIV as “a gift and a curse.
The gift, it made me realize that I didn't want to die. The curse is that I'm HIV
positive and I have to deal with the stigma of being HIV positive.” He went on to
note that “just being HIV positive helps me deal with just life, period, and just how I
see things. Taking care of my HIV is the most important.”
A number of participants described undergoing a “psychic shift” after
receiving their HIV diagnoses that made them see life as more valuable, realize that
they wanted to live and begin to prioritize preserving their health. Participant 8
recalled that “at first I looked at this HIV as a curse, but the way it made me take
control of my life and made me take control of my health, it was like a blessing in
disguise, so to speak, you know what I'm saying?...Because I want to live, and I
enjoy living.” Finally, some participants described coming to terms with their own
mortality as providing the critical impetus for getting sober. Participant 38 noted
152
that after being diagnosed, she realized that “I had the HIV, I was stuck with that.
But I could get rid of the drinking.” Participant 17 recalled of his HIV diagnosis:
It was one of the determining factors that stopped the drinking.
Before being infected, I didn't care, and didn't feel like I had
anything to live for. Once I became infected and realized that if I
didn't change my life I was going to die, and that I really didn't
want to die, and I really didn't want to live the way I was living, I
was able to do some change - make some changes in my life.
The last process that participants identified as a critical facet of recovery in
the context of HIV/AIDS was learning how to create and use systems of positive
social support. Participants described cultivating and maintaining supportive
relationships with non-HIV positive or recovering friends and family as especially
important. Participant 27 described the innate value of “picking up the phone and
calling someone, you know, that I could talk to that knows my history.” They
described family members and significant others as providing encouragement,
direction and support that kept them motivated and focused. Participants often
described their mothers, especially, as being uniquely willing to seek out care for
them and push them to utilize it, sometimes by going so far as to take them to
support group meetings or accompany them to medical appointments.
A number of participants noted, however, that that these supportive alliances
with old friends and family were made possible only by the disclosure of their HIV
status, which was often a difficult process. Some participants recalled relying on key
family members to whom they were particularly close to “spread the news” to other
family members in ways that would garner support. Participant 6 described being
able to spend significant amounts of time with his family after many years of not
153
doing so only because of his efforts to address their fears: “I kind of educated them
about it so they’re at peace with that now.” Participant 35 recalled that
At first, I said, "You know, I’m going not going to let everybody
see me taking all these pills," because I used to have to take a lot of
them. And the first reaction, from my sister, was, "If you’d quit
taking so many pills you might feel better." And then probably
about three weeks later, she looked at me and she said, "You know
what? If you don’t take those pills you’re going to die, huh?" And
I said, "Now, you’re getting it.” And you know what? From that
day on things got a lot easier.
Several participants speculated that if they had felt their family’s support or been
comfortable seeking and accepting help earlier, their alcohol misuse would not have
escalated to the degree that it did.
Participants also described the tremendous value of developing social support
systems within the recovery and HIV communities because of the issue-specific
support provided within them. Participant 29 noted that “I have support groups and
friends that used to be alcoholic, they clean for ten years, five or six years, and I
know them and I feel comfortable around them...it makes me feel so good, support
by people that understand it better than people that don’t have it.” Participant 3
described his recovery supports by saying “They really inspire me. They give me a
lot of high spirit and a lot of better thinking of myself and seeing myself as a better
person, understandings of myself...It helps keep me stronger than I was in the past.”
Participant 27 noted that “I go to support groups, and I’m able to sit in the room and
share a lot of that stuff. And when I leave I don’t want to go take a drink.” Some
participants also described how valuable recovery supports were for developing new
social circles that helped them avoid “shady areas” and “old using buddies.”
154
Participant 16 recalled, though, that in her struggle with sobriety, encountering such
people in a new milieu has provided support and reinforcement:
I see people at the meetings, you know...people I probably -- God
knows what I did with them. They got their wives, their kids and
everything now. And...the addict in me, of course, says, "Well, I'm
glad you're here, because if I ever went back out there, I'd be
looking for you, because you had a good deal." And he says, "If
you're ever looking for me," he said, "I'll be here Wednesday night,
six to eight, right here at this book study." And I looked at him
and I said, "Okay." And he said, "And I hope you're going to be
here, too," and I'm like, "Yeah."
For many participants, the process of establishing a positive social support
system also involved terminating relationships with partners, friends and family
members who were not supportive of them. Many participants described “changing
your friends and where you socialize” as an essential element of recovery because, as
participant 18 related, “It makes it impossible...when you're hanging out with people
like that, they never leave you alone. They're constantly at your door, especially
certain ones, just very persistent and very influential...almost like leeches and pests.”
Participant 1 recalled that she had to extricate herself from a long-term romantic
partnership in order to begin taking care of herself, noting that “I had to let him go
along with the drinking.” In addition to disengaging from relationships with active
substance abusers, participants also noted the need to avoid “negative” people,
especially HIV positive individuals who refused to address living with HIV/AIDS in
proactive or positive ways. Sadly, some participants noted the necessity of quashing
expectations of support from friends and family who had consistently proven
155
themselves unsupportive due to fear of and bias toward people with HIV/AIDS
and/or homosexuality.
Finally, many participants described having the opportunity to provide
support to others as what made them feel they had “come full circle” in terms of the
successful creation of a supportive social network. Participants described providing
support to others in the context of mutual aid and educational groups, through
mentoring those newly diagnosed with HIV or newly sober and through public
speaking and HIV prevention efforts. Many saw the process of providing support or
“giving back” to others, in relation to both recovery and HIV/AIDS, as critical to
their own sobriety and as something that gave value to their life experiences,
however negative. Participant 31 affirmed the importance of this process to the 12-
step model:
I believe, which is what the program has taught us, in one addict
helping another. And not to say that I’m going to counsel addicts,
but I know that I got to help a group of people. And I have so
much experience of heartache and abandonment and loss and
depression and low self-esteem that it’s obvious I have to share
that and help other peoples get over it, you know?
Participant 27 also reflected that he was glad to have gone through some of his most
“miserable” moments if doing so might help others in the long run:
Maybe it could help somebody else sometime to experience some
of the negative stuff that I’ve experienced in my life, just by telling
them, you know, this is what I went through. And it keep them
from going through it, they can take another choice. You know,
and that, that’s my biggest motivator, you know, with my life
experience. I could share my life experience with someone who
has not had these experiences yet, who may not have to have this
experience, if I share with them what I’ve been through.
156
Many participants identified that the value of these experiences lay in helping
them develop positive self-esteem and pride in their ability to do something right by
helping others. Participant 27 described sharing his life story with others as
something that gave him hope; “I feel like I’ve completed something to have had a
positive effect on another human being’s life. I don’t need no drink because I can do
that again.” Several participants commented that speaking to young people in their
communities was especially rewarding. Participant 32 noted that his experiences
speaking about HIV in schools were sometimes intimidating but powerful:
I think disclosing to, whether it be a high school student or middle
school students, that, you know, that they have the ability to
protect themselves and that they’re worth protecting...empowering
them and motivating them, to be empowered, to say, you know, if
you’re in a sexual situation that if they’re not going to have sex
with a condom then they’re not going to have sex is just amazing.
Many, like participant 32, reflected on the role that sharing their personal
experiences with others had played in their own transformations:
it’s so great, like in the 12-step program, to just, on a moment’s
notice...reveal who you are to another person. Very powerful to
tell your story. And then you get to know your story, you get to
know yourself. And then you can’t really talk about something
that you don’t really believe in or that’s really not truly you. So it
makes you very authentic.
Finally, others like participant 23 noted that these opportunities not only gave them
greater perspective on their own journeys but made them feel “glad to be at a place
now where I can actually look back and see, you know, that by going through what I
went through, I’m still here.”
157
Issues of Gender, Sexuality and Culture
In many ways, the narratives examined in relation to the three qualitative
research questions revealed few distinct differences based on gender; the patterns
described in relation to the etiology and impact of alcohol misuse in the context of
HIV/AIDS as well as the factors critical to recovery from it were largely similar for
men and women. However, gender-based differences did emerge with regard to
specific aspects of the contexts these processes occurred for male and female
participants. Most notably, female participants were far more likely than male
participants to discuss various aspects of their addiction, disease and recovery
experiences in the context of parenthood and in relation to caring for children and
meeting other familial responsibilities. Participant 20, for example, described her
HIV diagnosis itself as having occurred in the context of becoming a parent:
I was in [Hospital X], and the doctor came and told me that the
baby had it, and then they said it -- and I told the doctor, I said,
"What is that?" So I was like zero on the education about it. New
information...It was just really a blow. Yeah. I didn't know
nothing about the disease. I saw, '93, on TV, you know, the AIDS
commercials coming on. That was all I knew.
Participant 20 also noted that in the absence of any knowledge about HIV/AIDS, she
felt particularly overwhelmed by her care needs because “I had to find out how to
take care of a baby with AIDS and myself.” Indeed, women often described their
relationships with their families as both rewarding and stressful, because of the
burdens related to parental and other care giving responsibilities. Participant 39
recalled that when she was hospitalized with pneumonia, her primary concern was
that “all my children were there and they were looking at me, and they were
158
thinking, ‘Oh, you look just like grandmother.’” Many, like participant 40, described
these relationships and their burdens as being related to the ways that they coped
with HIV/AIDS, addiction and recovery in complex ways:
When I first stopped drinking alcohol, I wanted to be the perfect
mother and the perfect wife. But then I have to remember why I
started drinking alcohol; because I couldn’t be that perfect mother
and that perfect wife, and then I couldn’t cope. So in order for me
to cope in a healthy way, which is not using drugs and alcohol, I
have to make sure that everything is in order for me. It makes it
hard on my family sometimes because in a way I’m kind of selfish
and self-centered, but in a way then I can be there rather than not
be there. Take the positive with the negative.
Finally, women were much more likely than men to discuss the HIV-related
deaths of children and the loss of custody of children due to uncontrolled addiction
as consequences of and contributors to the depression, guilt and shame that propelled
their alcohol misuse and poor HIV self care. Participant 36 was diagnosed with HIV
at the same time that her nine month old infant was diagnosed with and subsequently
died of AIDS and described how this convergence of events made it impossible for
her to care for herself properly: “That was a lot of grief and, you know, in the
process of that, then I had a double diagnoses. I had to take care of myself with the
virus and my child with the virus, too. So that was overwhelming, too.” Participant
29 ruefully described her failure to “get it together” after completing treatment as
what led to the final loss of custody of her son. Participant 20 described
“remembering my situation with my son” (who was born HIV positive) as one of the
chief triggers that propelled her multiple relapses into alcohol misuse in subsequent
years.
159
Because the men interviewed for the qualitative analysis were far more likely
than women to describe themselves as gay or bisexual, they much more often
identified their sexuality as a major stumbling block to securing support from their
families and communities in relation to HIV/AIDS and alcohol misuse. Participant
23 felt that the tacit lack of acceptance of homosexuality and bisexuality in his
community as he matured had left him vulnerable to both HIV and the alcohol
problems that he subsequently developed:
When I first came out I didn’t even have a clue as to, as to what
that meant, you know? So, you know, there were no supposedly
role models or books that told you the right way to be gay. So it
was about experimentation which led to God knows what. I think
that if I had had someone who, who could have kind of given me
direction, you know, as to what to expect, you know, that might
have helped.
Other men described their experiences coping with HIV and alcohol misuse
as being colored by both generalized homophobia and personal alienation from their
families and communities based on their sexual orientation. Many, like participant
32, described these attitudes as being tightly interwoven with the stigma surrounding
HIV: “There is so much stigma and...if there is any other self hatred -- maybe
because you’re gay and culturally maybe because you’re African American and
you’re gay... In my situation, my family disowned me because they found out I was
HIV positive.” Many men viewed resolving these issues as seen as pivotal to their
recovery. For some, this resolution occurred through their ability to surmount past
rejections and reconnect with their families after recovery, while for others, it
160
required them to appraise their losses in ways that enabled them to move on from
them and develop alternate sources of social support.
The nature of participants’ experiences with religiosity and spirituality in the
context of HIV/AIDS and alcohol use also appeared to be influenced by issues of
gender and sexuality. As reviewed, many participants described the development of
a personal relationship with God as key to manifesting the changes they needed to
create in their lives in order to recover from alcohol abuse or dependence. As such,
most participants described both religiosity and spirituality as very important to them
and both men and women frequently emphasized the benefits of having a personal
spiritual practice. Participant 29, for example, explained that “I believe in God, you
know, and I always have and that does put the spirit within, you know, to keep me
going, so maybe I won’t feel as bad...my spirituality helps me to forgive myself.”
Formal ritual involvement with a church or religious community, however,
seemed to be an intensely positive experience and influence for some participants but
a negative one for others. This conflicting mixture of attitudes and experiences
about religion and spirituality appeared to be differentiated by gender and sexuality.
Women, who were predominantly heterosexual, were far more likely than the men
interviewed to described intensive involvement with traditional churches as an
important part of their lives and recovery. Participant 24 described her Church
attendance as key to maintaining her equilibrium:
I go to church...at least once a month. That helps me along the
way, also, because you got to have that. You can’t just pray to
Father every day, you have to...get more out of it...I get a lot of
people, I feed off of what the Lord has put the preacher to speak to
161
us about today, that’s what I get...I can take that and use that...so I
can have a level head about life and stay positive and stay strong
and stop hurting yourself, and all of this -- that’s what the preacher
is telling us when we go there, which we already know, but
sometimes like a parent still in our life, that’s why I go. I need to
hear what this person is saying. I understand what I’m thinking
and what I’m praying for, but what is he saying? So I’m getting all
the opinions.
Male participants, who were more likely than female participants to be gay or
bisexual, more often focused on the importance of their personal spiritual
experiences, rather than formal religious involvement, to their recovery. When
asked what aspect of his religious and spiritual experiences was most critical to his
recovery, participant 33 provided a response typical of the men in the sample:
I would say my religious and spiritual connection. God, you know,
he intervened, and he let me know that there was another way, that
I didn’t have to live, I didn’t have to continue living my life the
way that I was living it. That there was help and if I sought out
help that I could find a better way to live, and that’s what
happened, you know?
Participant 33 went on to specify that his personal spiritual practice included keeping
himself “spiritually centered,” which he accomplished through a variety of practices:
I might listen to gospel music, pray, meditate, and just know that
God, that’s what I call my higher power, my spiritual, you know,
provider, I just try to stay grounded and stay humble, and try to
accept people for who they are, which helps me stay spiritually
connected, you know? Staying in touch with who I am. And I do
that by trying my best to accept other people with their faults, with
their different backgrounds, different religions, by realizing that,
you know, we’re all different, but at the same time we’re all human
and we have all the same basic needs.
Much more often than female participants, male participants described their
spiritual practices as being cut off from the formal religious institutions in which
162
they were raised. A number of these men described direct experiences of rejection
based on religious attitudes toward homosexuality and, associatively, HIV/AIDS that
left them feeling alienated from mainstream churches. Participant 35 explained that
“at a Black church, I still don’t exist. There is no such thing as a Black man with
HIV. You don’t exist, don’t even mention that in our presence, you know?”
Despite these types of negative experiences, however, some men described the
power of their familiarity with the Church as causing them to turn to it in their most
desperate moments. Participant 18 described “showing up at...the Church...just on
my knees, begging, ‘Somebody please help me.’”
Still other men described formal religious involvement as a positive source of
support and guidance in their recovery process in the context of becoming intensely
involved with alternative churches and ministries. These alternate houses of worship
were typically created by and for gay men, gay men of color and/or those with
HIV/AIDS and as such served a very specific set of congregants. Participant 35 was
careful to distinguish his experiences with this kind of religious community from his
experiences with other Churches, noting “there’s a whole big difference between
Black churches and the church I go to.” Like many of the women interviewed,
participant 35 described his engagement with his religious community as critical to
his establishment of supportive social networks during and after recovery:
Now, I went back to church, and so that has a whole lot to do with
it. And it’s basically a gay church, but, you know, and when I
went there, I said, "What am I doing in here?" You know? But
now...it’s like I don’t care what I do Saturday night, Sunday
morning is, that’s my time there. And Wednesday nights I go
there, and like when I leave here I go there, and, you know, it’s --
163
and it’s fun, you know, it’s a community thing, so it keeps me
going.
Despite the predominance of this gender-based pattern, a few women also
described feeling alienated from their traditional religious communities in the wake
of their HIV diagnosis. Participant 39 attributed this feeling to the fear that
disclosure of her HIV status would evoke community censure. She likened her
situation as an HIV-positive Church member to that of a pregnant teenager in her
congregation:
So I know how they think, and I know that the majority of the
people do. And then people at Church sweep everything under the
rug, too, you know. So I go to Church, I praise God, and, you
know, I look at them...And then when a young person comes up
pregnant, then they make them repent. I said, "Why are you
making that child repent? She’s doing what’s normal. She love
that person. You missing something somewhere.” ...And I’ll stay
away from the church meeting or something, because I don’t agree,
they think I’m crazy.
Participant 39 also noted that this apprehension led her to censor herself in ways that
appeared to dilute the intensity and intimacy of her relationships within her religious
community and thus the potential social support she might receive through them.
Despite this awareness, she assessed trying to change the attitudes of her Church
community toward HIV-related issues as futile, saying “they don’t want to hear it,
and then they’re not open to it.”
Finally, it is critical to note that these gender and sexuality-related patterns
emerged within an overarching framework of consumer emphasis on the central
influence of culture in their lives. Participants illustrated the centrality of culture in
several ways. The first was through their contextualization of almost every
164
experience they described in terms of being African American. This was
characterized by frequent references to the importance and impact of a sense
connection to the larger African American community and the tight knit nature of
African American neighborhood, extended family and religious/spiritual
communities. Participants also emphasized the specificity of dealing with
HIV/AIDS and alcoholism in this cultural context because of the cultural patterns
highlighted. The second way in which participants emphasized the importance of
culture was through their descriptions of the adaptations and compromises involved
in living as African Americans within the larger majority culture. Many participants
asserted that the adoption of a defensive posture was required in order to survive as
an African American in the United States, regardless of the complications presented
by HIV/AIDS and alcohol-related problems.
Participants offered many examples of the many ways in which the African
American cultural context provided the potential for intensive social support. They
also, however, described the devastating isolation and alienation from this support
that might occur when community standards were transgressed. Participants
described many failings in their own lives, such as sexual promiscuity, sex work and
the betrayal of family members and family responsibilities, as having occurred in the
service of active addiction. When asked about the relative weight of these
transgressions in relation to an HIV/AIDS diagnosis, though, participants seemed to
concur that HIV/AIDS was less acceptable and more misunderstood, feared and
165
rejected in the African-American community than alcohol and other substance abuse.
Participant 12 echoed many participants’ assessments when he responded:
Least acceptable: HIV...Because there's so little that they -- that
we know about it. There's so little. I mean, you can just look.
Alcoholism and drug addiction, they've pretty much accepted that
as a way of life in our ghettos. I mean, they pretty much expect it.
They pretty much accept it. But now if someone was to walk in
with HIV, you'll notice how the certain prejudices come in.
Participant 21 noted that in his experience, this lack of acceptance of those
living with HIV/AIDS led to a very gender-specific lack of support for African
American men living with HIV/AIDS:
The African-American community is not ready for AIDS or HIV,
especially in black men...I think it kind of like started pretty much
in the gay community, and they got a grasp on it a lot further, so
they became more advocates for that type of base. There's not a lot
of advocates for, you know, African-American men, because we
don't want to talk about it, for the most part. It's pretty much kept
quiet, underneath the table. You want to keep it a secret. So
there's not a lot of support going on for that.
Others, like participant 31, asserted that this pattern of fear and secrecy impacted
both men and women in dangerous ways:
I feel like we should have been further along in dealing with
people with HIV than we are...especially in the African American
community. You know, so a down-low thing that’s going on
which is just not within the African American community, but, you
know, the rise of women getting HIV is just like, okay, let’s see
what the underlying stuff that’s going on. But so many people
want to stay blind, you know, and just say, “No, it’s them, them,
them.”
Many participants related this pattern to the centrality of the Church in African
American communities and resultant cultural attitudes about homosexuality and
166
HIV/AIDS. Referring to the influence of religious attitudes towards homosexuality
on cultural ones, participant 35 commented
It’s still a very big black stigma, you know? You cannot be Black
and that. And even though it’s the largest part of the HIV
population, they try not to make, you know, in the family structure,
Black people, they don’t want to admit it. You know?
Finally, participant 33 asserted that the interests of the African American
community might be served most expediently through the development of more
tolerant views with regard to HIV/AIDS and those living with it. He suggested that
increased attention to education and prevention efforts, specifically, might cut
through this atmosphere of fear and misunderstanding:
The key to that is education, so as African-Americans we must
educate ourselves, we must stay open, you know, to what’s going
on, not only in our communities but in the world and how it affects
us. And, we have to do that, we have to stay educated, we have to
stay focused on, you know, not only our health but, you know,
social situations, you know, our social status, which, you know, we
sometimes don’t do. We just look at the small picture, we don’t
look at the big picture...We have to teach our youth drugs and
alcohol are bad...whether they’re HIV positive or not. Because a
lot of times, drugs and alcohol will lead somebody to becoming
infected because their judgment is impaired, you know? Or
they’re drunk, "Oh, he looks okay, I don’t need a condom.”...so
hopefully education and outreach will be the critical areas where,
you know, where we can prevent the spread of HIV and AIDS in
our community.
167
CHAPTER 7: Discussion
This study is the first to explore the influence of gender, religiosity, social
support and social network composition on alcohol abuse and dependence among
HIV-positive African Americans. In addition, it utilized qualitative analysis to
explore consumer perspectives on the factors that influence the genesis and
resolution of alcohol misuse in the context of HIV/AIDS, as well as its impact on
HIV self care. This discussion addresses the findings from both the quantitative and
qualitative sections of the analysis, especially as they inform and relate to one
another. The findings are also discussed in relation to the literature reviewed,
limitations of the study are presented and recommendations are made for the
development of interventions based on the findings of this study and for further
research.
Discussion of Quantitative Findings
In alignment with the literature reviewed, bivariate analysis revealed
significant differences between men and women in the total sample in relation to the
variables under study; men evidenced significantly higher rates of current alcohol
abuse and dependence (Caetano, Clark & Tam, 1998; Green, Freeborn & Polen,
2001; Herd, 1997; Wilsnack, Vogeltanz, Wilsnack & Harris, 2000) and women
evidenced significantly higher levels of religiosity (Chatters, Taylor & Lincoln,
1999; Levin & Taylor, 1993) and perceived social support (Owens, 2003; Prado,
2004). In addition, bivariate analysis revealed that as the literature reviewed would
suggest, levels of religiosity and spirituality were significantly correlated in the total
168
sample (Miller & Thorensen, 2003; Zinnbauer et al, 1997). These analyses also
revealed that both religiosity and spirituality were significantly correlated with
perceived social support, a finding that is supported by the literature reviewed
(Mattis & Jagers, 2001; Prado et al, 2004; Turner-Musa & Wilson, 2006) but that
suggests the existence of relationships between these key variables that were neither
attended to nor included the models tested in this study. Based on these findings as
well as literature indicating a lack of conceptual clarity about the distinct and
overlapping properties of religiosity versus spirituality (Miller & Thorensen, 2003;
Zinnbauer et al, 1997), separate models were used to test the influence of both of
these variables in order to address Hypothesis 1.
Bivariate analysis also revealed significant differences between men and
women with regard to sexual orientation, in that women in the total sample
predominantly identified themselves as heterosexual while men predominantly
identified themselves as gay or bisexual. In addition, bivariate analysis identified
significant differences between those with and without current alcohol abuse or
dependence problems with regard to sexual orientation: In comparison to 11.5% of
those who identified themselves as heterosexual (N = 131) and 9.6% of those who
identified themselves as gay (N = 94), fully 27.7% of those who identified
themselves as bisexual (N = 47) were assessed as currently abusing or dependent on
alcohol. Based on these findings, sexual orientation was included and controlled for
in the models used to test Hypotheses 1 and 2.
169
Despite bivariate findings of differences in relation to key variables,
however, multivariate analysis failed to support any of the stated hypotheses. In
relation to Hypothesis 1, the models tested revealed no significant main effects for
religiosity, spirituality or gender, and no significant moderating effects for gender on
either religiosity or spirituality. In relation to Hypothesis 2, although the model
tested did reveal significant main effects for perceived social support, it indicated no
main or moderating effects for gender. Supplemental analysis of the perceived social
support subscales indicated that the significant differences between men’s and
women’s levels of perceived social support lay in their levels of perceived support
from family which, unlike perceived support from friends and significant others, did
not significantly influence current alcohol abuse and dependence. Furthermore all of
these models revealed significant main effects for sexual orientation, which resulted
in model significance that was not accounted for by the variables addressed in the
hypotheses.
Because logic based on the literature review tells us that more significant
relationships should be detected in these models with regard to both gender (Green,
Freeborn & Polen, 2001; Herd, 1997; Wilsnack, Vogeltanz, Wilsnack & Harris,
2000) and religiosity (Bazargan, Sherkat & Bazargan, 2004; Caetano & Clark,
1998a), it seems likely that Type II error played a role in these findings. With regard
to both Hypotheses 1 and 2, the potential to effectively analyze the moderating
influence of gender on religiosity, spirituality and social support may have been
compromised by the skewed distribution of gender in the total sample: Women
170
comprised only 26.8% of the total sample (N = 73) and only 13.5% of those in the
total sample assessed as currently alcohol abusing or dependent (N = 5). While this
imbalance is not unusual for a sample comprised of HIV-positive individuals, it does
pose analytical limitations when gender is proposed to function as a moderator.
It is also possible that the failure to detect significance in the model used to
test the moderated effects of religiosity may be related to measurement issues
inherent to the instrument used to assess religiosity and spirituality. In addition to
the aforementioned conceptual fuzziness surrounding the concepts of religiosity and
spirituality, reliability analysis revealed relatively low factor loadings of some of the
items for the religiosity section (as low as .57), indicating that these items are not as
conceptually cohesive as might be needed for the scale section to function optimally
within the model. Although the spirituality items appeared to demonstrate somewhat
more cohesiveness, the lack of significant differences found between men and
women and between those with and without alcohol problems with regard to
spirituality do not suggest that significant findings should have been expected at the
multivariate level. Finally, because the literature reviewed indicates that African
Americans almost uniformly exhibit very high levels baseline levels of religiosity
and/or spirituality when compared to whites and members of other ethnic groups
(Levin & Taylor, 1993; Galvan et al, 2007), these findings may speak to a lack of
cultural specificity in the measurement of these concepts.
Bivariate analysis of the sub sample leading up to the analysis of Hypotheses
3 and 4 revealed that those individuals included in the sub sample were more likely
171
to have a lifetime history of alcohol abuse or dependence and less likely to have
engaged in the heaviest drinking of their lives since receiving their HIV diagnoses
than those not included in the sub sample. However, bivariate analysis comparing
the two samples revealed that most of the significant differences found between men
and women in the total sample with regard to the variables under study were not
found in the sub sample; the only significant difference that persisted between the
genders was that of sexual orientation. Because of this finding, sexual orientation
was also included and controlled for in the models used to test Hypotheses 3 and 4.
Bivariate analysis of the sub sample also revealed gender-based differences in
specific aspects of social network composition, but not in those aspects of social
network composition that, based on the literature reviewed, (Bond, Kaskutas &
Weisner, 2003; Valente, Gallagher & Mouttapa, 2004) were included in the
multivariate analysis.
In relation to Hypothesis 3, the model tested revealed no main effects for
religiosity, spirituality or gender and no moderation by gender, indicating that no
moderation was present to be mediated by social network composition. In relation to
Hypothesis 4, the model tested revealed weakened main effects for social support but
no main or moderating effects for gender and thus again, no moderation present to be
mediated by social network composition. Additional analyses revealed that the
specific elements of social network composition selected for inclusion in the models
used to test Hypotheses 3 and 4 had neither significant main effects on current
172
alcohol abuse or dependence nor significant moderating effects on the influence of
perceived social support on current alcohol abuse or dependence.
Because logic based on the literature reviewed tells us, again, that more
significant relationships should be seen in these models with regard to gender
(Green, Freeborn & Polen, 2001; Herd, 1997; Wilsnack, Vogeltanz, Wilsnack &
Harris, 2000), religiosity (Bazargan, Sherkat & Bazargan, 2004; Caetano & Clark,
1998a) and social network composition (Bond, Kaskutas & Weisner, 2003; Valente,
Gallagher & Mouttapa, 2004) it seems likely that that Type II error played a role in
these findings as well. Failure to detect significance with regard to Hypotheses 3 and
4 may be related to the limitation in power associated with the reduced size of the
sample used to test the models associated with these hypotheses. This possibility is
suggested by the deterioration of the significance of bivariate relationships between
the total sample and the sub sample. Post-hoc power analysis indicating insufficient
power to detect significance in the sub sample confirms that this was likely a factor
in the findings related to Hypotheses 3 and 4.
Finally, the composition of the sample used for this study in relation to the
outcome variable of alcohol abuse and dependence may have compromised the
validity of the quantitative findings. Only 13.6% of the total sample participants (N
= 37) were assessed as currently abusing or dependent on alcohol. However, fully
75.4% (N = 205) of total sample participants had a lifetime history of alcohol abuse
or dependence. Thus, of the 235 participants (86.4% of the total sample) who were
not abusing or dependent on alcohol when interviewed, only 67 participants (24.6%
173
of the total sample and 28.5% of the “non abusing or dependent” group) had never
met the criteria for alcohol abuse or dependence at any point in their lives. Despite
the cross-sectional nature of the study, the conceptual clarity of comparing those
with and without current alcohol problems is somewhat muddied by the reality of
there being three significant groups with regard to the outcome variable; those with
current alcohol problems, those with no current or past alcohol problems and those
without current problems but with histories of alcohol problems of undetermined
intensity, duration and chronological proximity to the point of data collection.
Discussion of Qualitative Findings
The consumer narratives analyzed in the qualitative segment of this study
were largely recovery narratives provided by individuals who had achieved or were
struggling with recovery from alcohol abuse and dependence in the context of
HIV/AIDS. As such, many of the consumers interviewed were passionate about and
eager to share their experiences in an effort to help others understand the benefits
they had reaped through their recovery. Probably, at least in part because of this
dynamic, these narratives provided more than sufficient breadth and depth of
information to meaningfully address the qualitative research questions. This analysis
thus enabled the development of conceptual models of the factors that influence
alcohol abuse and dependence among HIV-positive African Americans, the ways in
which alcohol misuse interferes their HIV self care and the factors critical to their
recovery from alcohol abuse and dependence. These findings echoed many patterns
suggested in the literature reviewed about the juxtaposition of HIV/AIDS and
174
alcohol misuse in the African American community (Adimora & Schoenbach, 2002;
CDC, 2007b; Smith et al, 2003; Welch, 2000; Wright, 2001) and provided detailed
descriptions of consumer perspectives on the underlying mechanisms and meanings
of those patterns.
In relation to the factors that participants saw as influencing them toward
alcohol abuse and dependence in the context of HIV/AIDS, some participants
indicated that their alcohol misuse predated and was independent of their HIV
diagnosis, while others felt that being diagnosed with HIV had initiated, exacerbated
or re-intensified these problems. Participants also reported that depression, related to
both longstanding unresolved emotional issues and more contemporary
circumstantial issues, and a dearth of effective social support were the two factors
that most directly influenced them toward alcohol abuse and dependence in the
context of HIV/AIDS.
In relation to the impact of alcohol misuse on HIV self care, the participants
interviewed reported engaging in a range of HIV self care behaviors that included
those focused on their physical, psychological/emotional and spiritual needs as well
as those related to recovery and education about HIV/AIDS. Participants’ narratives
clearly supported the association of alcohol misuse with the neglect of those needs,
especially, as suggested in the literature reviewed, in relation to a delay in addressing
HIV care needs (Petry, 1999), the failure to engage effectively with medical care
providers (Flexner et al, 2001; Petry, 1999; Samet et al, 2003), poor medication
adherence (Cook et al, 2001; Flexner et al, 2001; Palepu, et al, 2003; Samet et al,
175
2004) and engagement in risky sexual behaviors (Chuang, Liebschutz, Horton &
Samet, 2003; Conigliaro et al, 2003; Parsons et al, 2004; Petry, 1999).
In relation to the factors critical to recovery from alcohol abuse and
dependence in the context of HIV/AIDS, various participants reported that their
participation in both structured chemical dependency treatment programs and 12 step
and peer support networks had been instrumental to their recovery. Almost all
participants, however, emphasized that the most critical foundational element to
recovering from alcohol abuse and dependence was their own readiness to change.
The processes that they emphasized as critical to making these changes included
increasing self-awareness of the underlying issues driving alcohol misuse and
creating and engaging with social networks that provide positive social support,
especially support specific to HIV and recovery issues. Many participants also
described the importance, for them, of developing personal relationships with God in
order to achieve these instrumental changes.
The ways that participants describe the issues and patterns associated with
their recovery processes echo the literature reviewed on the role of religiosity and
spirituality (Coleman & Holzemer, 1999; Latkin, Tobin & Gilbert, 2002), social
support (Beattie & Longabaugh, 1999; Peirce et al, 2000) and social network
composition (Bond, Kaskutas & Weisner, 2003; Valente, Gallagher & Mouttapa,
2004) in relation to alcohol misuse among HIV-positive African Americans. These
narratives also corroborate the assumptions inherent to the theoretical models
176
utilized in peer support-based recovery programs such as those based on the 12-step
model (Beattie & Longabaugh, 1999).
Triangulation of Qualitative and Quantitative Findings
Although the research questions addressed in the quantitative and qualitative
analyses were not identical, these analyses did overlap and the findings complement
one another in ways that facilitate the discussion of each in relation to the other. The
qualitative analysis of consumer narratives, especially, provided alternate
perspectives and emphases as well as a level of detail that enabled the examination
of gaps and contradictions encountered in the quantitative analyses. In several
instances these findings reveal contradictions and complexities in the variable
relationships examined that offer possible explanations for the lack of significance
found in relation to these variable relationships in the quantitative analysis.
As reviewed, despite bivariate trends and literature suggesting otherwise, the
quantitative findings revealed no main or moderating effects for gender on alcohol
abuse or dependence. In the qualitative narratives, however, it seems clear that
gender does, in fact differentiate the contexts in which these men’s and women’s
alcohol abuse and dependence unfolded and resolved. The qualitative findings
suggest that HIV-positive African American women maintain more intense
connections to their families than do their male counterparts. This finding supports
the quantitative finding that women’s social networks included significantly more
family members than men’s and that women reported significantly more perceived
social support from family members than did men. In addition, in support of the
177
literature reviewed (Owens, 2003), findings from the qualitative analysis suggest that
women’s greater engagement with family systems appears to provide both
opportunities for support and stress related to care giving and other responsibilities.
This may provide some explanation for why, in the quantitative analysis, women’s
greater perceived social support from family was not related to their lesser likelihood
of current alcohol abuse and dependence.
In addition, findings from the qualitative analysis strongly supported the
significance of sexual orientation that emerged in the quantitative analysis at the
bivariate level and as a confounding variable in the multivariate models. These
findings suggest that many of the gender-based differences outlined in the literature
reviewed and in the bivariate quantitative analysis may actually represent differences
based on sexual orientation. In short, men’s and women’s experiences differed
significantly at least in part because male participants were primarily gay or bisexual
while female participants were primarily heterosexual. Specifically, in the
qualitative narratives many male participants discussed experiences with
homophobia and related AIDS-phobia related to religious and cultural attitudes
toward homosexuality in African American communities that they describe as
pervasive. In many instances, they directly related these attitudes and experiences
to the lack of social support and depression that propelled their alcohol misuse. The
power, continuity and cultural specificity of these experience as described in these
narratives may provide some explanation of the unexpected strength of the
significance of sexual orientation in the quantitative models.
178
As expected from the review of the literature on African Americans and
religiosity (Levin & Taylor, 1993; Galvan et al, 2007) both the quantitative and
qualitative findings indicated that baseline levels of religiosity and spirituality were
quite high among study participants. However, as discussed, despite literature
attesting to the impact of spirituality and religiosity on alcohol misuse (Bazargan,
Sherkat & Bazargan, 2004; Caetano & Clark, 1998a), no main effects for either
religiosity or spirituality were found in the quantitative analysis. In addition, as
discussed, despite literature, bivariate and qualitative findings all supporting the
existence of gender-based differences in religiosity and spirituality (Green, Freeborn
& Polen, 2001; Wilsnack, Vogeltanz, Wilsnack & Harris, 2000), these differences
were not found to impact alcohol abuse or dependence in the multivariate models.
Within the qualitative narratives, however, the overwhelming majority of
participants described spiritual practices and, secondarily, religious participation as
intricately related to their recovery processes and as critical to their sobriety. This
supports the conclusion that issues related to the conceptualization and measurement
of these variables may have contributed to Type II error in relation to the
multivariate findings. In addition, the narratives examined in the qualitative analysis
suggests that gender-based differences around spirituality and religiosity existed
primarily in relation to participants’ experiences with and participation in formal
religious practice, rather than personal spirituality. In addition, many participants’
narratives suggest that these differences may be more directly related to sexual
orientation than to gender. This is largely attributed to both anticipatory perceptions
179
and direct experiences of negative religious attitudes toward homosexuality in the
African American community among the male participants interviewed. As such,
these qualitative findings provide some specificity about the gender-based
differences that exist in relation to religiosity as well as the etiology of these
differences. This, in turn, may provide some explanation for the contradictions
encountered between the bivariate and multivariate quantitative analysis with regard
to the relationships between gender, religiosity and spirituality and alcohol abuse and
dependence.
Findings from both the quantitative and qualitative analyses clearly support
the significance of the influence of social support on alcohol abuse and dependence
in this study. The quantitative and qualitative findings regarding the importance of
differences in social network composition, however, were less congruent. None of
the targeted aspects of social network composition (number of social network
members, number of social network members who encourage moderation or
abstinence and number of social network members who are heavy drinkers) were
found to have significant effects in relation to alcohol abuse and dependence in the
quantitative analysis. However, participant narratives analyzed in the qualitative
analysis revealed extensive discussion of the ways in which the composition of
participants’ social networks influenced their alcohol abuse and dependence and
recovery processes.
Most centrally, participants described socializing with other addicts and
avoiding “straight” social settings as a characteristic pattern of active alcohol misuse.
180
In addition, they described restructuring the composition of their social networks as a
key step in their recovery from alcohol abuse and dependence. More specifically, in
line with the literature reviewed (Bond, Kaskutas & Weisner, 2003; Valente,
Gallagher & Mouttapa, 2004), they described this process as critical to their
development of the supportive social systems needed to establish and maintain
sobriety. Also in line with this literature, participants described this restructuring
process as including both the establishment of social networks populated by those
supportive of the participants’ sobriety and the exclusion of those who were
unsupportive of their sobriety. Finally, according to participants’ descriptions, those
not supportive of sobriety included those who had alcohol or other substance abuse
problems as well as those who were rejecting or unsupportive of the participant due
to their attitudes about HIV/AIDS or issues related to the participant’s sexuality,
once again reinforcing the untested but clearly far-reaching underlying significance
of sexual orientation in relation to the issues under study.
Implications for Alcohol Interventions with HIV-Positive African Americans
The findings of both the quantitative and qualitative segments of this study
suggest that interventions with the goal of reducing alcohol abuse and dependence
and its impact on HIV self care among HIV-positive African Americans should
focus on utilizing and enhancing the sources of social support available to these
individuals. Social support-focused interventions typically emphasize helping
participants to recognize, gain access to and utilize supportive relationships in order
to address immediate concerns and develop skills for addressing future problems
181
(Belgrave, 1998a). Variations on this model for intervention are used in a variety of
chemical dependency treatment modalities, most commonly in those that rely on the
12-step model of providing semi-structured peer support for recovery from drug and
alcohol use.
As reviewed in the literature (Belgrave, 1998a), there is a historical tradition
of emphasis on the family, extended family and kin, religious and community
networks among African Americans that promotes a communal, relationship and
interpersonally oriented reliance on informal groups of similar others to address
problems and concerns. As such, the goals and structural elements of social support-
focused interventions are likely to be culturally congruent with and draw on the
strengths of traditional values and practices in the African American community.
However, the findings of both the quantitative and qualitative segments of this study
suggest that in the case of interventions addressing the needs of HIV-positive
African Americans around alcohol misuse, some tension with traditional African
American religious and cultural attitudes is also likely. Thus, in this instance, a
translational approach must be used to strike a delicate balance between utilizing
traditional sources of cultural strength like interpersonal inter-reliance while
simultaneously challenging culturally entrenched apprehensions and responses
among participants in the intervention and in the larger community. Efforts to
increase social support must therefore be effected in conjunction with efforts to
increase education about HIV/AIDS and issues of sexuality among religious and
community leaders and stakeholders.
182
The mechanics of a formal intervention focused on enhancing social support
must emphasize several specific structural elements. The intervention must be based
on shared goals and clear objectives that define the structure and content of the
interventions. These goals must represent congruence between leaders and
participants in order to establish the collaborative relationships essential to recruiting
and retaining participants by reinforcing the need for and purpose of participation
(Belgrave, 1998a). Such an intervention should also be shaped by a well defined
structure and format, including attention to elements such as the composition and
responsibilities of both leaders and participants, group norms, size, setting and
duration and formats for the presentation of materials and ideas (Belgrave, 1998a).
Finally, the content of the social support-focused intervention must be clearly
linked to its desired outcomes. Thus, it should include an initial review of the
purpose and nature of intervention. Content should then focus on the development
of group cohesion and explanation of the concept of social networks or social circles.
Exercises should be geared toward helping participants identify their own social
networks and other positive sources of social support and helping participants learn
how to identify, in a given problem situation, whether or not support is needed, what
kind of support is needed if so and who or what is the most appropriate source of that
support. Content should also focus on helping participants learn how to provide
support to others and on seeing social support as a source of empowerment versus
dependence. Finally, content should include the examination of how cultural values
both encourage and inhibit the participants’ access to and utilization of social
183
support. Strategizing around these issues should include an examination of
participants’ expressed and unconscious values and assumptions and of how these
attitudes impact their own behavior and expectations of others (Belgrave, 1998a).
Educational efforts targeting the larger community should be geared toward
building on the good will that religious and community leaders are assumed to have
toward HIV-positive, gay and bisexual community members who are suffering and
in need of assistance. These efforts must focus on addressing the issues of bias
identified in this analysis and on pointing out the potentially damaging effects of
these attitudes and practices in a culturally sensitive manner. These messages must
be delivered from the standpoint that actions taken in the past have been based on
ignorance of their ill effects rather than any direct intention to cause harm to
vulnerable community members. In addition, these efforts are most likely to be
effective if they are initiated by community-specific advocacy and supportive service
entities. Members of such groups possess a certain degree of community buy-in and
can speak directly and personally to the strengths and resources that religious
communities have to offer HIV-positive African Americans with alcohol abuse and
dependence problems.
Recommendations for Future Research
Discussion of both the quantitative and qualitative analyses presented here
suggests several potential directions for additional research that might build on the
findings of the current study. The first is the examination of the relationship between
sexual orientation, especially bisexuality, and alcohol abuse and dependence among
184
HIV-positive African Americans. The findings indicating that sexual orientation is a
significant factor contributing to alcohol misuse in this population appear to be
consistent between the quantitative and qualitative analytical segments of the study.
However, due to the small number of participants in both analyses and especially the
small number of those participants identified as bisexual, these results should be
examined in a context that allows a directed examination of this trend using a larger
study sample.
In addition, findings from the qualitative analysis in the current study,
especially, suggest that further research should focus on the influence and interaction
of social support and depression as the primary influences on alcohol abuse and
dependence among both male and female HIV-positive African Americans.
Investigation of this nature might lead to the development of models that more
parsimoniously account for increased odds of alcohol abuse and dependence in this
population and to stronger evidence for developing effective interventions in this
arena.
Study Limitations
This study is not without limitations, which are driven by a number of issues
in the execution and design of the quantitative segment of the study, in particular.
One of these issues is difficulty surrounding the accurate conceptualization and
measurement of the key variables of religiosity and spirituality, which might be
rectified through the use of a better validated, more cohesive and more culturally
specific measure of these variables. There are also concerns regarding limitations
185
based on the composition of the study sample. One of these issues is potentially
insufficient power to detect significance due to the limited sample size, especially in
the analyses utilizing the sub sample. Another issue related to sample composition is
the skewedness in the gender distribution of the sample, given the importance of
gender-based differences and the role of gender to the hypotheses and multivariate
analyses.
Additionally, there are concerns about the predominance of those with former
alcohol abuse or dependence problems in the group evaluated as free of alcohol
abuse and dependence, due to how this reduces the clarity of differences between
groups in relation to the outcome variable. The ability to generalize and predict
outcomes based on the quantitative findings of this study are necessarily limited
because of the relatively small size of the total sample and the cross sectional nature
of the data. Confidence in the findings arising out of the qualitative analysis must
also be limited due to the participation of a single researcher in the analysis and the
inability to conduct any sort of member check with participants in the open-ended
interviews.
Conclusion
Despite the bivariate significance of key variable relationships, the
quantitative findings of this study failed to support any of the four hypotheses posed.
In addition, the study’s overall focus on the significance of gender to the experiences
of alcohol abuse and dependence among HIV-positive African Americans was not at
all manifested by the quantitative results. The qualitative data, however, offered a
186
breadth and depth of information about alcohol abuse and dependence among HIV-
positive African Americans from the consumer perspective that provided rich
opportunities for further analysis. This analysis did lend support to the validity of
gender as a significant issue in relation to this phenomenon. It also highlighted the
significance of sexual orientation in relation to the relationships under study and
helped to highlight the mechanisms through which this significance may arise.
Finally, the qualitative analysis provided detailed information about the influences
on and impacts of alcohol abuse and dependence among HIV-positive African
Americans and information about consumer experiences with recovery from alcohol
misuse that hold significant implications for the development of interventions
intended to address this phenomenon.
Thus, it is hoped that despite the failure of this analysis to support any of the
stated hypotheses, results from this study might contribute to the delivery of effective
alcohol treatment services to HIV-positive African Americans and to an increased
awareness of the significance of alcohol-related problems in HIV treatment and
prevention efforts targeting this heavily impacted community.
187
REFERENCES
Adimora, A. A. & Schoenback, V. J. (2002). Contextual factors and the Black-White
disparity in heterosexual HIV transmission. Epidemiology, 13, 707-712.
Airhihenbuwa, C. O. & liburd, L. (2006). Eliminating health disparities in the
African American population: The interface of culture, gender and power.
Health Education and Behavior, 33, 488-501.
Baron, R. M. & Kenny, D. A. (1986). The moderator-mediator variable distinction in
social psychological research: conceptual, strategic and statistical
considerations. Journal of Personality and Social Psychology, 51, 1173-1182
Bazargan, S., Sherkat, D. E. & Bazargan, M. (2004). Religion and alcohol use
among African American and Hispanic inner-city emergency care patients.
Journal for the Scientific Study of Religion, 43, 419-428.
Beattie, M. C. & Longabaugh, R. (1999). General and alcohol-specific social support
following treatment. Addictive Behaviors, 24, 593-606.
Beatty, L. A., Gaiter, J. & Wheeler, D. (2004). HIV prevention research for African
Americans: Current and future directions. The Journal of Black Psychology,
30, 40-58.
Belgrave, F. Z. (1998a). Guidelines for implementing a social support intervention.
In F. Z. Belgrave, Psychosocial Aspects of Chronic Illness and Disability
Among African Americans. (pp. 67-86). Westport, CT: Auburn House.
Belgrave, F. Z. (1998b). The role of social support in adaptive functioning among
African Americans with chronic illnesses and disabilities. In F. Z. Belgrave,
Psychosocial Aspects of Chronic Illness and Disability Among African
Americans. (pp. 101-119). Westport, CT: Auburn House.
Bird, C. & Rieker, P. (1999). Gender matters: an integrated model for understanding
men’s and women’s health. Social Science and Medicine, 48, 745-755.
188
Bond, J., Kaskutas, L.A. & Weisner, C. (2003). The persistent influence of social
networks and Alcohol Anonymous on abstinence. Journal of Studies on
Alcohol, 64, 579-588.
Bowie, J. V., Ensminger, M. E. & Robertson, J. A. (2006). Alcohol-use problems in
young Black adults: Effects of religiosity, social resources and mental health.
Journal of Studies on Alcohol, 67, 44-53.
Brooks, R. A., Etzel, M. A., Hinojos, E., Henry, C. L. & Perez, M. (2005).
Preventing HIV among Latino and African American gay and bisexual men
in a context of HIV-related stigma, discrimination and homophobia:
Perspectives of providers. AIDS Patient Care and STDs, 19, 737-744.
Buseh, A. G., Stevens, P. E., McManus, P., Addison, J., Morgan, S. & Millon-
Underwood, S. (2006). Challenges and opportunities for HIV prevention and
care: Insights from focus groups of HIV-infected African American men.
Journal of the Association of Nurses in AIDS Care, 17, 3-15.
Caetano, R. (1997). Prevalence, incidence and stability of drinking problems among
whites, blacks and hispanics: 1984-1992. Journal of Studies on Alcohol 58,
565-572.
Caetano, R. & Clark, C. L. (1998a). Trends in alcohol consumption patterns among
whites, blacks and hispanics: 1984 and 1995. Journal of Studies on Alcohol
59, 659-668.
Caetano, R. & Clark, C. L. (1998b). Trends in alcohol-related problems among
whites, blacks and hispanics, 1984-1995. Alcoholism: Clinical and
Experimental Research, 22, 534-538.
Caetano, R. & Clark, C. L. (1998c). Trends in situational norms and attitudes toward
drinking among whites, blacks and hispanics: 1984-1995. Drug and Alcohol
Dependence, 54, 45-56.
189
Caetano, R., Clark, C. L. & Tam, T. (1998). Alcohol consumption among
racial/ethnic minorities: Theory and research. Alcohol Health and Research
World, 22, 233-238.
Cargill, V. A., Stone, V. E. & Robinson, M. R. (2004). HIV treatment in African
Americans: Challenges and opportunities. Journal of Black Psychology, 30,
24-39.
Catz, S. L., Gore-Felton, C. & McClure, J. B. (2002). Psychological distress among
minority and low-income women living with HIV. Behavioral Medicine, 28,
53-60.
Centers for Disease Control and Prevention. (2007a). CDC HIV/AIDS Fact Sheet:
A glance at the HIV/AIDS epidemic. Atlanta, GA: CDC, National Center
for HIV, STD and TB Prevention, Division of HIV/AIDS.
Centers for Disease Control and Prevention. (2007b). CDC HIV/AIDS Fact Sheet:
HIV/AIDS among African Americans. Atlanta, GA: CDC, National Center
for HIV, STD and TB Prevention, Division of HIV/AIDS.
Centers for Disease Control and Prevention. (2007c). CDC HIV/AIDS Fact Sheet:
HIV/AIDS Among Women. Atlanta, GA: CDC, National Center
for HIV, STD and TB Prevention, Division of HIV/AIDS.
Chatters, L. M., Taylor, R. J. & Lincoln, K. E. (1999). African American religious
participation: A multi-sample comparison. Journal for the Scientific Study of
Religion, 38, 132-145.
Chuang, C. H., Liebschutz, J. M., Horton, N. J. & Samet, J. H. (2003). Association
of physical and sexual abuse with high-risk sexual behavior in HIV-infected
persons with alcohol problems. Journal of General Internal Medicine 18,
142-143.
Ciambrone, D. (2002). Informal social networks among women with HIV/AIDS:
Present support and futuure prospects. Qualitative Health Research, 12, 870-
896.
190
Cokley, K. O. (2005). Racial(ized) identity, ethnic identity and Afrocentric values:
Conceptual and methodological challenges in understanding African
American identity. Journal of Counseling Psychology, 52, 517-526.
Coleman, C. L. (2004). The contribution of religious and existential well being to
depression among African American heterosexuals with HIV infection.
Issues in Mental Health Nursing, 25, 103-110.
Coleman, C. L. & Holzemer, W. L. (1999.) Spirituality, symptoms and psychological
well-being for African Americans with HIV infection. Journal of the
Association of Nurses in AIDS Care, 10, 42-50.
Coleman, C. L., Holzemer, W. L., Eller, L. S., Corless, I., Reynolds, N., Nokes, K.
M., Kemppainen, J. K., Dole, P., Kirksey, K., Seficik, L., Nicholas, P. &
Hamilton, M. J. (2006). Gender differences in use of prayer as a self care
strategy for managing symptoms in African Americans living with
HIV/AIDS. Journal of the Association of Nurses in AIDS Care, 17, 16-23.
Coleman, C. L. & Hummel, D. B. (2005). Sexual orientation a predictor of
depressive symptoms among HIV-infected African American men: A
descriptive correlational study. Archives of Psychiatric Nursing, 19, 236-241.
Conigliaro, J., Gordon, A. J., McGinnis, K. A., Rabenack, L. & Justice, A. C. (2003).
How harmful is hazardous alcohol use and abuse in HIV infection: Do health
care providers know who is at risk? Journal of Acquired Immune Deficiency
Syndromes, 33, 521-525.
Conigliaro, J., Maisto, S., McNeil, M., Kraemer, K., Conigliaro, R. & O’Connor, M.
(1998). How do black and white patients with alcohol problems differ?
Journal of General Internal Medicine, 13, 105.
Cook, C. H. L. (2004). Addiction and spirituality. Addiction, 99, 539-551.
Cook, R. L., Sereika, S. M., Hunt, S. C., Woodward, W. C., Erlen, J. A. and
Conigliaro, J. (2001). Problem drinking and medication adherence among
persons with HIV infection. Journal of General Internal Medicine 16, 83-88.
191
Cotton, S., Puchalski, C. M., Sherman, S. N., Mrus, J. M., Peterman, A. H.,
Feinberg, J., Pargament, K. I., Justice, A. C., Leonard, A. C. & Tsevat, J.
(2006a). Spirituality and religion in patients with HIV/AIDS. Journal of
General Internal Medicine, 21, S5-13.
Cotton, S., Tsevat, J., Szaflarski, M., Kudel, I., Sherman, S. N., Feinberg, J.,
Leonard, A. C. & Holmes, W. C. (2006b). Changes in religiousness and
spirituality attributed to HIV/AIDS: Are there sex and race differences?.
Journal of General Internal Medicine, 21, S14-20.
Crawford, I., Allison, K. W., Zamboni, B. D. & Soto, T. (2002). The influence of
dual identity development on the psychosocial functioning of African
American gay and bisexual men. The Journal of Sex Research, 39, 179-189.
Cunradi, C., Caetano, R., Clark, C. & Schaefer, J. (1999). Alcohol-related problems
and intimate partner violence among white, black and hispanic couples in the
U.S. Alcoholism: Clinical and Experimental Research, 23, 1492-1501.
Denton, M. & Walters, V. (1999). Gender differences in structural and behavioral
determinants of health: an analysis of the social production of health. Social
Science and Medicine, 48, 1221-1235.
Doyal, L. (2001). Sex, gender and health: the need for a new approach. British
Medical Journal, 323, 1061-1063.
Edwards, L. V. (2006). Perceived social support and HIV/AIDS medication
adherence among African American women. Qualitative Health Research,
16, 679-691.
Falkin, G. P. & Strauss, S. M. (2000). Drug-using women’s communications with
social supporters about HIV/AIDS issues. Journal of Drug Issues, 30, 801-
822.
Fein, G., Fletcher, D. J. & Di Scalfani, V. (1998). Effect of chronic alcohol abuse on
the CNS morbidity of HIV disease. Alcoholism: Clinical and Experimental
Research 22, 196S-200S.
192
First, M. B., Gibbon, M., Spitzer, R. L. & Williams, J. B. W. (2002). User’s Guide
for the SCID-I: Structured Clinical Interview for DSM-IV-TR Axis I
Disorders (Research Version). New York: Biometrics Research Department,
New York State Psychiatric Institute.
Flexner, C. W., Cargill, V. A., Sinclair, J., Kresina, T. F. & Cheever, L. (2001).
Alcohol use can result in enhanced drug metabolism in HIV
pharmacotherapy. AIDS Patient Care and STDs, 15, 57-58.
Fraser, D. (2000). QSR NVivo NUD*IST Vivo Reference Guide. Melbourne: QSR
International.
Galvan, F. H., Bing, E. G., Fleishman, J. A., London, A. S., Caetano, R., Burnam,
M. A., Longshore, D., Morton, S. C., Orlando, M. & Shapiro, M. (2002).
The prevalence of alcohol consumption and heavy drinking among people
with HIV in the United States: Results from the HIV Cost and Services
Utilization Study. Journal of Studies on Alcohol 63, 179-186.
Galvan, F. H., Burnam, A. B. & Bing, E. G. (2003). Co-ocurring psychiatric
symptoms and drug dependence or heavy drinking among HIV-positive
people. Journal of Psychoactive Drugs, SARC Supplement, 153-160.
Galvan, F. H., Collins, R. L., Kanouse, D. E., Pantoja, P. & Golinelli, D. (2007).
Religiosity, denominational affiliation and sexual behaviors among people
with HIV in the United States. Journal of Sex Research, 44, 49-58.
Gant, L. M. & Ostrow, D. G. (1995). Perceptions of social support and psychological
adaptation to sexually acquired HIV among white and African American
men. Social Work, 40, 215-224.
Gesensway, D. (2001). Reasons for sex-specific and gender-specific study of health
topics. Annals of Internal Medicine, 135, 935-938.
Glaser, B. G. & Strauss, A. L. (1967). The Discovery of Grounded Theory:
Strategies for qualitative research. New York: Aldine de Gruyter.
193
Green, C. A., Freeborn, D. K. & Polen, M. R. (2001). Gender and alcohol use: The
roles of social support, chronic illness and psychological well-being. Journal
of Behavioral Medicine, 24, 383-399.
Green, C. A., Perrin, N. & Polen, M. (2004). Gender differences in the relationships
between multiple measures of alcohol consumption and physical and mental
health. Alcoholism: Clinical and Experimental Research, 28, 754-764.
Green, J. E., Saveanu, R. V. & Bornstein, R. A. (2004). The effect of previous
alcohol abuse on cognitive functioning in HIV infection. The American
Journal of Psychiatry, 161, 249-254.
Griffin, M. L., Amodeo, M., Clay, C., Fassler, I. & Ellis, M. A. (2006). Racial
differences in social support: Kin versus friends. American Journal of
Orthopsychiatry, 76, 374-380.
Harper, F. D. (2001). Alcohol use and misuse. In R. L. Braithwaite & S. E. Taylor,
(Eds.), Health issues in the Black community. (pp. 403-418). San Francisco:
Jossey-Bass Publishers.
Hawe, P., Webster, C. & Shiell, A. (2006). A glossary of terms for navigating the
field of social network analysis. Journal of Epidemiology and Community
Health, 58, 971-975.
Heckman, T.G., Kochman, A., Sikkema, K.J., Kalichman, S.C., Masten, J. &
Goodkin, K. (2000). Late middle-aged and older men living with HIV/AIDS:
Race differences in coping, social support and psychological distress.
Journal of the National Medical Association, 92, 436-444.
Herd, D. (1997). Sex ratios of drinking patterns and problems among black and
whites: results from a national survey. Journal of Studies on Alcohol, 58, 75-
83.
Hicks, K., Allen, J. & Wright, E. (2005). Building holistic HIV/AIDS responses in
African-American urban faith communities: A qualitative multiple case study
analysis. Family Community Health, 28, 184-205.
194
The HIV Cost and Services Utilization Study. Retrieved from the World Wide Web:
http://www.rand.org/health/projects/chipts/projects/proj002.html.
Hough, E. S., Magnan, M. A., Templin, T. & Gadelrab, H. F. (2005). Social network
structure and social support in HIV-positive inner city mothers. Journal of
the Association of Nurses in AIDS Care, 16, 14-24.
Hudson, A. L., Lee, K. A., Miramontes, H. & Portillo, C. J. (2001). Social
interactions, perceived social support and level of distress in HIV-positive
women. The Journal of the Association of Nurses in AIDS Care, 12, 68-76.
Hurdle, D. E. (2001). Social support: A critical factor in women’s health and health
promotion. Health and Social Work, 26, 72-79.
Hurst, C., Montgomery, A. J., Davis, B. L., Killion, C. & Baker, S. (2005). The
relationship between social support, self care agency and self care practices
of African American women who are HIV-positive. The Journal of
Multicultural Nursing and Health, 11, 11-22.
Ironson, G., Stuetzle, R. & Fletcher, M. A. (2006). An increase in religiousness/
spirituality occurs after HIV diagnosis and predicts slower disease
progression over four years in people with HIV. Journal of General Internal
Medicine, 21, S62-68.
Isaki, L. & Kresina, T. F. (2000). Directions for biomedical research in alcohol and
HIV: Where are we now and where can we go? AIDS Research and Human
Retroviruses, 16, 1197-1207.
Jang, K., Livesley, W. J. & Vernon, P. (1997). Gender specific etiological
differences in alcohol and drug problems: a behavioral genetic analysis.
Addiction, 92, 1265-1276.
Jemmott, L. S., Jemmott, J. B. & Hutchinson, M. K. (2001). HIV/AIDS. In R. L.
Braithwaite & S. E. Taylor, (Eds.), Health issues in the Black community.
(pp. 403-418). San Francisco: Jossey-Bass Publishers.
195
Jipguep, M. C., Sanders-Phillips, K. & Cotton, L. (2004). Another look at HIV in
African American women: The impact of psychosocial and contextual
factors. Journal of Black Psychology, 30, 366-385.
Johnson, R. (2005). Gender differences in health-promoting lifestyles of African
Americans. Public Health and Nursing 22, 130-137.
Jones, D. J., Beach, S. R. H., Forehand, R. & Foster, S. E. (2003). Self-reported
health in HIV-positive African American women: The role of family stress
and depressive symptoms. Journal of Behavioral Health, 26, 577-599.
Kaskutas, L. A., Bond, J. & Humphreys, K. (2002). Social networks as mediators of
the effect of Alcoholics Anonymous. Addiction, 97, 891-900.
Kenny, D. A. (2006). Mediation. Retrieved from the World Wide Web:
http://davidakenny.net/cm/mediate.htm.
Kenny, D. A. (2004). Moderation. Retrieved from the World Wide Web:
http://davidakenny.net/cm/moderation.htm.
Knowlton, A. R., Hua, W. & Latkin, C. (2005). Social support networks and medical
service use among HIV-positive drug users: Implications to intervention.
AIDS Care, 17, 479-492.
Knowlton, A. R., Hua, W. & Latkin, C. (2004). Social support among HIV positive
injection drug users: Implications to integrated intervention for HIV
positives. AIDS and Behavior, 8, 357-363.
Lakey, B. & Cohen, S. (2000). Social support theory and measurement. In S. Cohen,
L. G. Underwood & B. H. Gottlieb (Eds.) Social Support Measurement and
Intervention: A Guide for Health and Social Scientists. (pp. 29-52). Oxford:
Oxford University Press.
196
Latkin, C. A., Tobin, K. E. & Gilbert, S. H. (2002). Shun or support: The role of
religious behaviors and HIV-related health care among drug users in
Baltimore, Maryland. AIDS and Behavior, 6, 321-3329.
Lefevre, F., O’Leary, B., Moran, M., Mossar, M., Yarnold, P. R., Martin, G. J. &
Glassroth, J. (1995). Alcohol consumption among HIV-infected patients.
Journal of General Internal Medicine, 10, 458-460.
Levant, R., Majors, R. & Kelley, M. (1998). Masculinity ideology among
young African American and European American women and men in
different regions of United States. Cultural Diversity and Mental Health,
4, 227-236.
Levin, J. S. & Taylor, R. J. (1993). Gender and age difference in religiosity among
Black Americans. The Gerontologist, 33, 16-23.
Lincoln, K.D., Chatters, L.M. & Taylor, R.J. (2005). Social support, traumatic
events and depressive symptoms among African Americans. Journal of
Marriage and Family, 67, 754-766.
Lewis, R. K., Redmond, M., Paschal, A. & Green, B. L. (2005). Health behavior
patterns of African Americans: Are we making progress toward the healthy
people 2010 goals? American Journal of Health Studies, 20, 149-155.
Liu, W. M. (2005). The study of men and masculinity as an important multicultural
competency consideration. Journal of Clinical Psychology, 61, 685-697.
Lorenz, K.A., Hays, R.D., Shapiro, M.F., Cleary, P.D., Asch, S.M. & Wenger, N.S.
(2005). Religiousness and spirituality among HIV-infected Americans.
Journal of Palliative Medicine, 8, 774-781.
Lorenz, K.A., Wenger, N., Hays, R.D., Cleary, P., Bozzette, S.A., Nederend, S. &
Shapiro, M.F. (1999). Description and predictors of religiosity and
spirituality in a nationally representative sample of patients with HIV
infection from the HIV cost and services utilization study (HCSUS).
Abstract Book of the Assn Health Services Resource Meeting, 16, 396-397.
197
MacKinnon, D. P., Lockwood, C. M., Hoffman, J. M., West, S. G. & Sheets, V.
(2002). A comparison of methods to test mediation and other intervening
variable effects. Psychological Methods, 7, 83-104.
Maisto, S. A., Carey, M. P., Carey, K. B., Gordon, C. M. & Schum, J. L. (2004).
Effects of alcohol and expectancies on HIV-related risk perception and
behavioral skills in heterosexual women. Experimental and Clinical
Psychopharmacology, 12, 288-297.
Malebranche, D. (2003). Black men who have sex with men and the HIV epidemic:
Next steps for public health. American Journal of Public Health, 93, 862-
864.
Mason, H., Simoni, J., Marks, G., Johnson, C. & Richardson, J. (1997). Missed
opportunities? Disclosure of HIV infection and support seeking among HIV+
African-American and European-American men. AIDS and Behavior, 3, 155-
162.
Mattis, J. S. & Jagers, R. J. (2001). A relational framework for the study of
religiosity and spirituality in the lives of African Americans. Journal of
Community Psychology, 29, 519-539.
McCallister, L. & Fischer, C. S. (1983). A procedure for surveying personal
networks. In R. S. Burt & M. J. Minor (Eds.), Applied Network Analysis: A
Methodological Introduction (pp. 75-88). Beverly Hills, CA: SAGE
Publications.
McNair, L. & Prather, C. (2004). African-American women and AIDS: Factors
influencing risk and reaction to HIV disease. Journal of Black Psychology,
30, 106-123.
Mellins, C. A., Ehrhardt, A. A., Rapkin, B. & Havens, J. F. (2000). Psychosocial
factors associated with adaptation in HIV-infected mothers. AIDS and
Behavior, 4, 317-328.
198
Midlife in the United States: A National Study of Health and Well Being. Retrieved
from the World Wide Web: http://midus.wisc.edu/midus1/.
Miller, W. L. & Crabtree, B. F. (1992). Primary care research: A multimethod
typology and qualitative road map. In Crabtree, B. F. & Miller, W. L. (eds.),
Doing Qualitative Research. (pp. 3-30). Newbury Park CA: Sage.
Miller, W. R. & Thorensen, C. E. Spirituality, religion and health: An emerging
research field. (2003). American Psychologist, 58, 24-35.
Minugh, P. A., Rice, C. & Young, L. (1998). Gender, health beliefs, health behaviors
and alcohol consumption. American Journal of Drug and Alcohol Abuse, 24,
483-497.
Morse, E., Morse, P., Klebba, K., Stock, M., Forehand, R. & Panayotova, E. (2000).
The use of religion among HIV-infected African American women. Journal
of Religion and Health, 39, 261-276.
Moynihan, C. (2006). Theories in health care and research: Theories of masculinity.
British Medical Journal, 317, 1072-1075.
Mrus, J. M., Williams, P. L., Tsevat, J., Cohn, S. E. & Wu, A. W. (2005). Gender
differences in health-related quality of life in patients with HIV/AIDS.
Quality of Life Research, 14, 479-491.
Muller, D., Judd, C. M. & Yzerbyt, V. Y. (2005). When moderation is mediated and
mediation is moderated. Journal of Personality and Social Psychology, 89,
852-863.
Myers, H. F. & Durvasula, R. S. (1999). Psychiatric disorders in African American
men and women living with HIV/AIDS. Cultural Diversity and Ethnic
Minority Psychology, 5, 249-262.
199
Obot, I. S. (1996). Problem drinking, chronic disease and recent life events. In H. W.
Neighbors & J. S. Jackson (Eds.), Mental health in Black America. (pp. 45-
61). Thousand Oaks, CA: SAGE Publications.
Ogunyemi, D., Ogunyemi, O., Ajayi, O. & Rodriguez, N. (2006). Psychosocial,
medical and gynecologic complications in a cohort of HIV-infected women
from an underserved minority community in Los Angeles. Journal of
Acquired Immune Deficiency Syndromes, 41, 255-257.
Orme, J. G. & Buehler, C. (2001). Introduction to multiple regression for categorical
and limited dependent variables. Social Work Research, 25, 49-61.
Owens, S. (2003). African American women living with HIV/AIDS: Families as
sources of support and of stress. Social Work, 48, 163-171.
Palepu, A., Horton, N., Tibbetts, N., Meli, S. & Samet, J. H. (2005). Substance abuse
treatment and hospitalization among a cohort of HIV-infected individuals
with alcohol problems. Alcoholism: Clinical and Experimental Research, 29,
389-394.
Palepu, A., Horton, N., Tibbetts, N., Meli, S. & Samet, J. H. (2003). Substance abuse
treatment and adherence to antiretroviral therapy among HIV-infected
persons with alcohol problems. Journal of General Internal Medicine, 18,
184-185.
Pargamet, K., McCarthy, S., Shah, P., Ano, G., Tarakeshwar, N., Wachholtz, A.,
Sirrine, N.,Vasconcelles, E., Murray-Swank, N., Locher, A. & Duggan, J.
(2004). Religion and HIV: A review of the literature and clinical
implications. Southern Medical Journal, 97, 1201-1209.
Parsons, J. T., Vicioso, K. L., Punzalan, J. C., Halkitis, P. N., Kutnick, A. &
Velasquez, M. M. (2004). The impact of alcohol use on the sexual scripts of
HIV-positive men who have sex with men. The Journal of Sex Research 41,
160-172.
200
Peirce, R. S., Frone, M. R., Russell, M., Cooper, M. L. & Mudar, P. (2000). A
longitudinal model of social contact, social support, depression and alcohol
use. Health Psychology, 19, 28-38.
Petry, N. M. (1999). Alcohol use in HIV patients: What we don’t know may hurt us.
International Journal of STD and AIDS, 10, 561-570.
Pierce, G. R., Sarason, I. G. & Sarason, B. R. (1995). Coping and social support. In
M. Zeidner & N. S. Endler, (Eds.), Handbook of Coping: Theory, Research,
Applications (pp. 434-451). New York: John Wiley & Sons, Inc.
Power, R., Koopman, C., Volk, J., Israelski, D. M., Stone, L., Chesney, M. A. &
Spiegel, D. (2003). Social support, substance use and denial in relationship to
antiretroviral treatment adherence among HIV-infected persons. AIDS
Patient Care and STDs, 15, 245-252.
Prado, G., Feaster, D., Schwartz, S., Pratt, I., Smith, L. & Szapocznik, J. (2004).
Religious involvement, coping, social support and psychological distress in
HIV-seropositive African-American mothers. AIDS and Behavior, 8, 221-
235.
Prado, G., Szapocznik, J., Mitrani, V., Mauer, M., Smith, L. & Feaster, D. (2002).
Factors influencing engagment into interventions for adaptation to HIV in
African American women. AIDS and Behavior, 6, 141-151.
Read, J. G. & Gorman, B. K. (2005). Gender inequalities in U.S. adult health: The
interplay of race and ethnicity. Social Science and Medicine, 62, 1045-1065.
Redelings, M., Frye, D. & Sorvillo, F. (2005). High incidence of HIV-associated
mortality among Black and Hispanic infants and women of childbearing age
in the United States 1990-2001. Journal of Acquired Immune Deficiency
Syndrome, 39, 496-498.
Redgrave, G., Swartz, A. & Romanski, A. (2003). Alcohol misuse by women.
International Reveiw of Psychiatry, 15, 256-268.
201
Ross, M. W., Essien, E. J. & Torres, I. (2006). Conspiracy beliefs about the origin of
HIV/AIDS in four racial/ethnic groups. Journal of Acquired Immune
Deficiency Syndrome, 41, 342-344.
Ryff, C. D., Singer, B. H., & Palmersheim, K. A. (2004). Social inequalities in
health and well-being: The role of relational and religious protective factors.
In O. G. Brim, C. D. Ryff & R. C. Kessler (Eds.), How healthy are we?: A
national study of well-being at midlife. (pp. 90 - 123). Chicago: University of
Chicago Press.
Samet, J. H., Horton, N. J., Meli, S., Freedberg, K. A. & Palepu, A. (2004). Alcohol
consumption and antiretroviral adherence among HIV-infected persons with
alcohol problems. Alcoholism: Clinical and Experimental Research, 28, 572-
577.
Samet, J. H., Horton, N. J., Traphagen, E. T., Lyon, S. M. & Freedberg, K. A.
(2003). Alcohol consumption and HIV disease progression: Are they related?
Alcoholism: Clinical and Experimental Research, 27, 862-867.
Sandelowski, M., Lambe, C. & Barroso, J. (2004). Stigma in HIV-positive women.
Journal of Nursing Scholarship, 36, 122-128.
Shambley-Ebron, D. Z. & Boyle, J. S. (2006). In our grandmothers’ footsteps:
Perceptions of being strong in African American women with HIV/AIDS.
Advances in Nursing Science, 29, 195-206.
Shear, M. K., Greeno, C., Kang, J., Ludewig, D., Frank, E., Swartz, H. A. &
Hanekamp, M. (2000). Diagnosis of nonpsychotic patients in community
clinics. American Journal of Psychiatry, 157, 581-587.
Shippy, R.A. & Karpiak, S. E. (2005). The aging HIV/AIDS population: Fragile
social networks. Aging and Mental Health, 9, 246-254.
Simoni, J. M., Demas, P., Mason, H. R. C., Drossman, J. & Davis, M. L. (2000).
HIV Disclosure among women of African descent: Associations with coping,
social support and psychological adaptation. AIDS and Behavior, 4, 147-158.
202
Simoni, J. M., Frick, P. A. & Huang, B. (2006). A longitudinal evaluation of a social
support model of medication adherence among HIV-positive men and women
on antiretroviral therapy. Health Psychology, 25, 74-81.
Simoni, J. M., Huang, B., Goodry, E. J. & Montoya, H. D. (2005). Social support
and depressive symptomatology among HIV-positive women: The mediating
role of self-esteem and mastery. Women & Health, 42, 1-15.
Simoni, J., Martone, M. & Kerwin, J. (2002). Spirituality and psychological
adaptation among women with HIV/AIDS: Implications for counseling.
Journal of Counseling Psychology, 49, 139-147.
Smith, K., Brutus, A., Cathcart, R., Gathe, J., Johnson, W., Jordan, W., Kwakwa, H.,
Nkwanyou, J., Page, C., Scott, R., Vaughn, W., Virgil, L. & Willaimson, D.
(2003). Optimizing care for African-American HIV-positive patients. AIDS
Patient Care and STDs, 17, 527-536.
Snowden, L. R. (1998). Racial differences in informal help seeking for mental health
problems. Journal of Community Psychology, 26, 429-438.
Strauss, A. & Corbin, J. (1998). Basics of qualitative research: Techniques and
procedures for developing grounded theory (second edition). Thousand Oaks,
CA: Sage Publications.
Steiner, J.L., Tebes, J. K., Sledge, W. H. & Walker, M. (1995). A comparison of the
structured clinical interview for DSM-III-R and clinical diagnoses. Journal of
Nervous & Mental Disease,183, 365-369.
Stewart, K. E., Cianfrini, L. R. & Walker, J. F. (2005). Stress, social support and
housing are related to health status among HIV-positive persons in the deep
south or the United States. AIDS Care, 17, 350-358.
Szaflarski, M., Ritchey, P. N., Leonard, A. C., Mrus, J. M., Peterman, A. H., Ellison,
C. G., McCullough, M. E. & Tsevat, J. (2006). Modeling the effects of
spirituality/religion on patients’ perceptions of living with HIV/AIDS.
Journal of General Internal Medicine, 21, S28-38.
203
Taylor, R. J., Hardison, C. B. & Chatters, L. M. (1996). Kin and nonkin as sources of
formal assistance. In H. W. Neighbors & J. S. Jackson (Eds.), Mental health
in Black America. (pp. 130-145). Thousand Oaks, CA: SAGE Publications.
Taylor, R. J., Mattis, J. & Chatters, L. M. (1999). Subjective religiosity among
African Americans: A synthesis of findings from five national samples.
Journal of Black Psychology, 25, 524-543.
Taylor, S. E. & Braithwaite, R. L. (2001). African American health: An overview. In
R. L. Braithwaite & S. E. Taylor, (Eds.), Health issues in the Black
community. (pp. 3-12). San Francisco: Jossey-Bass Publishers.
Trubo, R. (2004). CDC initiative targets HIV research gaps in black and hispanic
communities. Journal of the American Medical Association, 292, 2563-2564.
Turner, B. J., Laine, C., Cosler, L. & Hauck, W. W. (2003). Relationship of gender,
depression, and health care delivery with antiretroviral adherence in HIV-
infected drug users. Journal of General Internal Medicine, 18, 248-257.
Turner-Cobb, J. M., Gore-Felton, C., Marouf, F., Koopman, C., Kim, P., Israelski, D.
& Spiegel, D. (2002). Coping, social support and attachment style as
psychosocial correlates of adjustment in men and women with HIV/AIDS.
Journal of Behavioral Medicine, 25, 337-353.
Turner-Musa, J. O. & Wilson, S. A. (2006). Religious orientation and social support
on health-promoting behaviors of African American college students. Journal
of Community Psychology, 34, 105-115.
Walter, H., Guttierez, K., Ramskogler, K., Hertling, I., Dvorak, A. & Lesch, O. M.
(2003). Gender-specific differences in alcoholism: Implications for treatment.
Archives of Women’s Mental Health, 6, 253-258.
Valente, T.W., Gallagher, P. & Mouttapa, M. (2004). Using social networks to
understand and prevent substance use: A transdisciplinary perspective.
Substance Use & Misuse, 39, 1685-1712.
204
Wang, X., Douglas, S. D., Metzger, D. S., Guo, C., Li, Y., O’Brien, C. P., Song, L.,
Davis-Vogal, A. & Ho, W. (2002). Alcohol potentiates HIV-1 infection of
human blood mononuclear phagocytes. Alcoholism: Clinical and
Experimental Research, 26, 1880-1886.
Welch, K. J. (2000). Correlates of alcohol and/or drug use among HIV-infected
individuals. AIDS Patient Care and STDs, 14, 317-323.
Whetten, K., Reif, S., Lowe, K. & Eldred, L. (2004). Gender differences in
knowledge and perceptions of HIV resources among individuals living with
HIV in the southeast. Southern Medical Journal, 97, 342-349.
Whitehead, T. (1997). Urban low income African American men, HIV/AIDS and
gender identity. Medical Anthropology Quarterly, 11, 411-447.
Williams, P. B., Ekundayo, O., Udezulu, I. & Omishakin, A. M. (2003). An
ethnically sensitive and gender-specific HIV/AIDS assessment of African
American women.: A comparative study of urban and rural American
communities. Family & Community Health, 26, 108-123.
Williams, J. K., Wyatt, G. E., Resell, J., Peterson, J. & Asuan-O’Brien, A. (2004).
Psychosocial issues among gay and non gay identifying HIV-seropositive
African American and Latino msm. Cultural Diversity and Ethnic
Minority Psychology, 10, 268-286.
Willms, D. G., Best, A. J., Taylor, D. W., Gilbert, J. R., Wilson, D. M. C., Lindsay,
E. A. & Singer, J. (1992). A systematic approach for using qualitative
methods in primary prevention research. Medical Anthropology Quarterly 4,
391-409.
Wilsnack, R., Vogeltanz, N., Wilsnack, S. & Harris, R. (2000). Gender differences in
alcohol consumption and adverse drinking consequences: cross cultural
patterns. Addiction, 95, 251-265.
Wilson, M. N. (1986). The black extended family: An analytical consideration.
Developmental Psychology, 22, 246-258.
205
Witbrodt, J. & Kaskutas, L. A. (2005). Does diagnosis matter? Differential effects of
12-step participation and social networks on abstinence. The American
Journal of Drug and Alcohol Abuse, 31, 685-707.
Wohl, A. R., Johnson, D. F., Lu, S., Jordan, W., Beall, G., Currier, J. & Simon, P. A.
(2002). HIV risk behaviors among African American men in Los Angeles
county who self identify as heterosexual. Journal of Acquired Immune
Deficiency Syndromes, 31, 354-360.
Wohl, A. R., Lu, S., Odem, S., Sorvillo, F., Pegues, C. F. & Kerndt, P. R. (1998).
Sociodemographic and behavioral characteristics of African-American
women with HIV and AIDS in Los Angeles County, 1990-1997, Journal of
Acquired Immune Deficiency Syndromes and Human Retrovirology, 19, 412-
420.
Woodyard, J., Peterson, J. & Stokes, J. (2000). Let us go into the house of the Lord:
Participation in African American churches among young African American
men who have sex with men. Journal of Pastoral Care, 54, 451-460.
Wright, E. M. (2001). Substance abuse in African American communities. In S. L.
Ashenberg-Straussner, (Ed.), Ethnocultural factors in substance abuse
treatment. (pp 31-51). New York: The Guilford Press.
Yi, M. S., Mrus, J. M., Wade, T. J., Ho, M. L., Hornung, R. W., Cotton, S.,
Peterman, A. H., Puchalski, C. M. & Tsevat, J. (2006). Religion, spirituality,
and depressive symptoms in patients with HIV/AIDS. Journal of General
Internal Medicine, 21, S21-27.
Zanarini, M.C., Skodol, A.E., Bender, D., Dolan, R., Sanislow, C., Schaefer, E.,
Morey, L.C., Grilo, C.M., Shea, M.T., McGlashan, T.H. & Gunderson, J.G.
(2000). The Collaborative Longitudinal Personality Disorders Study:
reliability of axis I and II diagnoses. Journal of Personality Disorders, 14,
291-299
Zimet, G. D., Dahlem, N. W., Zimet, S. G & Farley, G. K. (1988). The
Multidimensional Scale of Perceived Social Support. Journal of Personality
Assessment, 52, 30-41.
206
Zimet, G. D., Powell, S. S., Farley, G. K., Werkman, S. & Berkoff, K. A. (1990).
Psychometric characteristics of the Multidimensional Scale of
Perceived Social Support. Journal of Personality Assessment, 55, 610-617.
Zinnbauer, B. J., Pargament, K. I., Cole, B., Rye, M., Butter, E. M., Belavich, T. G.,
Hipp, K. M., Scott, A. B. & Kadar, J. L. (1997). Religion and spirituality:
Unfuzzying the fuzzy. Journal for the Scientific Study of Religion, 36, 549-
564.
Zule, W., Flannery, B., Wechsberg, W. & Lam, W. (2002). Alcohol use among out-
of-treatment crack using African-American women. American Journal of
Drug and Alcohol Abuse, 28, 525-544.
207
APPENDIX A: SCID-I (for DSM-IV-TR)
Alcohol Use Screening (NOV 2002) Substance Use Disorders E. 207
E. SUBSTANCE USE DISORDERS
ALCOHOL USE DISORDERS (LIFETIME) SCREEN Q#1
YES NO E1a
IF SCREENING QUESTION #1 ANSWERED "NO,"
CHECK HERE ___
AND SKIP TO *NON-ALCOHOL SUBSTANCE USE DISORDERS,* E. 10
(IF NO: GO TO *NON-ALCOHOL
USE DISORDERS* E.9)
IF SCREENER NOT USED, OR IF QUESTION #1 IS ANSWERED "YES,"
CONTINUE:
What are your drinking habits like?
(How much do you drink?) (Has there
ever been a time in your life when you
had five or more drinks on one occasion?)
When in your life were you drinking the RECORD DATE OF HEAVIEST
most? (How long did that period last?) USE AND DESCRIBE PATTERN:
___________________________________ E1b
___________________________________ E1c
During that time . . .
how often were you drinking?
what were you drinking? how much?
During that time . . .
did your drinking cause problems for you?
did anyone object to your drinking?
IF ALCOHOL DEPENDENCE SEEMS LIKELY, E1d
CHECK HERE ___ AND SKIP TO *ALCOHOL
DEPENDENCE,* E. 4.
IF ANY INCIDENTS OF EXCESSIVE DRINKING OR
ANY EVIDENCE OF ALCOHOL-RELATED PROBLEMS, CONTINUE WITH
*ALCOHOL ABUSE,* ON NEXT PAGE.
IF NEVER HAD ANY INCIDENTS OF EXCESSIVE DRINKING AND
THERE IS NO EVIDENCE OF ANY ALCOHOL-RELATED PROBLEMS,
SKIP TO *NON-ALCOHOL SUBSTANCE USE DISORDERS,* E. 10.
208
?=Inadequate information, 1=Absent/False, 2=Sub-Threshold, 3=Threshold /True
*LIFETIME ALCOHOL ABUSE* ALCOHOL ABUSE
CRITERIA
Let me ask you a few more questions
about (TIME WHEN DRINKING
THE MOST OR TIME WHEN
DRINKING CAUSED MOST
PROBLEMS). During that time . . .
A. A maladaptive pattern of
alcohol use, leading to
clinically significant
impairment or distress, as
manifested by three (or more)
of the following occurring
within a twelve month period:
Have you ever missed work or school
because you were intoxicated, high,
or very hung over? (How often? What
about doing a bad job at work or
failing courses at school because of
your drinking?)
IF NO: What about not keeping
your house clean or not taking
proper care of your children
because of your drinking? (How
often?)
IF YES TO EITHER OF ABOVE:
How often? (Over what period of
time?)
(1) Recurrent alcohol use
resulting in a failure to fulfill
major role obligations at
work, school, or home (e.g.,
repeated absences or poor
work performance related to
alcohol use; alcohol-related
absences, suspensions, or
expulsions from school;
neglect of children or
household).
? 1 2 3 E2
Did you ever drink in a situation in
which it might have been dangerous
to drink at all? (Did you ever drive
while you were really too drunk to
drive?)
IF YES AND UNKNOWN: How
many times? (When?)
(2) recurrent alcohol use in
situations in which it is
physically hazardous (e.g.,
driving an automobile or
operating a machine when
impaired by alcohol use)
? 1 2 3 E3
Has your drinking gotten you into
trouble with the law?
IF YES AND UNKNOWN: How
often? (Over what period of time?)
(3) recurrent alcohol-related
legal problems (e.g., arrests
for alcohol-related disorderly
conduct)
? 1 2 3
E4
IF NOT ALREADY KNOWN: Has
your drinking caused problems with
other people, such as with family
members, friends, or people at work?
(Have you ever gotten into physical
fights when you were drinking? What
about having bad arguments about
your drinking?
IF YES: Did you keep on drinking
anyway? (Over what period of
time?)
(4) continued substance use
despite having persistent or
recurrent social or
interpersonal problems
caused or exacerbated by the
effects of the substance (e.g.,
arguments with spouse about
consequences of intoxication,
physical fights)
? 1 2 3
E5
209
AT LEAST ONE "A" ITEM CODED “3” 1 3 E6
(IF “1”)
IF ALCOHOL DEPENDENCE QUESTIONS HAVE
ALREADY BEEN ASKED (I.E., DEPENDENCE SEEMED
LIKELY AFTER ALCOHOL SCREENING ON E. 1
BUT FULL CRITERIA WERE NOT MET),
GO TO *NON-ALCOHOL USE DISORDERS, * E. 9.
IF ALCOHOL DEPENDENCE QUESTIONS HAVE
NOT YET BEEN EVALUATED AND THERE IS
ANY POSSIBILITY OF PHYSIOLOGICAL DEPENDENCE
OR COMPULSIVE USE,
GO TO *ALCOHOL DEPENDENCE,* ON PAGE E. 4.
OTHERWISE,
GO TO *NON-ALCOHOL USE DISORDERS,* E. 9.
(IF “3”)
ALCOHOL
ABUSE
IF ALCOHOL DEPENDENCE QUESTIONS HAVE ALREADY BEEN ASKED
(I.E., DEPENDENCE SEEMED LIKELY AFTER ALCOHOL SCREENING ON E.
1, BUT FULL CRITERIA WERE NOT MET),
GO TO *ALCOHOL ABUSE CHRONOLOGY,* E. 6.
IF ALCOHOL DEPENDENCE QUESTIONS HAVE NOT YET BEEN
EVALUATED,
CONTINUE WITH *ALCOHOL DEPENDENCE,* ON PAGE E. 4.
210
?=Inadequate information, 1=Absent/False, 2=Sub-Threshold, 3=Threshold/True
ALCOHOL DEPENDENCE ALCOHOL DEPENDENCE
CRITERIA
I'd now like to ask you some
more questions about (TIME
WHEN DRINKING THE
MOST OR TIME WHEN
DRINKING CAUSED MOST
PROBLEMS). During that
time . . .
A maladaptive pattern of alcohol use,
leading to clinically significant
impairment or distress, as manifested
by three (or more) of the following
occurring at any time in the same
twelve month period:
NOTE: CRITERIA FOR ALCOHOL
DEPENDENCE ARE NOT IN DSM-
IV-TR ORDER
Have you often found that
when you started drinking you
ended up drinking much more
than you were planning to?
IF NO: What about
drinking for a much longer
period of time than you
were planning to?
(3) alcohol is often taken in larger
amounts OR over a longer period than
was intended
? 1 2 3
E7
Have you tried to cut down or
stop drinking alcohol?
IF YES: Did you ever
actually stop drinking
altogether?
(How many times did you
try to cut down or stop
altogether?)
IF NO: Did you want to
stop or cut down? (Is this
something you kept
worrying about?)
(4) there is a persistent desire OR
unsuccessful efforts to cut down or
control alcohol use
? 1 2 3
E8
Have you spent a lot of time
drinking, being high, or hung
over?
(5) a great deal of time is spent in
activities necessary to obtain alcohol,
use alcohol, or recover from its
effects
? 1 2 3
E9
Have you had times when you
would drink so often that you
started to drink instead of
working or spending time at
hobbies or with your family or
friends, or engaging in other
important activities, such as
sports, gardening, or playing
music?
(6) important social, occupational, or
recreational activities given up or
reduced because of alcohol use
? 1 2 3
E10
211
IF NOT ALREADY KNOWN:
Has your drinking ever caused
any psychological problems
like making you depressed or
anxious, making it difficult to
sleep, or causing "blackouts?
IF NOT ALREADY KNOWN:
Has your drinking caused
significant physical problems
or made a physical problem
worse?
IF YES TO EITHER OF
ABOVE: Did you keep on
drinking anyway?
(7) alcohol use is continued despite
knowledge of having a persistent or
recurrent physical or psychological
problem that is likely to have been
caused or exacerbated by alcohol
(e.g., continued drinking despite
recognition that an ulcer was made
worse by alcohol consumption)
? 1 2 3
E11
Have you found that you
needed to drink a lot more in
order to get the feeling you
wanted than you did when you
first started drinking?
IF YES: How much more?
IF NO: What about finding
that when you drank the
same amount, it had much
less effect than before?
(1) tolerance, as defined by either of
the following:
(a) a need for markedly increased
amounts of alcohol to achieve
intoxication or desired effect
(b) markedly diminished effect
with continued use of the same
amount of alcohol
? 1 2 3
E12
Did you ever have any
withdrawal symptoms when
you cut down or stopped
drinking like . . .
(2) withdrawal, as manifested by
either (a) or (b):
(a) at least TWO of the following:
? 1 2 3
E13
. . . sweating or racing heart?
. . . hand shakes?
. . . trouble sleeping?
. . . feeling nauseated or
vomiting?
. . . feeling agitated?
. . . or feeling anxious?
-- autonomic hyperactivity (e.g.,
sweating or pulse rate > 100)
-- increased hand tremor
-- insomnia
-- nausea or vomiting
-- psychomotor agitation
-- anxiety
(How about having a seizure or
seeing, feeling, or hearing
things that weren't really
there?)
-- grand mal seizures
-- transient visual, tactile, or auditory
hallucinations or illusions
IF NO: Have you ever started
the day with a drink, or did
you often drink or take some
other drug or medication to
keep yourself from getting the
shakes or becoming sick?
(b) alcohol (or a substance from the
sedative / hypnotic/ anxiolytic class)
taken to relieve or avoid withdrawal
symptoms
E14
212
IF UNKNOWN: When did
(SXS CODED 3" ABOVE)
occur? (Did they all happen
around the same time?)
AT LEAST THREE
DEPENDENCE ITEMS
CODED "3" AND ITEMS
OCCURRED WITHIN
THE SAME TWELVE
MONTH PERIOD
1 3
E15
(IF “3” ALCOHOL DEPENDENCE, GO TO *CHRONOLOGY* E.7)
(IF “1”)
IF ALCOHOL ABUSE QUESTIONS (PAGES E:1-E.3)
HAVE NOT YET BEEN ASKED,
GO TO PAGE E.2 AND CHECK FOR ABUSE.
IF ABUSE QUESTIONS HAVE BEEN ASKED AND
ABUSE IS PRESENT, CODE "3."
OTHERWISE, IF QUESTIONS HAVE BEEN ASKED AND
ABUSE IS NOT PRESENT,
GO TO *NON-ALCOHOL USE DISORDERS,* E. 9
1 3
E16
(IF “1” GO TO *NON- ALCOHOL USE ABUSE DISORDER,* E. 9)
(IF “3” ALCOHOL ABUSE)
*ALCOHOL ABUSE
CHRONOLOGY*
How old were you when you
first had (ABUSE SXS
CODED "3")?
Age at onset of Alcohol
Abuse
(CODE 99 IF UNKNOWN)
___ ___
E17
IF UNCLEAR: During the
past month, have you had
anything at all to drink?
IF YES: Tell me more
about it.
(Has your drinking
caused you any
problems?)
Criteria for Alcohol Abuse
met at any time in past month
1 3
E18
(IF “1” PAST ABUSE, GO TO *NON-ALCOHOL USE DISORDER,*E. 10)
(IF “3” CURRENT ABUSE, GO TO *NON-ALCOHOL USE DISORDER,*E. 10)
213
*CHRONOLOGY FOR
DEPENDENCE*
How old were you when you
first had (LIST OF ALCOHOL
DEPENDENCE OR ABUSE
SXS CODED "3")?
Age at onset of Alcohol
Dependence or Abuse
(CODE 99 IF
UNKNOWN)
___ ___
E19
IF UNCLEAR: During the past
month, have you had anything
at all to drink?
IF YES: Tell me more about
it.
(Has your drinking caused
you any problems?)
Full criteria for Alcohol
Dependence met at any
time in past month (or
never had a month without
symptoms of Dependence
or Abuse since last onset of
Dependence)
? 1 3
E20
(IF “1”, GO TO *REMISSION SPECIFIERS* E. 8)
(IF “3”, CURRENT DEPENDENCE)
Indicate if:
1 - With Physiological Dependence (current evidence of tolerance or
withdrawal)
2 - Without Physiological Dependence (no current evidence of tolerance
or withdrawal)
NOTE SEVERITY OF DEPENDENCE FOR WORST WEEK OF PAST
MONTH (Additional questions about the effect of alcohol on social and
occupational functioning may be necessary.)
1 Mild: Few, if any, symptoms in excess of those required to make
the diagnosis, and the symptoms result in no more than mild
impairment in occupational functioning or in usual social
activities or relationships with others (or criteria met for
Dependence in the past and some current problems).
2 Moderate: Symptoms or functional impairment between "mild" and
"severe."
3 Severe: Many symptoms in excess of those required to make the
diagnosis, and the symptoms markedly interfere with
occupational functioning or with usual social activities or
relationships with others.
E22
GO TO NON-ALCOHOL USE DISORDERS, E. 9
214
*REMISSION SPECIFIERS FOR DEPENDENCE*
THE FOLLOWING REMISSION SPECIFIERS CAN BE APPLIED ONLY AFTER
NO CRITERIA FOR DEPENDENCE OR ABUSE HAVE BEEN MET FOR AT
LEAST ONE MONTH IN THE PAST.
Note: These specifiers do not apply if the individual is
On Agonist Therapy or In a Controlled Environment (below).
Number of months prior to interview when last had some
problems with Alcohol __ __ __
E23
E24
2 Early Full Remission: For at least one month, but less than twelve months,
no criteria for Dependence or Abuse have been met
______________________________________________________
Dependence / 1 month / 0-11 months
2 Early Partial Remission: For at least one month, but less than twelve months,
one or more criteria for Dependence or Abuse have been met
(but the full criteria for Dependence have not been met).
______________________________________________________
Dependence / 1 month / 0-11 months
3 Sustained Full Remission: None of the criteria for Dependence or Abuse
have been met at any time during a period of twelve months or longer.
______________________________________________________
Dependence / 1 month / 11+ months
4 Sustained Partial Remission: Full criteria for Dependence have not been met
for a period of twelve months or longer; however, one or more criteria for
Dependence or Abuse have been met.
______________________________________________________
Dependence / 1 month / 11+ months
Check ____ if In a Controlled Environment: The individual is in an
environment where access to alcohol and controlled substances is restricted
and no criteria for Dependence or Abuse have been met for at least the past
month. Examples are closely-supervised and substance-free jails,
therapeutic communities, and locked hospital units.
E26
215
APPENDIX B: HCSUS Religiosity & Spirituality Section
These questions are about being religious and being spiritual. Please think about
what these two words mean to you and answer the question with those meanings in
mind.
1. Please look at this list and tell me what your personal religious preference is:
1 Apostolic 38 Salvation Army
2 Assembly of God 39 Sanctified
3 Baptist (all types) 40 Seventh Day Adventist
4 Born-Again Christian 41 Spiritual
5 Brethren 42 Unitarian
6 Buddhist (all types, including Zen) 43 United Church of Christ
7 Disciples of Christ/Christian Church 44 Agnostic or Atheist
8 Catholic, Roman 45 No religious preference
9 Catholic, Ukrainian 46 Other
10 Catholic (all others)
11 Christian Reformed
12 Church of God
13 Congregational
14 Episcopalian or Anglican/Church of England
15 Evangelical
16 Hindu
17 Holiness
18 Jehovah’s Wittness
19 Jewish (Orthodox)
20 Jewish (Conservative)
21 Jewish (Reform)
22 Jewish (Reconstructionist)
23 Jewish (all others)
24 Lutheran
25 Mennonite
26 Methodist (all types, including United Brethren)
27 Mormon, Latter Day Saints
28 Muslim
29 Nazarene
30 Orthodox (Russian, Greek, Serbian)
31 Pentecostal
32 Presbyterian
33 Protestant, Interdenominational (you go to two or more Protestant Churches)
34 Protestant, no denomination
35 Protestant, other
36 Quaker, Society of Friends
37 Rastafarian
216
(Ask Question 2. only if response to Question 1. is 44, 45 or 46:)
2. Do you consider yourself a Christian?
1 Yes
2 No
3. How religious are you? Would you say:
1 Very
2 Somewhat
3 Not very
4 Not at all
4. How spiritual are you? Would you say:
1 Very
2 Somewhat
3 Not very
4 Not at all
5. How important is religion in your life? Would you say:
1 Very
2 Somewhat
3 Not very
4 Not at all
6. How important is spirituality in your life? Would you say:
1 Very
2 Somewhat
3 Not very
4 Not at all
7. How closely do you identify with being a member of your religious group?
Would you say:
1 Very
2 Somewhat
3 Not very
4 Not at all
5 Does not apply
217
8. How much do you prefer to be with other people who are the same religion that
you are? Would you say:
1 Very
2 Somewhat
3 Not very
4 Not at all
9. Which of the following do you believe: That it’s good to explore many different
religious or spiritual teachings, or that one should stick to one particular faith?
1 Explore different teachings
2 Stick to one faith
3 Neither
10. How often do you usually attend religious or spiritual services? Would you say:
1 More than once per week
2 About once per week
3 One to three times per month
4 Less than once per month
5 Never
11. When you have problems or difficulties in your family, work or personal life,
how often do you seek comfort through religious or spiritual means such as
praying, meditating, attending a religious or spiritual service or talking to a
religious or spiritual advisor? Would you say:
1 Often
2 Sometimes
3 Rarely
4 Never
12. When you have decisions to make in your life, how often do you ask yourself
what your religious or spiritual beliefs suggest you should do? Would you say:
1 Often
2 Sometimes
3 Rarely
4 Never
218
APPENDIX C: Multidimensional Scale of Perceived Social Support
Instructions: We are interested in how you feel about the following statements. Read each statement
carefully. Indicate how you feel about each statement.
Circle the “1” if you Very Strongly Disagree
Circle the “2” if you Strongly Disagree
Circle the “3” if you Mildly Disagree
Circle the “4” if you are Neutral
Circle the “5” if you Mildly Agree
Circle the “6” if you Strongly Agree
Circle the “7” if you Very Strongly Agree
Very Very
Strongly Strongly Mildly Mildly Strongly Strongly
Disagree Disagree Disagree Neutral Agree Agree Agree
1. There is a special person who
is around when I am in need. 1 2 3 4 5 6 7
2. There is a special person with whom
I can share joys and sorrows. 1 2 3 4 5 6 7
3. My family really tries to help me. 1 2 3 4 5 6 7
4. I get the emotional help & support
I need from my family. 1 2 3 4 5 6 7
5. I have a special person who is
a real source of comfort to me. 1 2 3 4 5 6 7
6. My friends really try to help me. 1 2 3 4 5 6 7
7. I can count on my friends when
things go wrong. 1 2 3 4 5 6 7
8. I can talk about my problems with
my family. 1 2 3 4 5 6 7
9. I have friends with whom I can
share my joys and sorrows. 1 2 3 4 5 6 7
10. There is a special person in my
life who cares about my feelings. 1 2 3 4 5 6 7
11. My family is willing to help me
make decisions. 1 2 3 4 5 6 7
12. I can talk about my problems with
my friends. 1 2 3 4 5 6 7
219
APPENDIX D: Social Network Questions
We would now like to ask you about some of the people in your life. Please name all of the
friends and family members, including spouses, partners or lovers that you see or talk on the
phone with once or more within a typical two-week period.
1. __________________________________________
2. __________________________________________
3. __________________________________________
4. __________________________________________
5. __________________________________________
6. __________________________________________
7. __________________________________________
8. __________________________________________
9. __________________________________________
10. __________________________________________
Name of Person Number 1:
_______________________________________________
1. What is this person’s gender? ___ Male ___ Female ___ Transgender/Other
2. What is this person’s age? __________ years
3. What is this person’s racial or ethnic identity?
a. _____ non-Latino White b. _____ African American
c. _____ Latino d. _____ Asian/Pacific Islander
e. _____ American Indian/Alaskan Native/Native American
f. _____ Multiethnic/multiracial
g. _____ Other (please describe: ______________________________)
220
4. What is your relationship to this person?
a. _____ spouse b. _____ partner c. _____ significant other
d. _____ other family member (please describe: ________________________)
e. _____ friend f. _____ neighbor g. _____ work colleague
h. _____ school colleague i. _____ fellow church member
j. _____ other (please describe: _________________________________)
5. How did you meet this person?
____________________________________________________________________
____________________________________________________________________
6. How long have you known this person? ______________________________
(Indicate in appropriate units: years/months/weeks/days)
7. Does this person live in the same city that you do?
_____ Yes _____ No
8. Does this person live within a five-minute drive from you?
_____ Yes _____ No
9. Do you feel especially close to this person?
_____ Yes _____ No
10. How often do you and this person “get together?”
____________________________________________________________________
11. Is this person a heavy or problem drinker?
_____ Yes _____ No
12. Does this person encourage you to reduce or stop your drinking?
_____ Yes _____ No
221
Name of Person Number _____ :
__________________________________________
1. What is this person’s gender? ___ Male ___ Female ___ Transgender/Other
2. What is this person’s age? __________ years
3. What is this person’s racial or ethnic identity?
a. _____ non-Latino White b. _____ African American
c. _____ Latino d. _____ Asian/Pacific Islander
e. _____ American Indian/Alaskan Native/Native American
f. _____ Multiethnic/multiracial
g. _____ Other (please describe: ______________________________)
4. What is your relationship to this person?
a. _____ spouse b. _____ partner c. _____ significant other
d. _____ other family member (please describe: ________________________)
e. _____ friend f. _____ neighbor g. _____ work colleague
h. _____ school colleague i. _____ fellow church member
j. _____ other (please describe: _________________________________)
5. How did you meet this person?
____________________________________________________________________
____________________________________________________________________
____________________________________________________________________
6. How long have you known this person? ______________________________
(Indicate in appropriate units: years/months/weeks/days)
222
7. Does this person live in the same city that you do?
_____ Yes _____ No
8. Does this person live within a five-minute drive from you?
_____ Yes _____ No
9. Do you feel especially close to this person?
_____ Yes _____ No
10. How often do you and this person “get together?”
____________________________________________________________________
____________________________________________________________________
11. Is this person a heavy or problem drinker?
_____ Yes _____ No
12. Does this person encourage you to reduce or stop your drinking?
_____ Yes _____ No
223
APPENDIX E: Open Ended Interview Questions
1. As a person living with HIV, what kinds of things do you do to take care of
your health and well-being?
2. Also as someone living with HIV, are there things that make it harder for you
to take care of your health and well-being? What kinds of barriers do you
encounter?
In what ways do they make things harder?
3. Since you were diagnosed with HIV, has your use of alcohol ever made it
harder for you to take care of your health and well-being?
If so, how? Can you give a few examples?
4. Looking at the ways that alcohol has affected you, would you say that its
impact on your well-being and self care has changed over time? What would
you say caused those changes?
5. What do you think might prevent your alcohol use from affecting how you
take care of your health and well-being?
6. Do you have any other thoughts on this issue that you’d like to share?

Asset Metadata

Creator Davis, E. Maxwell (author)

Core Title Gendered issues in alcohol abuse and dependence among HIV-positive African Americans

School School of Social Work

Degree Doctor of Philosophy

Degree Program Social Work

Publication Date 10/17/2007

Defense Date 08/21/2007

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag African Americans,alcohol abuse,alcohol dependence,gender,HIV/AIDS,OAI-PMH Harvest,social support

Place Name California (states), Los Angeles (counties), USA (countries)

Language English

Advisor Land, Helen M. (committee chair), Lutkehaus, Nancy (committee member), Palinkas, Lawrence A. (committee member)

Creator Email emdavis@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m871

Unique identifier UC1289133

Identifier etd-Davis-20071017 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-584288 (legacy record id),usctheses-m871 (legacy record id)

Legacy Identifier etd-Davis-20071017.pdf

Dmrecord 584288

Document Type Dissertation

Rights Davis, E. Maxwell

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Repository Name Libraries, University of Southern California

Repository Location Los Angeles, California

Repository Email uscdl@usc.edu

Abstract (if available)

Abstract African Americans are disproportionately impacted by HIV/AIDS and by many psychosocial issues that complicate their experiences living with this disease. Of these issues, alcohol abuse and dependence has been linked to accelerated HIV disease progression, reduced efficacy of HIV medications, poor adherence to medical regimens and risky sexual behaviors. This study uses the secondary analysis of quantitative and qualitative data collected from 272 HIV-positive African Americans in Los Angeles County to investigate this phenomenon. Analysis revealed that 13.6% of participants met the criteria for current alcohol abuse or dependence. Logistic regressions testing the influences of gender, religiosity, spirituality, social support and social network composition on current alcohol abuse and dependence revealed that social support was uniquely influential. Content analysis of the narratives of forty consumers with histories of alcohol abuse or dependence examined factors influencing both the exacerbation of and recovery from alcohol problems in the context of HIV/AIDS and the impact of these problems on HIV self care. These findings highlighted the influence of religiosity and spirituality but, again emphasized the centrality of social support. The qualitative findings also provided a unique opportunity to interpret the quantitative findings in relation to consumer perspectives and experiences. Discussion of these findings includes recommendations for using the influence of social support to lessen the impact of alcohol misuse on HIV treatment and prevention efforts in this heavily impacted community.