Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2014 Sep 6.
Published in final edited form as: AIDS Care. 2014 Jan 13;26(8):959–967. doi: 10.1080/09540121.2013.873765

Gender moderates the influence of psychosocial factors and drug use on HAART adherence in the context of HIV and childhood sexual abuse

Sarah M Wilson a,b, Kathleen J Sikkema a,b, Krista W Ranby c
PMCID: PMC4156823  NIHMSID: NIHMS622183  PMID: 24410324

Abstract

This study aimed to examine gender moderation within a stress and coping model of HIV medication adherence in adults with a history of childhood sexual abuse (CSA). Sequelae of CSA, including negative coping, psychological distress, and drug use, interfere with adherence to highly active antiretroviral treatment (HAART). These obstacles to adherence are likely moderated by gender. Gender may particularly influence the mediational effect of drug use on adherence. Participants included 206 adults living with HIV/AIDS and CSA. Categorical/continuous variable methodology (CVM) in a structural equation modeling (SEM) framework was used to test a multigroup model with women and men. Gender significantly moderated several effects in the model. For women, the effect of psychological distress on HAART adherence was mediated by drug use, and the effect of drug use on viral load was mediated by HAART adherence. Among men, drug use did not significantly impact adherence. Since gender appears to moderate the effect of drug use on medication adherence, it is particularly important to address drug use within the context of HIV disease management in women with a history of CSA. Further, interventions to increase HAART adherence should take trauma history, gender, and drug abuse into account when assessing efficacy.

Keywords: medication adherence, CSA = childhood sexual abuse, HIV = human immunodeficiency virus, drug abuse, SEM = structural equation modeling, gender difference

Introduction

HIV is a chronic illness that threatens the health of over 1 million individuals living in the United States (CDC, 2012). Among people living with HIV/AIDS, approximately 33% to 53% report a history of childhood sexual abuse (CSA) (Henny, Kidder, Stall, & Wolitski, 2007; Kalichman, Sikkema, DiFonzo, Luke, & Austin, 2002; Machtinger, Wilson, Haberer, & Weiss, 2012; Welles et al., 2009; Whetten et al., 2006). HIV-positive adults with CSA suffer from psychological sequelae that interfere with medication adherence, which in turn affects HIV disease progression (Cohen, Alfonso, Hoffman, Milau, & Carrera, 2001; Markowitz et al., 2011; Meade, Hansen, Kochman, & Sikkema, 2009). For HIV-positive individuals with CSA, unique factors leading to reduced HAART adherence include drug use, poor social support, and traumatic stress symptoms (Liu et al., 2006; Meade et al., 2009). While emerging research has focused on predictors of poor medication adherence in this population, there is a gap in empirical study of mechanisms and moderators of influences on adherence.

Drug use/abuse may represent one mechanism by which a history of CSA results in low HAART adherence. Drug and alcohol use significantly and uniquely predict poorer HAART adherence in a variety of populations (Golin et al., 2002; Hendershot, Stoner, Pantalone, & Simoni, 2009; Kreitchmann et al., 2012; Marks King et al., 2012). Accelerated HIV disease progression in drug users may be mediated by poor health maintenance behaviors, including HAART adherence (Milloy et al., 2012). Given that drug use/abuse is common in those who have experienced CSA, the impact of regular drug use on medication adherence should be examined in this population (Felitti et al., 1998; Molnar, Buka, & Kessler, 2001).

There is evidence to suggest that mediational pathways influencing HAART adherence may differ by gender. Overall, HIV-positive women demonstrate lower HAART adherence compared to their male counterparts (Berg et al., 2004; Tapp et al., 2011). Differences in mediational pathways may explain this gender disparity. For example, in a sample of older adults, depression mediated the relationship between negative coping and adherence for men only (Bianco, Heckman, Sutton, Watakakosol, & Lovejoy, 2011). Additionally, gender moderates the effect of alcohol use on HAART adherence: problematic alcohol use was associated with lower HAART adherence for women only (Berg et al., 2004). Given that CSA history is similarly prevalent in both men and women with HIV (Kalichman et al., 2002), gender effects should be evaluated in the context of CSA sequelae, such as negative coping, psychological distress, and drug use. For individuals with CSA, gender has yet to be examined as a moderating influence on HAART adherence, especially via variables such as depression, PTSD, and drug use.

The current study sought to test and expand upon an empirically-supported model of HAART adherence. Based on a theoretical foundation in stress and coping, Johnson and colleagues (2009) found that the effects of social support and negative coping on HAART adherence were fully mediated by negative affect. This model, which incorporated psychosocial and behavioral influences of HAART adherence, is similar to other theory-driven models (Gonzalez et al., 2007; Simoni, Frick, & Huang, 2006; Weaver et al., 2005). Utilizing a similar theoretical basis, the hypothesized model incorporated aspects of Lazarus and Folkman’s stress and coping theory (1984) within a model of behavioral mediation of HIV disease progression (Gore-Felton & Koopman, 2008). According to the theory of stress and coping, psychological symptoms are impacted by life stressors, coping strategies, and cognitive appraisals of stressors. The theoretical model of HIV disease progression hypothesizes that behavioral mechanisms, such as drug abuse and poor medication adherence, partially mediate the relationship between psychosocial factors and HIV disease progression (Gore-Felton & Koopman, 2008). In keep with these theoretical bases, we included gender as a moderator to examine how effects on HAART adherence differed for men and women.

In the current study, gender moderation was explored within a structural equation model of HAART adherence for HIV-positive adults with a history of CSA. Research goals were to: 1) explore the statistical fit of the negative affect mediation model for a sample of HIV-positive adults with CSA, 2) test whether drug use mediates the effect of psychological distress on HAART adherence, and 3) test whether model paths were moderated by gender. We hypothesized that gender would significantly moderate multiple mechanisms of HAART adherence. Specifically, we predicted that for women, drug use would have a stronger mediational effect on the relationship between psychological distress and adherence.

Method

Design

HIV-positive adults with CSA histories were recruited between March 2002 and January 2004 in New York City to participate in an intervention study on coping with HIV and sexual abuse, Living in the Face of Trauma (LIFT) (Sikkema et al., 2012).

Participants

All participants were recruited from AIDS service organizations and community health clinics in New York City between March 2002 and January 2004. Inclusion criteria were a) sexual abuse as a child and/or adolescent, defined as any unwanted touching of a sexual nature by an adult or by someone at least 5 years older than the participant when the incident occurred, b) current age 18 years or older, and c) HIV-positive serostatus. Exclusion criteria were a) acute distress attributable to sexual revictimization experienced within the past month, b) presence of impaired mental status, or c) extreme distress evidenced by suicidal intent or severe depressive symptomatology with a score of 30 or greater on the Beck Depression Inventory (Beck & Steer, 1993).

Of 33 individuals screened out for study eligibility, 21 did not meet study inclusion criteria (7 severely depressed, 7 not sexually abused, 6 cognitively impaired, and 1 acute sexual revictimization) and 41 declined or were not available for further participation in the study (23 not located, 13 not interested/unable to continue, 3 incarcerated, and 2 died/hospitalized). Thus, 271 individuals participated in baseline data collection. An additional 64 participants were excluded from the analysis because they were not prescribed HAART and had CD4 count over 350. One participant was also excluded for missing demographic data, leaving a final sample size of 206.

Procedure

Baseline assessment data were used for the current analyses. For further information on methods, see previous published studies (Meade et al., 2009; Meade et al., 2010; Sikkema et al., 2007; Sikkema et al., 2008).

Measures

The model was comprised of latent and observed variables, which were either continuous or dichotomous.

Latent life stress

The latent life stress variable was measured with the Life Problems, Stressors, and Concerns questionnaire (LPSC) (DeMarco, Ostrow, & DiFranceisco, 1999). Two LPSC items that assessed stress related to drug/alcohol use were excluded because the variable was intended to capture stress independent of drug use. Following statistical recommendations for parceling (Little, Cunningham, Shahar, & Widaman, 2002), the 26 remaining items were randomly parceled into four indicators of life stress (standardized factor loadings for the four indicators were .79, .78, .87, and .91).

Avoidant coping

Avoidant coping was measured using the Ways of Coping Questionnaire (Folkman & Lazarus, 1988) and the Coping with Illness scale (Namir, Wolcott, Fawzy, & Alumbaugh, 1987). Prior empirical work has found that when these scales are combined, five coping factors emerge (Sikkema, Hansen, Meade, Kochman, & Fox, 2009; Sikkema et al., 2012). The avoidant coping factor was used (α= .91) and self-destructive coping was omitted due to multiple items involving drug use.

Latent psychological distress

The latent psychological distress variable was measured with three indicators. The Impact of Events Scale (IES) (Horowitz, Wilner, & Alvarez, 1979) measures overall posttraumatic stress disorder severity (α = .92). The Center for Epidemiologic Studies Depression Scale (CES-D) (Radloff, 1977) is designed to measure depression severity in clinical populations (α = .92). The Symptom Checklist 90-R Global Severity subscale (SCL 90-R) (Derogatis & Cleary, 1977) is regarded as a measure of general psychological wellbeing (α = .98).

Drug use

Participants reported the number of days in the past month they used a variety of illicit drugs (i.e., cocaine, crack, heroin, or methamphetamine). Participants received a score of ‘1’ on drug use if they reported using an illicit drug at least 6 days in the prior month. Otherwise, they received a score of ‘0’ on drug use. This methodology for categorizing drug use was adapted from the National Survey on Drug Use and Health (SAMHSA, 2011) in order to capture habitual drug use.

HAART Adherence

Participants reported the percentage of time in the past month that they took all prescribed HAART. In accordance with guidelines of optimal adherence (Paterson et al., 2000), adherence was categorized as optimal (at least 90%) or suboptimal (below 90%). Individuals were also classified as suboptimally adherent if they fit in the following category: CD4 count < 350 but not taking HIV medication. This was intended to capture individuals whose CD4 count was low enough to necessitate HAART but were not taking medication (Panel on Antiretroviral Guidelines for Adults and Adolescents, 2009).

HIV disease progression

Participants reported last measured viral load and CD4 count. Due to its high skewness and kurtosis (5.99 and 38.02, respectively), high variance, and in keeping with previous empirical work, viral load was natural log transformed with a constant added to retain zero values (Lucas, Cheever, Chaisson, & Moore, 2001; Quinn et al., 2000). A linear transformation (divided by 100) was performed on CD4 count in order to reduce the range of scores. These variables were intended as a validity check on participants’ self-reports of medication adherence.

Statistical Methods

Prior to analyses, data were analyzed for patterns of missingness. Bivariate analysis included tests of gender differences in all predictor and outcome variables, as well as correlations among variables. The structural equation model was tested in Mplus Version 6 with continuous/categorical variable methodology (CVM) (Muthen, 1983), which uses a combination of full information maximum likelihood (FIML) and mean- and variance-adjusted weighted least squares estimation. Overall goodness of model fit was assessed with the χ2 statistic, the root mean square error of approximation (RMSEA), the comparative fit index (CFI), the standardized root mean square residual (SRMR), and the weighted root mean residual (WRMR). Goodness of fit was assessed with the following cutoffs: χ2 p-value < .05, RMSEA ≤ .05, CFI .≥ 95, and either SRMR < .08 or WRMR ≤.95 (Hoyle, 2011; Kline, 2010; Yu, 2002). The measurement model estimated all latent variables using a fully saturated model in which all latent variables were allowed to correlate. Given good fit of the measurement model, goodness of fit was then tested for the preliminary structural model, which included key variables, hypothesized paths, and demographic covariates (gender was not taken into account).

In order to test gender moderation in the adherence model, both the measurement model and structural model were considered successively. In keeping with accepted statistical procedures for testing multigroup models (Hoyle, 2011; Kline, 2010), the DIFFTEST command in Mplus was utilized to employ a chi-square difference test to compare the gender equivalence model (H0) to a series of gender difference models for each model path. After first establishing that male and female participants were equivalent on factor loadings in the measurement model, the structural model was examined. For paths that exhibited gender moderation, path magnitudes were allowed to be estimated freely by gender. Finally, indirect effects were examined for gender moderation.

Results

Bivariate gender differences

Gender differences emerged in bivariate statistical tests. Women and men differed in their years living with HIV, education level, ethnicity, and sexual orientation. Women had fewer years living with HIV, lower average educational attainment and were more likely to be African American and heterosexual (see Table 1). These variables were added as covariates in the model. See Table 2 for the variable correlation matrix by gender.

Table 1.

Sample characteristics by gender

Females
N = 94
Mean (SD)
N (%)		 Males
N = 112
Mean (SD)
N (%)		 Difference Test p value
Age 43.1 (7.2) 41.9 (6.7) t (204) = 1.24 .22
Years of Education 11.3 (1.9) 13.1 (2.5) t (201) = −5.71 <.001
Years Living with HIVa 8.3 (4.6) 11.2 (5.5) t (188) = −3.86 <.001
Ethnicity
 Non-Latino White 5 (5.3%) 19 (17.0%) χ2 (1) = 6.73 .009
 African American 72 (76.6%) 68 (60.7%) χ2 (1) = 5.92 .015
 Latino/a 14 (14.9%) 21 (18.8%) χ2 (1) = 0.54 .46
 Multiracial/Other 3 (3.2%) 4 (3.6%) χ2 (1) = 0.02 .88
Sexual Orientation
 Heterosexual 76 (80.9%) 11 (9.8%) χ2 (1) = 105.69 <.001
 Gay/Lesbian 7 (7.4%) 83 (74.1%) χ2 (1) = 92.31 <.001
 Bisexual 11 (11.7%) 18 (16.1%) χ2 (1) = 0.81 .37
Drug Useb 9 (9.6%) 15 (13.4%) χ2 (1) = 0.72 .40
 Cocaine 6 (6.4%) 10 (8.9%) χ2 (1) = 0.46 .50
 Crack 7 (7.4%) 9 (8.0%) χ2 (1) = 0.03 .88
 Methamphetamine 0 (0.0%) 4 (3.6%) χ2 (1) = 3.43 .064
 Heroin 0 (0.0%) 1 (0.9%) χ2 (1) = 0.84 .36
Optimal Adherence 63 (67.0%) 78 (69.6%) χ2 (1) = 0.16 .69
CD4 Countc 462 (361) 375 (257) t (172) = 1.85 .066
Ln Viral Loadc 2.52 (5.52) 3.12 (5.79) t (142) = −0.62 .54
Open in a new tab
a

Data on years living with HIV were missing for 16 participants.

b

Drug use was defined as using crack, cocaine, heroin, or methamphetamine at least 6 days in the past month.

c

Data on self-reported CD4 Count and viral load were incomplete, with missing data points numbering 32 and 62, respectively

Table 2.

Correlation matrix by gender

1 2 3 4 5 6 7 8 9 10 11
1. Life stressa - .640 .400 .334 −.216 .083 −.108 −.023 .079 −.048 −.091
2. Psychological distressa .530 - .827 .521 −.337 .129 −.168 .118 −.174 −.102 −.296
3. Avoidant copingb .278 .520 - .531** −.309* .305** −.211 .013 −.082 .041 −.122
4. Drug usec .099 .186 .068 - −.324 .283 −.140 .299 .076 −.418* −.309
5. Optimal adherencec −.019 −.035 .234* −.164 - −.426** .555** .252 .004 .238 −.153
6. Ln Viral Loadb .005 .010 −.083 −.095 −.313* - −.384** .110 −.063 −.017 .073
7. CD4b −.010 −.019 .216* −.119 .457** −.370** - .268 −.098 .143 .183
8. African Americanc −.043 .122 .217 −.333 −.275 .061 −.102 - .028 .138 −.204
9. Years educationb −.140 .065 −.135 .018 .101 .060 −.065 −.174 - .165 .051
10. Sexual orientationd .151 −.015 −.095 .096 .123 −.043 −.217 .273 −.275* - .335*
11. Years with HIVb −.093 −.018 −.098 −.045 −.238 −.060 .099 .096 −.021 −.258* -
Open in a new tab

Note. Correlations coefficients are shown for males and females separately; coefficients for females (n = 94) appear above the diagonal, and males (n = 112) appear below the diagonal.

Due to the categorical/continuous variable (CVM) methodology utilized in this analysis, correlations differ in type. Types include: Pearson (for categorical variable pairs), tetrachoric (for dichotomous variable pairs), polychoric (for dichotomous/trichotomous pairs), biserial (for dichotomous/continuous pairs), or polyserial (for trichotomous/continuous pairs). For correlations with latent variables, significance level cannot be calculated in Mplus v6 (L. K. Muthen, 2012).

a

Latent variable.

b

Continuous variable.

c

Dichotomous variable.

d

Trichotomous variable.

*

p < .05.

**

p < .01

Preliminary model

Since preliminary fit of the measurement model was not within cutoff values, modification indices were utilized. A path accounting for correlation between two life stress indicators was added to the model. After this modification, overall fit of the measurement model was within acceptable parameters (χ2 = 25.57, p = .01; CFI = .99; RMSEA = .07; SRMR = .02). Chi-square difference tests of the measurement model did not yield any significant gender differences on factor loadings. Goodness of fit indices for the preliminary structural model (not taking gender into account) were also within acceptable parameters, χ2 (69) = 81.11, p = .15; CFI = .98; RMSEA = .03; WRMR = .52.

Test of structural gender differences

In testing group invariance in the structural model, gender significantly moderated four paths: 1) the effect of psychological distress on drug use, 2) the effect of drug use on medication adherence, 3) the effect of avoidant coping on psychological distress, and 4) the correlation between stress and avoidant coping (see Figure 1 for final model). The multigroup model accounting for gender demonstrated good fit: χ2 (152) = 159.68, p = .32; CFI = .98; RMSEA = .02; WRMR = .75. In contrast to women, two paths were non-significant in the male sample. For men, psychological distress did not significantly predict drug use, nor did drug use significantly predict HAART adherence. However, for women these paths demonstrated strong effects.

Figure 1.

Figure 1

Open in a new tab

Final multigroup model illustrating gender moderation

Note. All paths show standardized estimates. Bolded paths reflect significant gender moderation; separate estimates by gender are noted with female estimates first. Circles denote latent variables, while rectangles denote observed variables. Covariates not depicted are ethnicity (African American vs. not), sexual orientation, years of education, and years living with HIV.

* p < .05. ** p < .01

Gender was further explored in the context of mediational paths, and gender differences were found in two indirect effects. For women, two indirect effects were significant: a) the effect of psychological distress on adherence, mediated by drug use, β = −.32, p = .018; and b) the effect of drug use on viral load, mediated by adherence, β = .24, p = .048. Female participants demonstrated a marginal indirect effect of psychological distress on viral load through adherence and drug use, β = .12, p = .085. For male participants, no indirect paths were significant.

Discussion

The purpose of this study was to explore gender moderation and drug use mediation within a model of HAART adherence in a sample of people living with HIV/AIDS and CSA. Study results indicated that gender significantly moderated influences on HAART adherence, most notably the effect of drug use. For women, but not men, drug use mediated the association between psychological symptoms and adherence. Results overall provide strong evidence for gender differences in key psychosocial influences on HAART adherence and HIV disease progression.

Study findings expand upon previous research that demonstrates the importance of gender when examining HAART adherence. For women, the relationship between psychological distress, drug abuse, and HAART adherence is likely stronger than for men, though this gender moderation may be especially evident in individuals with CSA. This finding corroborates prior evidence that gender moderates the influence of alcohol use on HAART adherence. Specifically, women, but not men, report lower levels of HAART adherence when they are problem drinkers (Berg et al., 2004). This is noteworthy because in the general population, substance use coping and negative consequences of substance use are associated more with male gender (Bolton, Robinson, & Sareen, 2009). Taken together, these studies suggest that women with HIV and CSA may be more likely a) to use substances as a coping strategy and b) to experience a negative impact of substance use on HAART adherence.

For men with a history of CSA, no study variables significantly predicted either HAART adherence or drug use. This suggests two hypotheses for further testing. First, men’s adherence may be more strongly affected by medication side effects, beliefs/attitudes about HAART. Past evidence suggests that these relationships may be particularly strong for men, but has not specifically tested gender moderation of the influence of these factors on adherence (Berg et al., 2004; Gonzalez et al., 2007). Secondly, HIV-positive men with a history of CSA may use drugs for different reasons than women. In the current sample, drug use was equally prevalent in men and women; however, the associations with drug use differed significantly between genders. There may be culturally-influenced factors that determine predictors and outcomes of drug use in individuals with HIV and CSA. While drug use increases transmission risk behavior (Morin et al., 2005), it may not necessarily lead to decreased HAART adherence.

Another incongruity with prior literature relates to the moderating effect of gender on the relationship between psychological distress and HAART adherence. In the current study, only women exhibited an association between psychological distress and HAART adherence. This finding contradicts results from a study of older adults with HIV, in which depression was associated with lower HAART adherence for men only (Bianco et al., 2011). This difference in findings could be explained by divergent study populations – the current study’s sample had experienced significant trauma, were younger, and included a higher percentage of African American participants. This discrepancy in study results advocates for the inclusion of Gender X Age interaction when investigating adherence in a wide age range.

There are some limitations to the study. First, the cross-sectional design limits conclusions regarding directionality of relationships. In order to address this limitation, both theoretical foundations and prior empirical evidence were considered to provide justification for directionality in the model. Additionally, HAART adherence, CD4 count and viral load were assessed by self-report, which could result in decreased reliability. However, self-reported HAART adherence does demonstrate validity when assessed against pill count and viral load (Simoni et al., 2006); further, self-report measures of CD4 count and viral load are comparable to electronic medical records (Kinsler, Cunningham, Mohanty, & Wong, 2008). Given the effect sizes of the relationships between CD4 count, viral load, and self-reported HAART adherence, it is likely that the self-reported variables demonstrate an adequate level of validity. Finally, data for this study were collected 10 years ago, which could adversely affect generalizability to current HIV-positive populations. This is a concern particularly with regard to study of adherence, since simplified HAART medication regimens have the potential to improve adherence overall (Langebeek et al., 2013). However, the effects of CSA on HIV disease trajectory have continued throughout the past decade (Pellowski, Kalichman, Matthews, & Adler, 2013). Thus, it is likely that these findings exhibit current relevance.

Despite these limitations, the study is strengthened by a number of methodological elements. Study hypotheses had a firm basis in theory and previous empirical work, expanding on a previously validated model of negative affect mediation of HAART adherence (Johnson et al., 2009). Additionally, analysis utilized robust statistical methods to assess gender moderation. Finally, the sample included a high ratio of female participants, as well as individuals at multiple phases of disease progression.

Several implications arise for clinical practice and future research. First, gender may moderate multiple influences on HAART adherence, such as medication side effects, food instability, social support, housing status, and trust of health care providers (Cooper, Gellaitry, Hankins, Fisher, & Horne, 2009; Gonzalez et al., 2007; Kalichman & Grebler, 2010; Milloy et al., 2012; Simoni et al., 2006). Future research should consistently test gender moderation of HAART adherence. Secondly, it is critical to treat substance abuse as part of HIV management in women, since alcohol and drug use particularly impact adherence for women. Existing interventions address poor adherence in the context of psychological dysfunction and substance use, such as Cognitive Behavioral Therapy for Adherence and Depression and antiretroviral Directly Observed Therapy (Nahvi et al., 2012; Safren et al., 2009; Safren et al., 2012). Our findings suggest that these interventions could potentially be targeted by gender. One such targeted intervention for HIV-positive women with CSA histories reduced sexual risk behavior and increased HAART adherence (Wyatt et al., 2004). Interventions that focus on underlying gender differences, psychological symptoms, as well as behavioral mechanisms may demonstrate improved efficacy in yielding long-term optimal HAART adherence.

Acknowledgments

Funding: Funding sources include This research was supported by grants R01-MH062965 and R01-MH078731 from the United States National Institute of Mental Health and P30-AI064518 from the United States National Institute of Allergy and Infectious Disease.

Footnotes

The authors have no conflicts of interest to declare.

References

  1. Beck AT, Steer RA. BDI: Beck depression inventory manual. New York: Psychological Corporation; 1993. [Google Scholar]
  2. Berg KM, Demas PA, Howard AA, Schoenbaum EE, Gourevitch MN, Arnsten JH. Gender differences in factors associated with adherence to antiretroviral therapy. Journal of General Internal Medicine. 2004;19(11):1111–1117. doi: 10.1111/j.1525-1497.2004.30445.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Bianco JA, Heckman TG, Sutton M, Watakakosol R, Lovejoy T. Predicting Adherence to Antiretroviral Therapy in HIV-Infected Older Adults: The Moderating Role of Gender. AIDS and Behavior. 2011;15(7):1437–1446. doi: 10.1007/s10461-010-9756-2. [DOI] [PubMed] [Google Scholar]
  4. Bolton JM, Robinson J, Sareen J. Self-medication of mood disorders with alcohol and drugs in the National Epidemiologic Survey on Alcohol and Related Conditions. Journal of Affective Disorders. 2009;115(3):367–375. doi: 10.1016/j.jad.2008.10.003. [DOI] [PubMed] [Google Scholar]
  5. Bornovalova MA, Ouimette P, Crawford AV, Levy R. Testing gender effects on the mechanisms explaining the association between post-traumatic stress symptoms and substance use frequency. Addictive Behaviors. 2009;34(8):685–692. doi: 10.1016/j.addbeh.2009.04.005. http://dx.doi.org/10.1016/j.addbeh.2009.04.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Centers for Disease Control and Prevention. Monitoring selected national HIV prevention and care objectives by using HIV surveillance data— United States and 6 U.S. dependent areas—2010. 2012 Retrieved from HIV Surveillance Supplemental Report. [Google Scholar]
  7. Cohen MA, Alfonso CA, Hoffman RG, Milau V, Carrera G. The impact of PTSD on treatment adherence in persons with HIV infection. General Hospital Psychiatry. 2001;23(5):294–296. doi: 10.1016/s0163-8343(01)00152-9. [DOI] [PubMed] [Google Scholar]
  8. Cooper V, Gellaitry G, Hankins M, Fisher M, Horne R. The influence of symptom experiences and attributions on adherence to highly active anti-retroviral therapy (HAART): a six-month prospective, follow-up study. AIDS Care. 2009;21(4):520–528. doi: 10.1080/09540120802301824. [DOI] [PubMed] [Google Scholar]
  9. DeMarco F, Ostrow D, DiFranceisco W. General and AIDS-Specific Stress, Coping, and Psychological Distress in the Biracial Coping & Change Study Cohort of Gay Men. AIDS and Behavior. 1999;3(3):177–186. doi: 10.1023/A:1025496209190. [DOI] [Google Scholar]
  10. Derogatis LR, Cleary PA. Factorial invariance across gender for the primary symptom dimensions of the SCL-90. British Journal of Social and Clinical Psychololgy. 1977;16(4):347–356. doi: 10.1111/j.2044-8260.1977.tb00241.x. [DOI] [PubMed] [Google Scholar]
  11. Felitti VJ, Anda RF, Nordenberg D, Williamson DF, Spitz AM, Edwards V, Marks JS. Relationship of childhood abuse and household dysfunction to many of the leading causes of death in adults - The adverse childhood experiences (ACE) study. American Journal of Preventive Medicine. 1998;14(4):245–258. doi: 10.1016/s0749-3797(98)00017-8. [DOI] [PubMed] [Google Scholar]
  12. Folkman S, Lazarus RS. Manual for the Ways of Coping Questionnaire: Research Edition. Palo Alto, CA: Consulting Psychological Press; 1988. [Google Scholar]
  13. Golin CE, Liu HH, Hays RD, Miller LG, Beck CK, Ickovics J, Wenger NS. A prospective study of predictors of adherence to combination antiretroviral medication. Journal of General Internal Medicine. 2002;17(10):756–765. doi: 10.1046/j.1525-1497.2002.11214.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Gonzalez JS, Penedo FJ, Llabre MM, Duran RE, Antoni MH, Schneiderman N, Horne R. Physical symptoms, beliefs about medications, negative mood, and long-term HIV medication adherence. Annals of Behavioral Medicine. 2007;34(1):46–55. doi: 10.1007/BF02879920. [DOI] [PubMed] [Google Scholar]
  15. Gore-Felton C, Koopman C. Behavioral mediation of the relationship between psychosocial factors and HIV disease progression. Psychosomatic Medicine. 2008;70(5):569–574. doi: 10.1097/Psy.0b013e318177353e. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Hendershot CS, Stoner SA, Pantalone DW, Simoni JM. Alcohol use and antiretroviral adherence: Review and meta-analysis. JAIDS. 2009;52(2):180–202. doi: 10.1097/QAI.0b013e3181b18b6e. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Henny KD, Kidder DP, Stall R, Wolitski RJ. Physical and sexual abuse among homeless and unstably housed adults living with HIV: Prevalence and associated risks. AIDS and Behavior. 2007;11(6):842–853. doi: 10.1007/s10461-007-9251-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Horowitz M, Wilner M, Alvarez W. Impact of Event Scale: A measure of subjective stress. Psychosomatic Medicine. 1979;41:209–218. doi: 10.1097/00006842-197905000-00004. [DOI] [PubMed] [Google Scholar]
  19. Hoyle RH. Structural equation modeling for social and personality psychology. Thousand Oaks, CA: SAGE Publications, Inc; 2011. [Google Scholar]
  20. Johnson CJ, Heckman TG, Hansen NB, Kochman A, Sikkema KJ. Adherence to antiretroviral medication in older adults living with HIV/AIDS: a comparison of alternative models. AIDS Care. 2009;21(5):541–551. doi: 10.1080/09540120802385611. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Kalichman SC, Grebler T. Stress and poverty predictors of treatment adherence among people with low-literacy living w with HIV/AIDS. Psychosomatic Medicine. 2010;72(8):810–816. doi: 10.1097/PSY.0b013e3181f01be3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Kalichman SC, Sikkema KJ, DiFonzo K, Luke W, Austin J. Emotional adjustment in survivors of sexual assault living with HIV-AIDS. Journal of Traumatic Stress. 2002;15(4):289–296. doi: 10.1023/A:1016247727498. [DOI] [PubMed] [Google Scholar]
  23. Kinsler J, Cunningham W, Mohanty S, Wong M. Are medical records a more reliable and valid source of CD4 count, viral load, and outpatient visit data than self-reports? A comparison with electronic medical records as the gold standard. Health Services and Outcomes Research Methodology. 2008;8(4):187–200. doi: 10.1007/s10742-008-0036-9. [DOI] [Google Scholar]
  24. Kline RB. Principles and practice of structural equation modeling. 3. New York: Guilford Press; 2010. [Google Scholar]
  25. Kreitchmann R, Harris DR, Kakehasi F, Haberer JE, Cahn P, Losso M, Team NLS. Antiretroviral adherence during pregnancy and postpartum in Latin America. AIDS Patient Care and STDs. 2012;26(8):486–495. doi: 10.1089/apc.2012.0013. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Langebeek N, Sprenger HG, Gisolf EH, Reiss P, Sprangers MAG, Legrand JC, Nieuwkerk PT. A simplified combination antiretroviral therapy regimen enhances adherence, treatment satisfaction and quality of life: results of a randomized clinical trial. HIV medicine. 2013 doi: 10.1111/hiv.12112. Advance online publication. [DOI] [PubMed] [Google Scholar]
  27. Lazarus RS, Folkman S. Stress, Appraisal, and Coping. New York: Springer Publishing Company, Inc; 1984. [Google Scholar]
  28. Little TD, Cunningham WA, Shahar G, Widaman KF. To parcel or not to parcel: Exploring the question, weighing the merits. Structural Equation Modeling. 2002;9(2):151–173. [Google Scholar]
  29. Liu HH, Longshore D, Williams JK, Rivkin I, Loeb T, Warda US, Wyatt G. Substance abuse and medication adherence among HIV-positive women with histories of child sexual abuse. AIDS and Behavior. 2006;10(3):279–286. doi: 10.1007/s10461-005-9041-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Lucas GM, Cheever LW, Chaisson RE, Moore RD. Detrimental effects of continued illicit drug use on the treatment of HIV-1 infection. JAIDS. 2001;27(3):251–259. doi: 10.1097/00126334-200107010-00006. [DOI] [PubMed] [Google Scholar]
  31. Machtinger EL, Wilson TC, Haberer JE, Weiss DS. Psychological trauma and PTSD in HIV-positive women: A meta-analysis. AIDS and Behavior. 2012;16(8):2091–2100. doi: 10.1007/s10461-011-0127-4. [DOI] [PubMed] [Google Scholar]
  32. Markowitz SM, O’Cleirigh C, Hendriksen ES, Bullis JR, Stein M, Safren SA. Childhood sexual abuse and health risk behaviors in patients with HIV and a history of injection drug use. AIDS and Behavior. 2011;15(7):1554–1560. doi: 10.1007/s10461-010-9857-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Marks King R, Vidrine DJ, Danysh HE, Fletcher FE, McCurdy S, Arduino RC, Gritz ER. Factors associated with nonadherence to antiretroviral therapy in HIV-positive smokers. AIDS Patient Care and STDs. 2012;26(8):479–485. doi: 10.1089/apc.2012.0070. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Meade CS, Drabkin AS, Hansen NB, Wilson PA, Kochman A, Sikkema KJ. Reductions in alcohol and cocaine use following a group coping intervention for HIV-positive adults with childhood sexual abuse histories. Addiction. 2010;105(11):1942–1951. doi: 10.1111/j.1360-0443.2010.03075.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Meade CS, Hansen NB, Kochman A, Sikkema KJ. Utilization of medical treatments and adherence to antiretroviral therapy among HIV-positive adults with histories of childhood sexual abuse. AIDS Patient Care and STDs. 2009;23(4):259–266. doi: 10.1089/apc.2008.0210. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Milloy MJS, Marshall BDL, Kerr T, Buxton J, Rhodes T, Montaner J, Wood E. Social and structural factors associated with HIV disease progression among illicit drug users: A systematic review. AIDS. 2012;26(9):1049–1063. doi: 10.1097/QAD.0b013e32835221cc. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Molnar BE, Buka SL, Kessler RC. Child sexual abuse and subsequent psychopathology: Results from the National Comorbidity Survey. American Journal of Public Health. 2001;91(5):753–760. doi: 10.2105/ajph.91.5.753. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Morin SF, Steward WT, Charlebois ED, Remien RH, Pinkerton SD, Johnson MO, Chesney MA. Predicting HIV transmission risk among HIV-Infected men who have sex with men - Findings from the healthy living project. JAIDS. 2005;40(2):226–235. doi: 10.1097/01.qai.0000166375.16222.eb. [DOI] [PubMed] [Google Scholar]
  39. Muthen B. Latent variable structural equation modeling with categorical data. Journal of Econometrics. 1983;22(1–2):43–65. [Google Scholar]
  40. Muthen LK. Tech 4 output. Message board conversation. 2012 Nov 30; Retrieved from http://www.statmodel.com/discussion/messages/11/11276.html.
  41. Nahvi S, Litwin AH, Heo M, Berg KM, Li X, Arnsten JH. Directly observed antiretroviral therapy eliminates adverse effects of active drug use on adherence. Drug and Alcohol Dependence. 2012;120(1):174–180. doi: 10.1016/j.drugalcdep.2011.07.025. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Namir S, Wolcott DL, Fawzy FI, Alumbaugh MJ. Coping with AIDS - Psychological and health implications. Journal of Applied Social Psychology. 1987;17(3):309–328. [Google Scholar]
  43. Panel on Antiretroviral Guidelines for Adults and Adolescents. Guidelines for the use of antiretroviral agents in HIV-1-infected adults and adolescents. U.S. Department of Health and Human Services; 2009. Available at: http://www.aidsinfo.nih.gov/ContentFiles/AdultandAdolescentGL.pdf. [Google Scholar]
  44. Paterson DL, Swindells S, Mohr J, Brester M, Vergis EN, Squier C, Singh N. Adherence to protease inhibitor therapy and outcomes in patients with HIV infection. Annals of Internal Medicine. 2000;133(1):21–30. doi: 10.7326/0003-4819-133-1-200007040-00004. [DOI] [PubMed] [Google Scholar]
  45. Quinn TC, Wawer MJ, Sewankambo N, Serwadda D, Li CJ, Wabwire-Mangen F Rakai Project Study G. Viral load and heterosexual transmission of human immunodeficiency virus type 1. New England Journal of Medicine. 2000;342(13):921–929. doi: 10.1056/NEJM200003303421303. [DOI] [PubMed] [Google Scholar]
  46. Radloff LS. The CES-D Scale: A self-report depression scale for research in the general population. Applied Psychological Measurement. 1977;1(3):385–401. doi: 10.1177/014662167700100306. [DOI] [Google Scholar]
  47. Safren SA, O’Cleirigh C, Tan JY, Raminani SR, Reilly LC, Otto MW, Mayer KH. A randomized controlled trial of cognitive behavioral therapy for adherence and depression (CBT-AD) in HIV-infected individuals. Health Psychology. 2009;28(1):1–10. doi: 10.1037/a0012715. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Safren SA, O’Cleirigh CM, Bullis JR, Otto MW, Stein MD, Pollack MH. Cognitive behavioral therapy for adherence and depression (CBT-AD) in HIV-infected injection drug users: A randomized controlled trial. Journal of Consulting and Clinical Psychology. 2012;80(3):404–415. doi: 10.1037/a0028208. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Sikkema KJ, Hansen NB, Kochman A, Tarakeshwar N, Neufeld S, Meade CS, Fox AM. Outcomes from a group intervention for coping with HIV/AIDS and childhood sexual abuse: Reductions in traumatic stress. AIDS and Behavior. 2007;11(1):49–60. doi: 10.1007/s10461-006-9149-8. [DOI] [PubMed] [Google Scholar]
  50. Sikkema KJ, Hansen NB, Meade CS, Kochman A, Fox AM. Psychosocial predictors of sexual HIV transmission risk behavior among HIV-positive adults with a sexual abuse history in childhood. Archives of Sexual Behavior. 2009;38(1):121–134. doi: 10.1007/s10508-007-9238-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Sikkema KJ, Ranby KW, Meade CS, Hansen NB, Wilson PA, Kochman A. Reductions in traumatic stress following a coping intervention were mediated by decreases in avoidant coping for people living with HIV/AIDS and childhood sexual abuse. Journal of Consulting and Clinical Psychology. 2012 doi: 10.1037/a0030144. No pagination specified. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Sikkema KJ, Wilson PA, Hansen NB, Kochman A, Neyfeld S, Ghebremichael MS, Kershaw T. Effects of a coping intervention on transmission risk behavior among people living with HIV/AIDS and a history of childhood sexual abuse. JAIDS. 2008;47(4):506–513. doi: 10.1097/QAI.0b013e318160d727. [DOI] [PubMed] [Google Scholar]
  53. Simoni J, Kurth A, Pearson C, Pantalone D, Merrill J, Frick P. Self-report measures of antiretroviral therapy adherence: A review with recommendations for HIV research and clinical management. AIDS and Behavior. 2006;10(3):227–245. doi: 10.1007/s10461-006-9078-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Simoni JA, Frick PA, Huang B. A longitudinal evaluation of a social support model of medication adherence among HIV-positive men and women on antiretroviral therapy. Health Psychology. 2006;25(1):74–81. doi: 10.1037/0278-6133.25.1.74. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Substance Abuse and Mental Health Services Administration. Results from the 2010 National Survey on Drug Use and Health: Summary of national findings. Rockville, MD: U.S. Department of Health and Human Services; 2011. NSDUH Series H-41, HHS Publication No. (SMA) 11–4658. [Google Scholar]
  56. Tapp C, Milloy MJ, Kerr T, Zhang R, Guillemi S, Hogg RS, Wood E. Female gender predicts lower access and adherence to antiretroviral therapy in a setting of free healthcare. BMC Infectious Diseases. 2011;11 doi: 10.1186/1471-2334-11-86. [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Vanderkolk BA, Fisler RE. Childhood abuse and neglect and loss of self-regulation. Bulletin of the Menninger Clinic. 1994;58(2):145–168. [PubMed] [Google Scholar]
  58. Weaver KE, Llabre MM, Duran RE, Antoni MH, Ironson G, Penedo FJ, Schneiderman N. A stress and coping model of medication adherence and viral load in HIV-positive men and women on highly active antiretroviral therapy (HAART) Health Psychology. 2005;24(4):385–392. doi: 10.1037/0278-6133.24.4.385. [DOI] [PubMed] [Google Scholar]
  59. Welles SL, Baker AC, Miner MH, Brennan DJ, Jacoby S, Rosser BS. History of childhood sexual abuse and unsafe anal intercourse in a 6-city study of HIV-positive men who have sex with men. American Journal of Public Health. 2009;99(6):1079–1086. doi: 10.2105/AJPH.2007.133280. [DOI] [PMC free article] [PubMed] [Google Scholar]
  60. Whetten K, Leserman J, Lowe K, Stangl D, Thielman N, Swartz M, Van Scoyoc L. Prevalence of childhood sexual abuse and physical trauma in an HIV-positive sample from the deep south. American Journal of Public Health. 2006;96(6):1028–1030. doi: 10.2105/ajph.2005.063263. [DOI] [PMC free article] [PubMed] [Google Scholar]
  61. Wyatt GE, Longshore D, Chin D, Carmona JV, Loeb TB, Myers HF, Rivkin I. The efficacy of an integrated risk reduction intervention for HIV- positive women with child sexual abuse histories. AIDS and Behavior. 2004;8(4):453–462. doi: 10.1007/s10461-004-7329-y. [DOI] [PubMed] [Google Scholar]
  62. Yu C. PhD Doctoral Dissertation. University of California; Los Angeles: 2002. Evaluating cutoff criteria of model fit indices for latent variable models with binary and continuous outcomes. [Google Scholar]

RESOURCES