Epidemiology & Social
Attitudes toward HIV treatments influence unsafe sexual and injection practices among injecting drug users
Tun, Waimar; Celentano, David D; Vlahov, Davida; Strathdee, Steffanie A
From the Infectious Disease Program, Department of Epidemiology, Johns Hopkins University, Bloomberg School of Public Health, Baltimore, Maryland, and the aCenter for Urban Epidemiologic Studies, New York Academy of Medicine, New York, New York, USA.
Correspondence to D. D. Celentano, Johns Hopkins University, Bloomberg School of Public Health, 615 North Wolfe Street, Room E6008, Baltimore, Maryland 21205, USA.
Received: 25 July 2002; revised: 20 November 2002; accepted: 19 March 2003.
Objective: To determine if HIV treatment-related attitudes are associated with unprotected sex and needle sharing among HIV-seropositive and -seronegative injecting drug users (IDU) in Baltimore, Maryland.
Design and methods: IDU participating in a cohort study seen between December 2000 and July 2001 completed an interviewer-administered questionnaire on attitudes toward HIV treatment and risk behaviors (593 HIV-seronegative, 338 HIV-seropositive), including: perceived HIV transmissibility through unprotected sex and needle sharing, and safer sex and injection fatigue. Logistic regression was used to examine the role of attitudinal factors on needle sharing and unsafe sex.
Results: Almost two-thirds of sexually active participants engaged in unprotected sex and approximately half of those injecting drugs shared needles. Among HIV-seropositive IDU, perception of reduced HIV transmissibility through unprotected sex was significantly associated with unprotected sex [adjusted odds ratio (AOR), 3.33; 95% confidence interval (CI), 1.05–10.55). Safer injection fatigue was independently associated with needle sharing among HIV-seropositive IDU (AOR, 6.55; 95% CI, 1.69–25.39). Among HIV-seronegative IDU, safer sex fatigue and safer injection fatigue were independently associated with unprotected sex (AOR, 3.12; 95% CI, 1.17–8.35) and needle sharing (AOR, 5.15; 95% CI, 2.33–11.37), respectively.
Conclusion: Among HIV-seropositive IDU, perceiving that HIV treatments reduce HIV transmission was significantly associated with unprotected sex. Risk reduction fatigue was strongly associated with unsafe sexual and injection behaviors among HIV-seronegative individuals. HIV prevention interventions must consider the unintended impact of HIV treatments on attitudes and risk behaviors among IDU.
The advent of highly active antiretroviral therapy (HAART) has decreased HIV-related mortality and morbidity, dramatically changing the face of the HIV/AIDS epidemic. HIV-infected patients with access to HIV treatment and care are living longer and healthier lives. Many individuals receiving HAART experience increased CD4 cell counts, decreased viral load, longer AIDS-free and survival time, with improved quality of life [1,2]. However, these treatment advances raise concerns about the potential for relapse to risky behaviors over time.
The effectiveness of HAART, as well as the depiction of healthy individuals in pharmaceutical advertisements, will no doubt play a role in affecting attitudes toward HIV infection, transmission risk, and, most importantly, risk behaviors. Risk-taking behaviors are a function of perceptions of disease severity, susceptibility, and benefits of and barriers to risk reduction [3,4]. The perception that HIV is less transmissible due to decreasing viral load or that HIV is less lethal due to more effective HIV treatments may lead to reduced caution in sexual and injection practices. Several studies have reported reduced concerns about HIV and increases in unsafe sex, anal/rectal gonorrhea, and HIV seroincidence among the gay and bi-sexual populations in the HAART era [5–9]. A Centers for Disease Control survey conducted in seven states among HIV-seronegative and untested individuals, including injecting drug users (IDU), men who have sex with men (MSM), and heterosexuals, revealed that people were generally less careful about sexual and drug use behaviors because of new HIV treatments .
However, little is known about attitudes regarding HIV and HIV-related risk behaviors among the drug injecting population since the advent of HAART. IDU represent one-third of all reported AIDS cases in the USA . There is some evidence among IDU that indicate a slight increase in unprotected sex, needle sharing and use of ‘shooting galleries’ among HAART initiators . However, another study among French IDU showed a strong association between HAART use and decreased sexual risk . Therefore, a better understanding of attitudes towards HIV treatments and their relationship to risky sexual and injection behaviors among IDU is valuable.
This paper assesses attitudes of IDU to HIV infection, HIV treatments and HIV-related risk behaviors and their association with risky sexual and injection practices. This paper will determine whether HIV risk perceptions are altered due to HAART, a concern commonly noted in the HIV research and treatment community. We hypothesized that treatment optimism, perception of reduced HIV severity and reduced importance of risk reduction and belief in reduced HIV transmissibility would be correlated with high-risk sexual and injection-related behaviors. Additionally, we hypothesized that fatigue from safer sex and injection practices would also be associated with risky sexual and injection practices.
Study population and data collection methods
The AIDS Link to Intravenous Experiences (ALIVE) Study is a longitudinal study of the natural history of and risk factors for HIV infection among community-recruited IDU in Baltimore City. From 1988 to 1989, 2960 IDU were enrolled into ALIVE. An additional 338 participants were recruited in 1994 to replenish the cohort; loss to follow-up in this cohort is minimal (approximately 10% annually). A detailed description of the ALIVE study recruitment and data collection methods has been published previously . Briefly, individuals were eligible for the study if they were at least 18 years of age, reported injecting illicit drugs within the previous 11 years, and were AIDS-free at the time of enrollment. Participants were recruited through extensive community outreach in order to recruit street IDU; over 85% reported being recruited by a friend or another study participant. All HIV-seropositive participants and a sample of HIV-seronegative participants were asked to participate in the follow-up study. Eligible participants returned to the study clinic every 6 months for a repeat interview on sexual and drug use behaviors, a complete physical examination, and venipuncture for laboratory analyses, including testing for HIV infection. HIV-seropositive participants were also asked about HIV medications initiated and/or used in the previous 6 months. All data were collected through interviewer-administered questionnaires, with the exception of data on sexual and drug use behaviors, which were collected through audio-computer assisted self-interview (ACASI). HIV infection was determined using enzyme-linked immunosorbent assay (Genetic Systems, Seattle, Washington, USA) and confirmed with Western blot assay (DuPont, Wilmington, Delaware, USA). The Committee on Human Research at the Johns Hopkins University Bloomberg School of Public Health approved this research study.
The study sample for the present analysis includes all ALIVE participants who attended a semi-annual visit between 1 December 2000 and 1 July 2001, and completed a survey on attitudes about HIV treatments and risk behaviors. Every ALIVE study participant who had a study visit during this time period was eligible to complete the attitudes survey.
HIV treatment-related attitudinal measures
Beginning in December 2000, all ALIVE participants also completed a survey on attitudes to HIV infection, HIV treatment, and HIV-related risk behaviors. The 27-item questionnaire was developed based on previously used questionnaires assessing such attitudes of gay and bi-sexual men in the Multi-center AIDS Cohort Study [6,7]; items regarding injection-related behaviors were newly constructed. Questionnaire items were pilot-tested on 20 selected participants, including HIV-seronegative and -seropositive participants (those receiving HAART and those not receiving any therapy), who attended a regularly scheduled follow-up visit. Responses to open-ended questions regarding the format of the questionnaire items were used to revise the items accordingly. It should be noted that items referring to HIV treatment did not refer specifically to HAART, but rather to any antiretroviral therapies to ensure comprehension by all respondents regardless of their HIV serostatus or HIV treatment status. Respondents were asked to answer questions regarding their beliefs and feelings towards ‘HIV medicines that are used to fight against the HIV virus.’ Each item was scored on a 5-point Likert scale, ranging from strongly disagree (1 point) to strongly agree (5 points), and was assigned to one of the scales described below. A response of ‘don't know’ was treated as missing data.
A principal components factor analysis using a varimax rotation yielded two factors: (i) perceived HIV transmissibility; and (ii) safe sex and injection fatigue. Details of this analysis have been described previously .
Scores for each scale were calculated by averaging the summed scores of the items. The scale scores were used as independent variables in the analysis assessing the association between the attitudinal factors and high-risk behaviors. The scores were categorized as low (score < 2.0), moderate (2.0 ≤ score ≤ 3.0) and high (score > 3.0).
Perceived HIV transmissibility
This scale consisted of four items assessing beliefs of reduced HIV transmissibility from engaging in unprotected sex or needle sharing with an HIV-seropositive individual receiving antiretroviral therapy (ART) or an HIV-seropositive person with an undetectable viral load (Cronbach's α, 0.74). A higher score reflects the belief in reduced HIV transmissibility. For the analysis determining the associations between attitudinal factors and high-risk behaviors, this scale was divided further into two subscales of ‘perceived HIV transmissibility through unprotected sex’ and ‘perceived HIV transmissibility through needle sharing’ due to their reference to two distinct activities. Each subscale included two questionnaire items. Due to the small number of items included in the split subscales, Cronbach's α decreased to 0.63 for ‘perceived transmissibility through unprotected sex’ and 0.61 for ‘perceived transmissibility through needle sharing', although both scales attained the minimally acceptable level of reliability.
Safer sex and injection fatigue
This scale consisted of six items assessing the degree to which respondents are tired of using condoms when having sex or using clean needles when injecting drugs. The Cronbach's α for the 6-item scale was 0.70. This scale was also divided further into two subscales of ‘safer sex fatigue’ (Cronbach's α, 0.58) and ‘safer injection fatigue’ (Cronbach's α, 0.58), again due to their reference to two distinct activities. Each subscale consisted of three items.
Outcome variables of interest included self-reported unprotected sex and needle sharing in the previous 6 months. Unprotected sex was defined as ‘not always using a condom’ in the past 6 months and needle sharing was defined as ‘sharing a needle or syringe at least once’ in the past 6 months. Separate analyses were conducted for unprotected sex and needle sharing.
Comparisons of demographic, clinical and risk behavior variables were made between the study sample and those excluded from the study sample. The chi-square test was used to assess differences in proportions, Student's t test for differences in means of continuous variables, and the Wilcoxon rank sum test to test for differences in medians. The chi-square test was also used to assess differences in proportions of HIV-seropositive and -seronegative individuals agreeing with attitude questionnaire items. Similar comparisons were made across HIV treatment groups among the HIV seropositive participants. The treatment groups consisted of those on HAART, those on antiretroviral regimens besides HAART (non-HAART), and those not on any treatment (no treatment). Responses were considered in agreement if a respondent indicated either ‘strongly agree’ or ‘somewhat agree'. Conversely, a response was considered a disagreement if the respondent indicated ‘strongly disagree', ‘somewhat disagree’ or ‘neither disagree or agree'.
To assess the association between attitudinal factors and high-risk behaviors, logistic regression was conducted to identify determinants of unprotected sex and needle sharing. The analysis for unprotected sex was restricted to the subsample of individuals who engaged in any sexual intercourse within the last 6 months (n = 535) and the analysis for needle sharing was restricted to those who injected drugs within the last 6 months (n = 459). Variables were included in the final multivariate models based on bivariate statistical significance (P < 0.05), and their identification in previous studies as factors associated with high-risk behaviors. Additionally, a variable was included in both the final models if it was significant in either of the multivariate models for unprotected sex or needle sharing in order to keep the models uniform. Additionally, multivariate models constructed for the HIV-seropositive group adjusted for CD4 cell count levels and HAART use.
All 1035 participants who had a follow-up visit between December 2000 and July 2001 completed the treatment belief questionnaire; however, a total of 931 participants (90.0%) had complete data on the 10 attitude questionnaire items of interest and were included in this analysis. Those who refused to answer or indicated ‘don't know’ for any of the items were excluded. Table 1 shows the demographic and clinical characteristics and risk behaviors of the study sample, and those excluded, as well as of the two subsamples that are later used to examine factors associated with unprotected sex and needle sharing. The majority of the study sample was African–American (93.2%), with a mean age of 44.8 years (SD, 6.8), and approximately one-third was currently employed. Those in the study sample did not differ from those excluded from the analysis with respect to race, age and employment status (P > 0.05). Of note, the study sample was more likely to be male (68.2% versus 58.3%; P = 0.04), inject drugs (49.7% versus 22.3%; P < 0.001) and slightly less likely to have sexual intercourse (57.8 versus 68.0%; P = 0.05) compared to those who were excluded. However, the groups were similar with regard to the two outcome variables: engaging in unprotected sex, and needle sharing. While the two groups had similar proportions that were HIV-seropositive (approximately one-third) and similar median CD4 cell counts, those included in the present analysis were less likely to be receiving any HIV treatment (67.0% versus 83.8%; P = 0.04).
Descriptive analysis of attitudinal measures
Table 2 shows the percentage of participants who agreed with the attitude questionnaire items stratified by HIV serostatus and by HIV treatment status. Approximately 10–20% of the total study sample agreed that HIV is less likely to be transmitted through needle sharing and unprotected sex due to current treatments and undetectable viral load. HIV-seropositive IDU were significantly more likely to score higher than HIV-seronegative IDU on both the perceived transmissibility through unprotected sex scale (P = 0.009) and the perceived transmissibility through needle sharing scale (P = 0.03). Of the total study sample, approximately 15–30% expressed experiencing fatigue with practicing safer sex and 10–40% expressed experiencing safer injection fatigue. HIV-seropositive IDU had significantly higher scale scores than HIV-seronegative individuals for both safer sex fatigue (P = 0.02) and safer injection fatigue (P = 0.006). Comparisons were also made across HIV treatment groups (HAART, non-HAART treatment and no treatment) among the HIV seropositive participants but no differences were observed in any of the scale scores. Due to the differences in attitudes between HIV-seropositive and -seronegative participants, and the fact that HIV-seropositive individuals generally have a different HIV-related risk behavior profile, stratified analyses were conducted in identifying attitudinal factors associated with high-risk behaviors.
HIV treatment-related attitudes and high-risk behaviors
Table 1 shows the characteristics of participants included in the analyses examining the association between the HIV treatment-related attitudes and high-risk behaviors. The univariate analyses identifying factors associated with unprotected sex (Table 3) show that, among HIV-seronegative participants who were engaging in sexual intercourse, those who injected drugs more than daily [odds ratio (OR), 1.85; 95% confidence interval (CI), 1.04–3.30], shared needles (OR, 1.89; 95% CI, 1.02–3.49). Additionally, those with high scores (i.e., > 3.0) indicating safer sex fatigue (OR, 2.84; 95% CI, 1.14–7.09) were significantly more likely to have unprotected sex, compared to those with low scores (i.e., < 2.0). Belief in reduced HIV transmissibility through unprotected sex was not significantly associated with unprotected sex. In the multivariate model, safer sex fatigue was independently associated with unprotected sex [adjusted odds ratio (AOR), 3.12; 95% CI, 1.17–8.35].
Among the HIV-seropositive IDU, the univariate model in Table 3 shows that being female (OR, 4.87; 95% CI, 2.27–10.44), those who injected more than daily (OR, 3.24; 95% CI, 1.38–7.64), and shared needles (OR, 2.82; 95% CI, 1.26–6.31) were significantly more likely to engage in unprotected sex. African–Americans (OR, 0.18; 95% CI, 0.03–0.97) were significantly less likely to engage in unprotected sex. Additionally, those with the highest level of safer sex fatigue (OR, 4.00; 95% CI, 1.54–10.40) and those with moderate (i.e., 2.0 ≤ score ≤ 3.0) and high scores indicating belief in reduced HIV transmissibility through unprotected sex (OR, 2.42; 95% CI, 1.10–5.33 and OR, 7.75; 95% CI, 2.13–28.14, respectively) were significantly more likely to engage in unprotected sex compared to those in the lowest levels of the respective scales. In the multivariate model, factors that were independently associated with unprotected sex were being female (AOR, 7.00; 95% CI, 3.00–18.88), CD4 cell count between 201 and 500 × 106/l (AOR, 0.16; 95% CI, 0.04–0.63), and moderate belief of reduced HIV transmissibility through unprotected sex (AOR, 3.33; 95% CI, 1.05–10.55). Although it failed to attain statistical significance, a dose-response was observed in the association between belief in reduced HIV transmissibility and unprotected sex; those who had the strongest belief in reduced HIV transmissibility had a five-fold increased odds of engaging in unprotected sex (AOR, 4.92; 95% CI, 0.97–24.91) relative to those with the weakest belief in reduced HIV transmissibility.
Table 4 shows results from univariate and multivariate models of needle sharing among participants injecting drugs. Among HIV-seronegative IDU, high scores indicating safer injection fatigue (OR, 4.85; 95% CI, 2.42–9.71), and moderate scores indicating belief in reduced HIV transmissibility through needle sharing (OR, 1.75; 95% CI, 1.05–2.91) were significantly associated with needle sharing compared to those with low scores on the respective scales. In the multivariate model, only safer injection fatigue remained independently associated with needle sharing. Those who scored the highest on the safer injection fatigue scale were five times more likely to share needles (AOR, 5.15; 95% CI, 2.33–11.37) relative to those scoring low on the scale.
Among the HIV-seropositive IDUs, univariate models indicated that only a high level of safer injection fatigue (OR = 4.20; 95% CI, 1.46–12.09) was significantly associated with needle sharing. In the multivariate model, safer injection fatigue remained the only factor independently associated with needle sharing. Those who scored the highest on the safer injection fatigue scale were over six times more likely to share needles (AOR, 6.55; 95% CI, 1.69–25.39).
The results from this study indicate that the use and knowledge of HAART appears to have had an impact on IDUs’ perceptions of HIV transmissibility among HIV-infected IDU. Further, safer injection and safer sex fatigue, which were associated with high-risk sexual and injection behaviors, were common in this drug injecting population.
Attitudes that sexual transmission of HIV is less likely to occur because of current HIV treatments and consequent undetectable viral loads was an important factor associated with unprotected sex among HIV-seropositive individuals in this study. This finding supports current concerns raised by the public health community that high-risk populations have become complacent about behaviors that reduce the risk of transmission of HIV and other blood-borne pathogens.
One of the most striking findings of this study was the differential finding among HIV-seropositive and -seronegative participants in the association between beliefs about HIV transmissibility and unprotected sex. The belief in reduced HIV transmissibility through unprotected sex was strongly associated with unprotected sex among HIV-seropositive individuals, but no association was observed among HIV-seronegative IDU. This suggests that HAART may have critically influenced sexual behaviors among those living with HIV. These results confirm the findings of Ostrow and colleagues in a gay/bisexual population , which has experienced a more significant impact of HIV treatments due to greater awareness and higher utilization of HAART. In that study, ‘reduced HIV concern,’ which measured personal concern about having unsafe sex and assumptions of HIV transmissibility in light of the availability of combination therapies and undetectable viral load, was independently associated with unprotected sex for both HIV-seropositive and -seronegative men. However, a dramatically stronger association existed for HIV-seropositive than for -seronegative men [6,7]. The lack of association among the HIV-seronegative individuals in our study sample may be due to insufficient personal knowledge about current treatments. It is plausible that the IDU in our study are less familiar with HAART than gay men, and thus, less susceptible to HAART-influenced risk perceptions and associated behavior change. This finding points to the need for intensified risk reduction counseling for HIV-infected individuals that focuses on their own HIV transmission risk and that specifically address safer sexual and injection practices in the context of HAART. While there is evidence for reduced likelihood of HIV transmission through unprotected sex [16,17], the potential for sexual transmission remains [18,19].
Although approximately 10–20% of the study population believed that potent HIV treatments reduced HIV transmissibility through needle sharing, such a perception was not associated with needle sharing among both HIV-seropositive and -seronegative individuals.
Among HIV-seronegative individuals, only safer sex fatigue was independently associated with unprotected sex. Similarly, safer injection fatigue was the only factor independently associated with needle sharing among HIV-seronegative IDU. This finding suggests that issues pertaining to risk-reduction fatigue may be an essential component to comprehensive harm reduction activities involving IDU populations, especially given that safer sex and injection fatigue was fairly common even in the HIV-seronegative population (10–40%).
Another important result of the study was the dramatically increased odds of HIV-seropositive women (AOR, 7.00) engaging in unprotected sex compared to HIV-seropositive men. One potential explanation is that the prevalence of trading money or drugs for sex is higher in female study participants. In our sample of IDU, women were three times more likely to sell sex for drugs or money than were the men (P < 0.001). Additionally, women may be less likely to disclose their HIV status to partners because of fear of the physical, emotional, and financial implications of disclosure [20,21]. As a result, they may find themselves in situations where they are engaging in unprotected sex as a method of avoiding disclosure. Furthermore, there is evidence that women with abusive partners report significantly less condom use . Lastly, women may be more likely to lack the skills necessary to negotiate safer sex with their partner than are the men.
This analysis also found a significant decreased odds of unprotected sex among HIV-seropositive African–Americans (AOR, 0.11), compared to HIV-seropositive non-African–Americans. This may be due to African–Americans having a greater number of sex partners than non-African–Americans, which has been reported in other studies . Further, findings from a study of Baltimore drug users suggests that those with casual sex partners and in less secure relationships were more likely to report engaging in protected sex . This may help to explain why we see a significant decreased odds of unprotected sex in the present study.
The finding that HIV-seropositive individuals who were more immunocompromised (CD4 cell count < 200 × 106/l) were more likely to engage in unprotected sex is most likely due to the cross-sectional nature of this analysis. Univariate analysis among HIV-seropositive individuals showed that those having unprotected sex were significantly more likely to be injecting drugs more often than daily compared to those practicing safe sex. Studies have shown that current injectors were less likely to be on HAART [25,26], and may be less likely to be adherent to the HIV treatment regimens , resulting in poorer clinical, immunological and virological outcomes . Thus, frequent injectors may be more likely to have lower CD4 cell counts, either as a result of not receiving HIV treatment or having a poorer adherence and immunological response to HIV treatment. Hence our findings most likely reflect the fact that poorer immunological status is a function of a riskier lifestyle.
Our study had several limitations that must be considered. First, due to the cross-sectional nature of the analysis, we cannot infer causal inferences about HIV treatment-related attitudes and high-risk behaviors. It is difficult to discern whether the behavior is subsequent to attitudes, or vice versa. A study examining the relationship between beliefs about HIV treatment and risky sexual behavior among gay and bisexual men suggests that belief in reduced HIV transmissibility may partially be a result of the unsafe behavior . Using mediation analysis, they showed a path from high-risk sexual behavior to feeling of increased susceptibility, and subsequently to belief in reduced HIV transmissibility. This finding is consistent with the theory of cognitive dissonance, which suggests that contradicting behaviors and attitudes can result in change in beliefs (i.e., rationalizing engaging in unsafe behavior by believing that HIV treatments reduce the risk of HIV transmission). While a longitudinal analysis could shed light on the causal effects, regardless of the directionality of the relationship, it remains important for prevention efforts to address the deleterious belief that HIV is less transmissible due to HIV treatments. Second, our categorization of the attitude scales may have caused some misclassification of people's attitudes. However, the use of categories yielded OR with dose response effects for both scales, which suggests that misclassification was minimal. Third, the inclusion of the response category ‘neither disagree or agree’ as a ‘disagree’ response could have affected our results. However, the percentage of respondents who responded ‘neither disagree or agree’ to the questionnaire items was minimal, ranging from 1 to 5%. Further, when results were re-calculated including ‘neither disagree or agree’ as an ‘agree’ response, this did not affect inferences made from results. Fourth, sexual and injection behavior data were based on self-report, and might be subject to some bias. However, there is evidence that self-report of risk behaviors among IDU generally is valid . Lastly, IDU may be less aware of antiretroviral therapies and therefore the impact of HIV treatments may not be sufficient to affect HIV-related risk behaviors. It will be important to assess changing attitudes and impact on behavior as therapies become more widely prescribed in this population.
Despite these limitations, some important implications can be drawn from these results. A critical implication is the need for HIV prevention programs to address issues of risk-reduction fatigue. This is a crucial issue especially for HIV-seropositive individuals who must maintain safe behaviors for the rest of their lives or risk transmission to others. Further investigation of correlates or mediating factors of safe sex and injection fatigue may shed light on potential modes of intervention to help individuals cope with such fatigue. For safe injection fatigue, the most effective intervention would be drug abuse treatment.
While many factors beyond those examined in this analysis may be associated with risky sexual and injection practices, the associations seen here indicate that HIV treatments have had an impact on people's attitudes regarding HIV transmission. This is particularly true for individuals living with HIV. Given that HIV-infected individuals are living healthier and longer lives due to HIV treatments, interventions need to focus on educating high-risk populations of the need for continued safer sexual and injection practices regardless of effective HIV treatments. Additionally, issues of risk-reduction fatigue need to be initiated by clinicians who provide care to HIV-infected individuals. HIV treatment-related attitudes will need to be considered in future research on HIV-related risk behavior and prevention strategies, as we have demonstrated that they are associated with risky sexual and injection practices.
The authors thank D. Ostrow and P. Vanable for the development of the original scale, The ALIVE Study interviewers for their thoughtful suggestions to the questionnaire, and J. Vertefeuille and I. Kuo for their insightful comments on the manuscript.
Sponsorship: Supported by grant DA04334 and DA12568 from the National Institute on Drug Abuse, and in part by a National Research Service Award 1-F31-MH12660 to Waimar Tun from the National Institute of Mental Hygiene.
1. Detels R, Tarwater P, Phair JP, Margolick J, Riddler SA, Munoz A. Effectiveness of potent antiretroviral therapies on the incidence of opportunistic infections before and after AIDS diagnosis. AIDS 2001, 15:347–355.
2. Vittinghoff E, Scheer S, O'Malley P, Colfax G, Holmberg SD, Buchbinder SP. Combination antiretroviral therapy and recent declines in AIDS incidence and mortality. J Infect Dis 1999, 179:717–720.
3. Falck RS, Siegal HA, Wang J, Carlson RG. Usefulness of the health belief model in predicting HIV needle risk practices among injection drug users. AIDS Educ Prevent 1995, 7: 523–533.
4. Longshore D, Anglin MD. Intentions to share injection paraphernalia: an empirical test of the AIDS risk reduction model among injection drug users. Int J Addict 1995, 30:305–321.
5. Kalichman SC, Rompa D, Austin J, Luke W, DiFonzo K. Viral load, perceived infectivity, and unprotected intercourse. J Acquir Immune Defic Syndr 2001, 28:303–305.
6. Ostrow DG, Fox K, Chmiel JS, Silvestre A, Visscher BR, Jacobson LP, et al. Attitudes towards highly active antiretroviral therapy are associated with sexual risk taking among HIV-infected and uninfected homosexual men. AIDS 2002, 16:1–6.
7. Vanable PA, Ostrow DG, McKirnan DJ, Taywaditep KJ, Hope BA. Impact of combination therapies on HIV risk perceptions and sexual risk among HIV-positive and HIV-negative gay and bisexual men. Health Psychol 2000, 19:134–145.
8. Katz MH, Schwarcz SK, Kellogg TA, Klausner JD, Dilley JW, Gibson S, et al. Impact of highly active antiretroviral treatment on HIV seroincidence among men who have sex with men: San Francisco. Am J Public Health 2002, 92:388–394.
9. Miller M, Meyer L, Boufassa F, Persoz A, Sarr A, Robain M, et al. Sexual behavior changes and protease inhibitor therapy. SEROCO Study Group. AIDS 2000, 14:F33–F39.
10. Lehman J, Hecht F, Wortley P, Lansky A, Stevens M, Fleming P. Are at-risk populations less concerned about HIV infection in the HAART era? Seventh Conference on Retroviruses and Opportunistic Infections. San Francisco, January–February 2000 [abstract 198].
11. Centers for Disease Control and Prevention. HIV/AIDS Surveillance Report, 12(2). Atlanta: CDC; 2000.
12. Vlahov D, Safaien M, Lai S, Strathdee SA, Johnson L, Sterling T, et al. Sexual and drug risk-related behaviours after initiating highly active antiretroviral therapy among injection drug users. AIDS 2001, 15:2311–2316.
13. Bouhnik AD, Moatti JP, Vlahov D, Gallais H, Dellamonica P, Obadia Y. Highly active antiretroviral treatment does not increase sexual risk behaviour among French HIV infected injecting drug users. J Epidemiol Commun Health 2002, 56: 349–353.
14. Vlahov D, Anthony JC, Munoz A, Margolick J, Nelson KE, Celentano DD, et al. The ALIVE study, a longitudinal study of HIV-1 infection in intravenous drug users: description of methods and characteristics of participants. NIDA Res Monogr 1991, 109:75–100.
15. Tun W. HIV-related risk behaviors and attitudes towards HIV treatment in the era of highly active antiretroviral therapy among injection drug users. [Dissertation] Baltimore: Johns Hopkins Bloomberg School of Public Health; 2002.
16. Gray RH, Wawer MJ, Brookmeyer R, Sewankambo NK, Serwadda D, Wabwire-Mangen F, et al. Probability of HIV-1 transmission per coital act in monogamous, heterosexual, HIV-1-discordant couples in Rakai, Uganda. Lancet 2001, 357:1149–1153.
17. Quinn TC, Wawer MJ, Sewankambo N, Serwadda D, Li C, Wabwire-Mangen F, et al. Viral load and heterosexual transmission of human immunodeficiency virus type 1. Rakai Project Study Group. N Engl J Med 2000, 342:921–929.
18. Mayer KH, Boswell S, Goldstein R, Lo W, Xu C, Tucker L, et al. Persistence of human immunodeficiency virus in semen after adding indinavir to combination antiretroviral therapy. Clin Infect Dis 1999, 28:1252–1259.
19. Zhang L, Ramratnam B, Tenner-Racz K, He Y, Vesanen M, Lewin S, et al. Quantifying residual HIV-1 replication in patients receiving combination antiretroviral therapy. N Engl J Med 1999, 340:1605–1613.
20. Gielen AC, Fogarty L, O'Campo P, Anderson J, Keller J, Faden R. Women living with HIV: disclosure, violence, and social support. J Urban Health 2000, 77:480–491.
21. Gielen AC, McDonnell KA, Burke JG, O'Campo P. Women's lives after an HIV-positive diagnosis: disclosure and violence. Matern Child Health J 2000, 4:111–120.
22. Wingood GM, DiClemente RJ. The effects of an abusive primary partner on the condom use and sexual negotiation practices of African-American women. Am J Pub Health 1997, 87:1016–1018.
23. Wohl AR, Lu S, Odem S, Sorvillo F, Pegues CF, Kerndt PR. Sociodemographic and behavioral characteristics of African-American women with HIV and AIDS in Los Angeles County, 1990–1997. J Acquir Immune Defic Syndr Hum Retrovirol 1998, 19:413–420.
24. Sherman SG, Latkin CA. Intimate relationship characteristics associated with condom use among drug users and their sex partners: a multilevel analysis. Drug Alcohol Depend 2001, 64:97–104.
25. Andersen R, Bozzette S, Shapiro M, St.Clair P, Morton S, Crystal S, et al. Access of vulnerable groups to antiretroviral therapy among persons in care for HIV disease in the United States. HCSUS Consortium. HIV cost and services utilization study. Health Serv Res 2000, 35:389–416.
26. Celentano DD, Galai N, Sethi AK, Shah NG, Strathdee SA, Vlahov D, et al. Time to initiating highly active antiretroviral therapy among HIV- infected injection drug users. AIDS 2001, 15:1707–1715.
27. Gebo K, Moore RD. Do social stressors, heath beliefs, and illicit drug use correlate with adherence to antiretroviral therapy. Annual Infectious Diseases Society of America (IDSA) Meeting. New Orleans, September 2000 [abstract 360].
28. Tuldra A, Fumaz CR, Ferrer MJ, Bayes R, Arno A, Balague M, et al. Prospective randomized two-Arm controlled study To determine the efficacy of a specific intervention To improve long-term adherence to highly active antiretroviral therapy. J Acquir Immune Defic Syndr 2000, 25:221–228.
29. Huebner DM, Gerend MA. The relation between beliefs about drug treatments for HIV and sexual risk behavior in gay and bisexual men. Ann Behav Med 2001, 23:304–312.
30. Darke S. Self-report among injecting drug users: a review. Drug Alcohol Depend 1998, 51:253–263.
HIV; antiretroviral therapy; attitudes; sexual behavior; intravenous substance abuse
© 2003 Lippincott Williams & Wilkins, Inc.
Highlight selected keywords in the article text.