This is the first study to report on gender differences in HIV risk behaviors among opioid-dependent youth receiving outpatient bup/nx treatment, and to examine changes in drug and sexual behaviors as a function of gender and treatment. The majority of males and females in this sample reported one or more HIV risk behaviors in the month prior to treatment, including injection risk (e.g., sharing needles, splitting drug solution), unprotected intercourse, and multiple sex partners. These behaviors place opioid-dependent youth at high risk for infection with HIV and other diseases. This is especially alarming given that 18% of this sample were seropositive for hepatitis C virus at baseline, and 4 of the 83 participants (5%) who were seronegative at baseline converted to seropositive status by week 12.41
Over the course of treatment, there were significant reductions in opioid use and IDU in both treatment conditions, but these reductions were greater for participants in BUP. By the end of treatment, only 15% of BUP participants were injecting compared to 35% of DETOX participants. This finding of decreased opioid use and IDU associated with bup/nx treatment is consistent with previous trials conducted among adults10, 40, 43
Given that opioid-dependent youth are at such high risk for relapse to drug use,48
these results provide further support that bup/nx treatment is a safe and effective option that should be considered. However, among participants who continued to inject drugs, there was no reduction in injection risk over time. This suggests that extended bup/nx therapy alone maybe insufficient for stopping behavioral risk among persistent drug injectors.
While males and females were equally likely to be injectors, females were significantly more likely to engage in injection risk behavior. For example, in the month prior to enrollment, 77% of female injectors compared to 35% of male injectors reported injection risk. Specifically, females were more likely to share a cooker, cotton, or rinse water and to fix drugs together and split the solution. These types of injection risk behaviors may be an increasingly important route of HIV and hepatitis C transmission.49
While further research is needed to better understand this gender difference, studies with adult injectors suggest that intimate partnerships may play a role. Women are more likely than men to attribute their initiation to and continued use of heroin to social reasons, particularly the influence of an opioid-using partner.50–52
Furthermore, the primary reason injectors report sharing equipment is the use of drugs with sex partners.52–54
Qualitative and social network research with young drug users may help elucidate the interpersonal dynamic and context associated with injection risk in females.
Interestingly, females responded particularly well to extended bup/nx therapy. Females in BUP had larger declines in IDU compared to females in DETOX, whereas males in both groups showed only modest declines. The cause of this gender difference is unclear, but it cannot be explained by reductions in opioid use, as no gender difference in opioid use was observed. Due to social and economic disparities, women may be more likely to depend upon male partners to obtain drugs, perpetuating a pattern of using, injecting, and sharing drugs with them. Young females may be particularly vulnerable to such dependency. The physiological benefits of extended bup/nx therapy may, for some females, weaken this dependency, allowing them to change their patterns of drug use, including less IDU. Future research should test this and other potential explanations.
In terms of sexual risk, males and females were equally likely to be sexually active and not use condoms, but males were more likely to have multiple partners. This gender difference is consistent with nationally representative surveys of adolescents and young adults.55, 56
Having multiple sex partners, particularly concurrent partners, is associated with sexually transmitted infections,57, 58
and this subgroup of young males with multiple sex partners may play an important role in the transmission of HIV and other infections among drug-using youth.
As in other studies of treatment-seeking opioid abusers,43, 59
there was a slight drop in the rate of sexual activity during treatment. It is unclear from our results whether this was a deliberate attempt by participants to reduce their sexual activity or a reflection of their increased focus on recovery during the early part of treatment. This change was not sustained, however, as the rate of sexual activity returned to nearly baseline levels by the end of treatment. More importantly, there was no change in the rate of multiple partners or unprotected intercourse. Many other studies have found that drug treatment is not associated with reductions in sex risk behaviors.10, 43, 46, 60
Thus, bup/nx and other treatments seem to have greater impact on drug risk and may be insufficient for promoting sex risk reduction.
While bup/nx therapy is an important component of HIV prevention for opioid-dependent youth, additional risk reduction counseling may be needed to promote greater decreases in both drug and sex risk behaviors. In one of the only randomized controlled trials to target drug using-adolescents, St. Lawrence and colleagues (2002) tested a 12-session group treatment based on the Information-Behavior-Motivation (IBM) Model among 161 adolescents in residential treatment.61
Compared to participants in the information only condition, participants who received both information and behavioral skills components had greater increases in condom use and sexual abstinence. Among adolescents in general, theoretically-based group interventions that are tailored to meet the needs of specific sub-groups of adolescents can effectively increase condom use and reduce number of sex partners.62
To the best of our knowledge, no interventions have been developed to reduce injection drug risk among adolescents, though many trials have included adolescents and young adults. For example, early in the HIV epidemic, Des Jarlais and colleagues (1992) tested the effects of a 4-session intervention based on Social Learning Theory among heroin sniffers with a mean age of 27 years.63
Participants who were randomized to the intervention rather than the control group were less likely to transition to IDU. Comprehensive reviews of the adult literature suggest that interventions can effectively reduce both drug and sex risk behaviors among IDUs.64–66
Successful programs have been theory-based and include the following components: HIV/AIDS education; assessment of personal risk and responsibility; behavioral skills training in safer sex and drug behaviors; development of intrapersonal skills (e.g., problem solving); reinforcement of positive changes; and discussion of practical and emotional issues related to HIV risk reduction.64
Thus, youth receiving bup/nx or other pharmacological treatments for opioid dependence may benefit from multi-component HIV prevention services that include cognitive-affective-behavioral skills training delivered in group formats. Clinical trials are needed to test the effectiveness of such programs among youth receiving psychopharmacological treatment.
Several limitations of this study should be noted. First, the sample size was modest, although this is the largest randomized clinical trial of bup/nx treatment for adolescents conducted to date. It is also the first to examine gender differences in and the effects of bup/nx treatment on HIV risk among opioid-dependent youth. Second, the follow-up rate was 67%, but there was no difference by treatment condition or gender, and participants who completed the week 12 assessment did not differ from those who did not on baseline demographic characteristics, substance abuse, and HIV risk behavior. Third, no HIV testing was performed and self-reported HIV risk data are subject to response bias. However, the latter remains the standard assessment method for obtaining personal data, and other studies of substance abusers have documented test-retest reliability and predictive validity of self-reported sexual and drug use behaviors.67, 68
Finally, while the use of a convenience sample of treatment-seeking volunteers raises the possibility of selection bias, the multi-site design of this study attempts to improve generalizability. Yet, results may not generalize to adolescents who are not seeking addiction treatment or are unwilling to participate in a clinical trial.