Sexual HIV transmission in the context of injection drug use: implications for interventions

Abstract

In recent years, there has been increased awareness of sexual HIV transmission between injection drug users (IDUs) and from IDUs to the ‘‘general’’ population. This commentary will highlight both aspects of these HIV transmission dynamics, with special attention on the implications for interventions and future research. It is argued that, based on the existing literature, a narrow focus on injection-related risks is an ineffective prevention strategy, regardless of the stage of the HIV epidemic. Interventions that target specific subgroups warrant attention of high-risk IDUs such as those who are sex workers, MSM, incarcerated and/or HIVinfected. Future research should expand efforts in behavioural surveillance to keep abreast of potential changes in HIV/STD transmission dynamics and inform efforts to guide policies. In early landmark studies of IDU populations, most HIV infections appeared to be due to parenteral risks (Des Jarlais, Friedman & Hopkins, 1985; Chaisson et al., 1987; Schoenbaum et al., 1989), leading to the presumption that sexual risks among IDUs were negligible. Other studies suggested that sexual risks were present, but were overshadowed by parenteral risks (van Ameijden & Coutinho, 1998; Marx et al., 1991; Battjes et al., 1990). Recent studies have shown otherwise. In particular, in a 10 year prospective study among IDUs in Baltimore where HIV prevalence was initially high ( /25%), predominant risk factors among males included not only needle-sharing, but sex with men. Among females, factors consistent with high risk heterosexual activity (e.g. having a recent STD and an IDU sex partner) outweighed drug-related risks (Strathdee et al., 2001). Results based on a series of crosssectional studies among young IDUs who were new initiates to injection drug use in Baltimore were similar, suggesting that sexual risks may be important throughout the span of an injection drug use career (Strathdee et al., 2002). Moreover, in a prospective study of IDUs in San Francisco, both homosexual activity and female sex work were associated with HIV seroconversion (Kral et al., 2001), which underscores the consistency of these findings in both high and low HIV prevalence settings. Interestingly, in the latter study, females with IDU sex partners were less likely to seroconvert than other female IDUs, suggesting that the context of sexual partnerships involving IDUs likely influences the degree of risk. Surprisingly, interventions focused on IDUs have tended to overlook sexual risks. In a 1996 survey of needle exchange programmes in the US, Paone et al. reported that only 18% offered on-site STD testing and medical referrals (Paone et al., 1999). Similarly, substance abuse treatment programmes have seldom had a dual focus on injection and sexual risk reduction, which has likely accounted at least in part for reported inconsistencies in study findings of the impact of these interventions on sexual risk behaviours (MacGowan et al., 1997; Lollis et al., 2000; Avants et al., 2000). In a follow-up study of clients enrolled in methadone maintenance programmes in two US states, inconsistent condom use was significantly associated with enrollment in programmes where condoms were available only upon request and abstinence and monogamy between uninfected partners were promoted (MacGowan et al., 1997). These examples illustrate the extent to which a broader, integrated approach to risk reduction is needed in such settings. A recent meta-analysis of 33 intervention studies focused on reducing sexual risks among drug users found a 40% risk reduction overall, which justifies the need to incorporate these interventions into existing programmes that target high risk drug users (Semaan et al., 2002). * Tel.: /1-410-614-4255; fax: /1-410-955-1383. E-mail address: sstrathd@jhsph.edu (S.A. Strathdee). International Journal of Drug Policy 14 (2003) 79 /81