Sex Work and Male-With-Male Sex Sharply Raise HIV Risk for People Who Inject Drugs, Global Analysis Finds

People who inject drugs already face some of the highest rates of HIV and hepatitis C virus (HCV) infection in the world, but a sweeping new global analysis shows that those risks climb dramatically when injection drug use overlaps with sex work or sexual activity between men. The findings, drawn from a systematic review and meta-analysis published in eClinicalMedicine, challenge the long-standing practice of treating sex workers, men who have sex with men, and people who inject drugs as separate populations for prevention planning, and argue instead for integrated services that address the full spectrum of overlapping risks.

The research team, led by Joshua Dawe of the University of Bristol together with collaborators across multiple countries, built their analysis on an established database of cohort studies that estimate new infections among people who inject drugs. Their searches covered literature published between January 1, 2000 and June 9, 2026, spanning MEDLINE, Embase, and PsycINFO. After screening more than 82,000 records and re-reviewing 436 studies that reported HIV or HCV incidence, the investigators identified 62 studies examining whether sex work or male-with-male sexual activity modified infection risk. Crucially, they also contacted the authors of 36 studies to request unpublished estimates, achieving a 67 percent response rate, which brought the final pool to 57 studies contributing 164 estimates. Notably, 70 percent of those estimates were unpublished, a deliberate strategy to reduce publication bias and strengthen statistical power.

The headline result concerns HIV acquisition among men who inject drugs and also report recent male-with-male sexual activity. Across 13 studies, the pooled incidence rate ratio was 2.35, meaning these men acquired HIV at more than twice the rate of men who inject drugs without recent male-with-male sex. The 95 percent confidence interval ran from 1.51 to 3.66, and the estimate grew slightly stronger in adjusted analyses at 2.66. Lifetime male-with-male sexual exposure told a similar story, with a pooled crude incidence rate ratio of 2.44 for HIV acquisition. By contrast, the analysis found no statistical evidence linking male-with-male sex to HCV acquisition, with a pooled incidence rate ratio of 1.28 for recent exposure whose confidence interval crossed the null. The authors point to transmission biology as a likely explanation: HCV spreads far more efficiently through blood-borne exposure than through sexual contact, so a sexual behavior alone would not be expected to drive HCV incidence the way it drives HIV incidence.

Sex work showed a different and, in some ways, more complicated pattern. Among mixed-sex samples of people who inject drugs, recent sex work was associated with a 1.72-fold higher rate of HIV acquisition across 20 studies, and a 1.59-fold higher rate of HCV acquisition across 22 studies, with both associations persisting in adjusted analyses. But the sex-stratified results revealed striking asymmetries. For HIV, the association was strong among women who inject drugs, at an incidence rate ratio of 2.30, while no evidence of an association emerged among men, at 1.43. For HCV, the pattern reversed: recent sex work was associated with elevated risk among men who inject drugs, at 1.51, but not among women, at 1.26. These divergent findings suggest that different mechanisms operate for different outcomes and populations, and that a single pooled estimate can obscure risks that are concentrated in specific subgroups.

Quantifying how common these overlapping behaviors are underscores why the risk findings matter. The pooled analysis estimated that around one in nine people who inject drugs, or 10.6 percent, had recently engaged in sex work, with the figure far higher among women, at 27.2 percent, than among men, at 4.0 percent. Among men who inject drugs, roughly 4.8 percent reported recent male-with-male sex, while 21.9 percent reported lifetime exposure. In other words, a substantial fraction of the population that harm reduction services serve also belongs to populations traditionally served by separate sexual health programs, yet the services themselves rarely overlap.

Methodologically, the review took care to focus on recent exposure, defined as within the past 12 months, rather than lifetime exposure, because recent behavior aligns more closely with the period at risk for incident infections. The primary analyses used crude rather than adjusted estimates, a choice the authors justify on the grounds that factors commonly adjusted for, such as injecting frequency, housing instability, or sexual practices, may lie on the causal pathway between sex work or male-with-male sex and infection, meaning that adjustment could attenuate the very effects the analysis sought to detect. Adjusted estimates were examined as secondary analyses and generally produced similar pooled effects. Study quality was assessed with a modified Newcastle-Ottawa Scale, and funnel plots together with Egger’s tests found little evidence of publication asymmetry, with one exception among published studies of male-with-male sex and HIV.

The mechanisms behind the elevated risks are likely to be structural as much as behavioral, the authors argue. Sex workers who inject drugs are more likely to engage in street-based or survival sex work, which is associated with limited condom negotiation, rushed or unsafe injections, more frequent encounters with law enforcement, and greater exposure to physical and sexual violence. Criminalization of both sex work and drug use in most countries intensifies policing and incarceration, which disrupt access to health care and, as prior reviews from the same group have shown, independently elevate HIV and HCV risk. For men who have sex with men who use drugs, prior studies indicate higher rates of condomless anal sex and participation in sexualized drug use, including chemsex, which can involve multiple partners and injection practices that heighten transmission risk. Stigma related to sexuality and sex work further deters engagement with health services, compounding vulnerability.

The heterogeneity across studies was moderate to high for most pooled estimates, and subgroup analyses did not fully explain it. Some patterns were nonetheless informative. The association between sex work and HIV appeared stronger in studies with higher proportions of women, higher levels of recent incarceration, and higher-income settings, while no clear association emerged in low- and middle-income settings, though the authors caution that most subgroup estimates rested on small numbers of studies with wide confidence intervals. Intriguingly, the association between recent sex work and HIV was substantially larger in low-incidence settings, with an incidence rate ratio of 2.38 in studies where background HIV incidence was below two cases per 100 person-years, compared with 1.24 in higher-incidence settings, a pattern the authors partly attribute to the compression of relative measures when baseline risk is high.

The review acknowledges several limitations. Most estimates came from high-income countries, limiting generalizability to low- and middle-income settings where evidence remains scarce. Exposure was based on self-report and may be underreported where behaviors are heavily stigmatized, and in four studies lifetime male-with-male exposure was defined by sexual identity rather than behavior, potentially misclassifying men who have sex with men but identify as heterosexual. Transgender people who inject drugs could not be analyzed separately because insufficient data were available. Unpublished estimates, while valuable for reducing publication bias, did not undergo formal peer review, and responding cohorts may differ systematically from non-responding ones.

The central message for public health is a call to dismantle the silos that organize HIV and HCV prevention around single populations or behaviors. The authors advocate integrated, low-threshold services that combine harm reduction measures, including sterile injecting equipment and opioid agonist therapy, with sexual health care such as pre-exposure prophylaxis, condoms, sexually transmitted infection testing, and reproductive health services. Routine assessment of both injection-related and sexual risks, with prevention tailored to individuals rather than to behavioral categories, could reach people whose overlapping vulnerabilities are currently missed. They also emphasize the need for research into the factors driving elevated risk among people who belong to more than one affected population, particularly in low- and middle-income countries, so that interventions can be designed around the real, intersecting lives of the people they aim to protect.

Subject of Research: HIV and hepatitis C acquisition risk among people who inject drugs who also engage in sex work or male-with-male sexual activity

Article Title: Sex work or male-with-male sexual activity and the risks of HIV and hepatitis C acquisition among people who inject drugs: a global systematic review and meta-analysis

Article References: Sex work or male-with-male sexual activity and the risks of HIV and hepatitis C acquisition among people who inject drugs: a global systematic review and meta-analysis. (n.d.). Original publication

Image Credits: AI Generated

DOI: Not provided

Keywords: HIV, hepatitis C, people who inject drugs, sex work, men who have sex with men, systematic review, meta-analysis, harm reduction, incidence rate ratio, public health, stigma, integrated services