Exchange sex is a known risk factor for HIV infection in the developing world and among female drug users in the United States, but few studies have explored its impact on heterosexual men and women more broadly. In our study of high-risk heterosexuals from New York City, we found a high prevalence of exchange sex, which was linked with several sexual and substance use risk factors. Exchange sex was associated with HIV infection in both men and women, but the relationship was confounded by other behavioral risks like male-to-male sex and injection drug use. Most of those with exchange partners engaged in unprotected sex with both exchange and non-exchange partners, with important variations by payment direction.
Our estimates of exchange sex prevalence were higher than some previous studies of high-risk heterosexuals in the United States, which found around 9% of men16
exchanged sex, but similar to studies of non-injection drug users12,14
At least 2 factors may influence our high estimates. First, several prostitution strolls (public areas where sex workers solicit clients) were located in our targeted high-risk areas. Second, we defined exchange sex partners broadly, as the trading of sex for goods like money or drugs,15
which may highlight the potentially blurry distinction between exchange and casual partnerships. One illustration of this is found by comparing the mean and median number of exchange partners among sex exchangers in the past year (17.4 versus 5): While there is small group (probably commercial sex workers) with a very high number of partners influencing the average, most exchangers have relatively few exchange partners, which may be evidence of non-commercial bartering. Participants in ethnographic interviews, conducted as part of our study’s formative research phase, often said less formal “sex-for-presents” relationships (which fall under our study definition of exchange sex) between older men and younger women were common in their communities. Research in Africa has explored the complexities of exchanging goods in main and long-term casual partnerships.5
An important age differential emerged from our analysis. In men, exchange sex was most likely in men aged 40–50 years old, while in women it was most likely those aged 30–39. This is also reflected in our finding that those who only sold sex were significantly more likely to be female and also have an exchange partner at last sex who was 5 or more years older. This reflects the traditional dynamic of older men paying younger women for sex, which presents an HIV risk to women since age gaps signify partnerships with older men who have accumulated more lifetime partners.28
Our findings on payment directionality by gender were consistent with studies in the United States, Africa, and elsewhere: most who only bought sex were men (93%) and most who only sold sex were women (91%).29
Previous research has largely focused on these traditional gender dynamics in sex exchange,16,30
and few studies have examined the overlap between buying and selling sex across gender. We found that exchangers who both bought and sold sex were a small group (12% of exchangers), but had the most sexual partners and the highest rates of unprotected sex. There were smaller gender differentials in this group (68% male versus 32% female). This again may reflect our expansive definition of exchange sex that includes informal trading outside the realm of commercial sex work. The dual directionality of payment in informal exchange partnerships has important implications. Sex ratio imbalances in urban African-American communities, due to disproportionately high rates of incarceration and mortality among young African-American men, may have led to the emergence of a market for male sexual services sold to female consumers.31–33
This is an area that deserves further research attention.
Among both men and women, exchange sex was strongly associated with high rates of multiple partnerships and unprotected sex. Similarly, a study of US men found that those with exchange partners were more likely to have multiple partners and be generally unwilling to use condoms.16
In formative ethnographic research, we found that sex workers initially said they always used condoms, but on further questioning admitted to condomless sex with long-term “steadies” who were willing to pay a premium.
This clustering of exchange sex with other behavioral risks indicates not only high levels of risk to persons with exchange partners but also community-level risks. This is illustrated in 2 findings. First, half of past-year exchangers (57%) had unprotected sex with both an exchange partner and a non-exchange partner, with significantly higher rates of this among exchangers who were paid for sex. Past research has examined how sexual activity between core risk groups (e.g., sex workers) and “bridging” groups (e.g., their non-exchange partners) has driven HIV and STD infections in the general heterosexual population in Africa and Asia.34,35
Our study points to the same phenomenon domestically. Second, men with male partners in the past year were over twice as likely to have (heterosexual) exchange partners in the same time frame. Previous research on a mixed MSM/heterosexual population has found links between bisexual identity and sex exchange among men.17
This dual heterosexual/same-sex and exchange/non-exchange overlap may be a central community-level transmission risk to otherwise low-risk members of these sexual networks.36
Men and women who used drugs or alcohol in our sample were more likely to have exchange partners, consistent with previous research studying heterosexual subgroups of substance users.13–15
Drugs are often used as a direct currency for sex, or indirectly when money exchanged for sex is used to buy drugs.37
Over half of men and women in our sample reported weekly non-injection drug use, and this was independently associated with exchange partnerships for both men and women. Nearly three quarters of participants who had an exchange partner at last sex reported drug or alcohol use during sex; concurrent substance use is generally linked to lower rates of condom use and increased risk of HIV infection.38,39
Finally, HIV prevalence was high in our study, even after removing participants with an MSM or IDU history, and nearly all of those testing positive were unaware of their status. In bivariate analyses, exchange partnerships were associated with HIV infection in both men and women, and HIV prevalence among those with exchange partners was 14.1% in men and 10.8% in women. These were similar to the prevalence levels observed among sex workers in Africa and Southeast Asia.1,40
In multiple logistic regression, the adjusted coefficients were lower when other key HIV behavioral risks were controlled for. This suggests (especially for women since the association lost statistical significance) that sex exchange may not always be an independent risk for HIV infection among heterosexual populations with high partner numbers and frequent unprotected sex. It may be impossible to disentangle exchange sex from these other risk factors in this group. Future studies should continue to investigate the confounding effects of these related behavioral risks, but our study suggests that exchange partnerships may be an HIV risk both directly and indirectly, given the overlap of this phenomenon with related behaviors.
Our study has several limitations. First, participants were asked to categorize their past year sex partners into 3 mutually exclusive groups (main, casual, and exchange), which may underestimate exchange partnerships. Second, the study design is cross-sectional and we cannot determine the time and cause of HIV infection. Third, all data except HIV infection were self-reported and are subject to the common issues with survey research, including recall error and social desirability biases. The latter may have contributed to the low prevalence of self-reported HIV infection among those who tested positive. Finally, RDS techniques for sampling and analysis are still developing, and these results are not necessarily representative of the larger target population of heterosexuals residentially or socially connected to high-risk areas in NYC.