HIV-negative white men who have sex with men are more likely than their black counterparts to believe they know the HIV status of their most recent partner, according to an analysis of survey data from Atlanta.1 While 88% of white respondents thought they knew the HIV status of their most recent male partner, only 78% of black men did. However, whites were more likely than blacks to say they would be willing to have unprotected anal sex with a partner they believed to be HIV-negative; indeed, among respondents who had had anal sex during their last sexual encounter, the proportion of whites who had used a condom was smaller than the proportion of blacks who had done so (45% vs. 73%). Overall, white respondents had had unprotected anal sex with a greater number of HIV-negative partners in the past six months than had blacks, regardless of whether they were the receptive or insertive partner, while blacks had had a greater number of partners of unknown HIV status than had whites.

The prevalence of HIV is substantially higher among black than white men who have sex with men. Researchers hypothesized that this racial disparity might be driven by differences in serosorting—the practice of choosing partners whose HIV status is the same as one’s own, which some men feel precludes the need to use condoms—or in beliefs about this approach. To explore this possibility, investigators surveyed 1,051 men attending two gay pride festivals in Atlanta in 2006. The self-administered questionnaire, which respondents completed anonymously, included questions about men’s demographic characteristics, partners in the past six months and risk behaviors during their most recent sexual encounter. It also asked about their serosorting beliefs (e.g., the extent to which their willingness to have unprotected sex depended on a potential partner’s HIV status) and their serostatus disclosure beliefs (e.g., their comfort level with disclosing their HIV status before having sex). Participants were asked to estimate their HIV risk by thinking about their sexual behavior during the past six months and marking their risk on a 248-millimeter visual scale ranging from having no risk (e.g., being abstinent) to having an extremely high risk (e.g., having unprotected receptive anal sex with an HIV-positive partner). The researchers compiled descriptive statistics and conducted logistic regression analyses to compare outcomes and associations by race. No HIV testing was performed, so the reported status of participants and their partners reflected participants’ beliefs and may not have been accurate.

The analysis focused on responses from 549 HIV-negative men, of whom 67% were white and 33% black. Compared with black respondents, whites were older (mean, 35 vs. 32 years) and more educated, had a higher income and were more likely to be in an exclusive sexual relationship (47% vs. 33%). On average, more than a year had elapsed since respondents’ most recent HIV test.

In the six months prior to the survey, blacks had had insertive or receptive anal intercourse with a greater number of partners than had white men (mean, 5.1 vs. 3.6), but the number of HIV-positive partners with whom participants had had unprotected anal sex did not differ by race. Whites had had unprotected anal sex with a greater number of HIV-negative partners in the past six months than had blacks, regardless of whether they were the receptive partner (1.0 vs. 0.5) or the insertive partner (0.8 vs. 0.6). Blacks, on the other hand, had engaged in unprotected anal intercourse with a greater number of partners of unknown HIV status than had whites, both for receptive sex (0.2 vs. 0.1) and for insertive sex (0.3 vs. 0.1).

The proportion of respondents who said they knew the HIV status of their last sexual partner was lower among blacks than among whites (78% vs. 88%). Five percent of whites and 8% of blacks said their most recent partner had been HIV-positive. The proportion of men who had had anal intercourse during their most recent sexual encounter was similar among blacks (68%) and whites (67%); however, among those who had had anal sex, the proportion who had used a condom was lower among whites than among blacks (45% vs. 73%).

Whites reported greater ease with disclosure of their HIV status than blacks (mean score, 5.1 vs. 4.8 on a six-point scale), as well as more positive attitudes toward serosorting (2.6 vs. 2.3, also on a six-point scale). Whites’ estimates of their HIV risk were lower than those of blacks—53 out of a possible 248 on the visual analog scale, versus 71 for blacks.

Finally, results of the multivariate regression analysis largely mirrored those of the univariate analyses. Compared with whites, blacks had lower odds of being in a committed relationship, reported a higher mean perceived HIV risk, and had had unprotected insertive sex with a lower number of HIV-negative partners but a higher number of partners of unknown HIV status. Beliefs concerning disclosure of HIV status did not vary by race.

The researchers caution that their data are self-reported and cross-sectional, and that the findings may not be generalizable to men in other locales. Nonetheless, they suggest that racial differences in “partner selection strategies”—in particular, black men’s elevated likelihood of having sex with partners whose HIV status is unknown—“may explain [racial] differences in HIV infection rates.” Although black men were more likely than whites to report condom use, this protective behavior may not compensate for the risk that results from having sex with partners of unknown HIV status. More information on how black men who have sex with men “strategically protect themselves (or do not protect themselves) from HIV” is needed to tailor interventions more effectively.

R. Hutchinson