Substantial proportions of HIV-positive adults engage in oral, anal or vaginal sex without telling their partner that they are infected, according to findings from a national survey; the risks associated with such behavior often are exacerbated by characteristics of the partnership.1 Four in 10 gay or bisexual men with HIV, and two in 10 infected heterosexual men and women, have had sex with a partner to whom they have not revealed their HIV status; for gay or bisexual men, most such sexual encounters have occurred within a nonexclusive partnership. Thirteen percent of partnerships between an infected individual and one who is HIV-negative or whose HIV status is unknown have involved unprotected intercourse without disclosure.
The data, from a national probability sample of HIV-infected men and women aged 18 or older, were gathered through in-person interviews in late 1998. At the time of interview, all respondents--606 men who identified themselves as gay or bisexual, 287 men who identified themselves as heterosexual and 504 women--had been receiving medical care for more than two years. The analyses were based on information they provided about their sexual activity in the previous six months.
Overall, about half of respondents were at least 40 years of age; 51% were white, 32% black, 13% Hispanic and 3% members of other racial or ethnic groups. The majority had at least a high school education. One in five had injected drugs before learning that they had HIV; four in 10 had AIDS. Background characteristics differed significantly among the three groups of respondents. For example, gay or bisexual men were more likely than heterosexual men or women to be white and to have AIDS; heterosexual men were the most likely to be 40 or older and to have injected drugs; and women were, on average, the youngest.
Twenty-eight percent of gay or bisexual men had not had sex within the previous six months; the proportion was about the same among women (34%) but significantly greater among heterosexual men (39%). Reports of sex only after disclosure of one's HIV status were significantly less common among gay or bisexual men (29%) than among heterosexual men (41%) or women (48%). Similarly, 42% of gay or bisexual men reported having had any sex without telling their partner of their HIV infection, compared with 17-19% of other respondents. Eight in 10 gay or bisexual men who reported having had sex without disclosing their HIV status (35% overall) said that they had done so in the context of nonexclusive partnerships; for heterosexual men and women, episodes of intercourse without disclosure were about evenly divided between exclusive and nonexclusive partnerships. The proportion who said they had had unprotected anal or vaginal intercourse without disclosing their infection status was significantly higher among gay or bisexual men (16%) than among other respondents (5-7%).
Roughly half of each group of respondents reported having had any serodiscordant partners (i.e., partners who were HIV-negative or whose HIV status was unknown) during the previous six months; 30% of gay or bisexual men and 9-10% of others had had at least two such partners. Thirteen percent of all partnerships in which the couple were serodiscordant involved unprotected intercourse in the absence of the infected partner's disclosure of his or her HIV status; the proportions in all three groups were similar. In each group of men, most unprotected sex in the absence of disclosure occurred within nonexclusive partnerships and involved a partner whose HIV status was unknown. Among women, such sexual encounters were equally likely to have occurred within nonexclusive and exclusive partnerships.
Using their findings, the analysts estimate that 45,300 gay or bisexual men, 8,000 heterosexual men and 7,500 women with HIV have sex without disclosing their HIV status to their partner; these totals may include as many as 17,400 gay or bisexual men, 2,000 heterosexual men and 2,900 women who have unprotected intercourse without disclosure. While acknowledging that these estimates are subject to a number of limitations, the analysts nevertheless conclude that they "suggest that substantial numbers of new HIV infections could occur among partners of HIV-positive persons who do not disclose their status." They add that the different rates of intercourse without disclosure among the groups studied may indicate that "the norms regarding disclosure [are] quite different" among these groups. Therefore, they recommend that HIV prevention interventions for HIV-positive populations "focus on specific relationships and contexts in which disclosure is most likely to affect behavior."--D. Hollander
1. Ciccarone DH et al., Sex without disclosure of positive HIV serostatus in a US probability sample of persons receiving medical care for HIV infection, American Journal of Public Health, 2003, 93(6):949-954.