Unprotected Anal Sex is Not Uncommon Among Men with HIV Infection

D. Hollander

First published online:

Three in 10 HIV-positive men interviewed in 1995-2000 reported that in the previous year, they had had unprotected anal intercourse with a steady male partner who was HIV-negative or whose infection status was unknown; the proportion was almost as high—two in 10—among men who had already been aware that they were infected.1 In the latter group, those who had no more than a high school education, who identified themselves as heterosexual, who did not know their partner's HIV status or who had used crack were at increased risk of engaging in anal intercourse without using a condom.

The interviews were conducted as part of an ongoing surveillance project involving individuals in 12 states with newly reported HIV infection or AIDS. The project, designed by the Centers for Disease Control and Prevention, collects information that is not routinely gathered through HIV and AIDS case reporting. Participants must be at least 18 years old, speak English or Spanish, and be medically able to complete a 45-minute interview. During the study period, 970 men reported having had a steady male sex partner who was not infected or whose serostatus they did not know in the year before the interview. Analysts examined data from these men to assess the prevalence and predictors of unprotected anal intercourse.

Roughly half of the men in the sample were members of racial or ethnic minority groups, half were in their 30s and half had more than a high school education. Eighty-four percent identified themselves as gay, and 12% as bisexual; 1% said that they were heterosexual. The vast majority (83%) had AIDS. Only 1% of men reported using injection drugs during the past year, and 6% reported using crack. At the time of interview, 20% had known of their HIV infection for six months or less, 10% for 7-12 months and 69% for more than a year.

Overall, 79% of men had had anal intercourse in the year before the interview, and 29% had done so at least once without using a condom. The proportion who had had unprotected anal intercourse declined significantly as the length of time since HIV diagnosis increased: from 52% among men who had known of their infection for no more than six months to 30% among those who had been aware of it for 7-12 months and 21% among those who had known for more than a year.

HIV diagnoses made within the previous year may or may not have preceded reported occurrences of unprotected anal intercourse. Therefore, to assess behavior of men who knowingly put their partners during the previous year at risk of infection by not using condoms, the analysts focused on those whose infection had been detected more than a year earlier. In this subgroup, 36% of men who had had unprotected intercourse reported never using condoms, and 64% reported using them some of the time. Thirteen percent said that they had engaged only in insertive unprotected anal intercourse, 44% in receptive unprotected anal intercourse and 44% in both.

In analyses adjusting for men's background characteristics and risk-related behaviors, men who had been aware of their infection status had elevated odds of having engaged in unprotected anal intercourse if they had a high school education or less (odds ratio, 1.8), their partner's HIV status was unknown (1.8) or they had used crack in the previous year (3.1). The odds also were elevated among men who identified themselves as heterosexual (8.3), but very few men were in this group and the confidence interval around the odds ratio was very wide. Compared with men whose infection had been diagnosed 1-4 years before interview, those who had known for five or more years that they were HIV-positive were significantly less likely to have engaged in anal intercourse without a condom (0.6). Race and ethnicity were not associated with having had unprotected anal intercourse, but the proportions of black and Hispanic men who had one or more of the identified risk factors (70% and 74%, respectively) were significantly higher than the proportion among white men (47%).

While the analysts acknowledge that their sample may not be representative of all HIV-infected U.S. men who have sex with men, they suggest that their findings have several implications for prevention programs. For example, they comment that programs should take a "holistic approach to the health care and prevention needs of HIV-positive men," help infected men develop the communication skills needed to discuss their infection status and safer-sex behaviors with partners, and emphasize the importance of HIV counseling and voluntary testing for the partners of infected men.

—D. Hollander


1. Denning PH and Campsmith ML, Unprotected anal intercourse among HIV-positive men who have a steady male sex partner with negative or unknown HIV serostatus, American Journal of Public Health, 2005, 95(1):152-158.