Changes in seroadaptive practices from before to after diagnosis of recent HIV infection among men who have sex with men

Abstract

Objective: We assessed changes in sexual behavior among men who have sex with men (MSM), before and for several years after HIV diagnosis, accounting for adoption of a variety of seroadaptive practices.

Methods: We collected self-reported sexual behavior data every 3 months from HIV-positive MSM at various stages of HIV infection. To establish population level trends in sexual behavior, we used negative binomial regression to model the relationship between time since diagnosis and several sexual behavior variables: numbers of (a) total partners, (b) potentially discordant partners (PDP; i.e., HIV-negative or unknown-status partners), (c) PDPs with whom unprotected anal intercourse (UAI) occurred, and (d) PDPs with whom unprotected insertive anal intercourse (uIAI) occurred.

Results: A total of 237 HIV-positive MSM contributed 502 interviews. UAI with PDPs occurred with a mean of 4.2 partners in the 3 months before diagnosis. This declined to 0.9 partners/3 months at 12 months after diagnosis, and subsequently rose to 1.7 partners/3 months at 48 months, before falling again to 1.0 partners/3 months at 60 months. The number of PDPs with whom uIAI occurred dropped from 2.4 in the pre-diagnosis period to 0.3 partners/3 months (an 87.5% reduction) by 12 months after enrollment, and continued to decline over time.

Conclusion: Within months after being diagnosed with HIV, MSM adopted seroadaptive practices, especially seropositioning, where the HIV-positive partner was not in the insertive position during UAI, resulting in a sustained decline in the sexual activity associated with the highest risk of HIV transmission.

Figure 1. Mean number of partners of various types per 3 months since HIV diagnosis among HIV-positive MSM in San Francisco, 2009–2010.

An immediate drop in the total number of male partners in the first year of infection was followed by increases in number of partners over the following 3–4 years. The trend was similar for potentially serodiscordand partners (PDPs) although they comprised only 1/3 to 1/2 of total partnerships. However, unprotected anal intercourse (UAI) with PDPs occurred in far fewer partnerships throughout the follow-up period. Partnerships in which the HIV-positive participant was the insertive partner during unprotected anal intercourse (uIAI) accounted for fewer than 10% of all partnerships and in very few of those partnerships did the participant have sufficient plasma viral load (VL >500 copies/ml) to present a significant transmission risk.