Perceived HIV status is a key determinant of unprotected anal intercourse within partnerships of men who have sex with men in Amsterdam.

The practice of unprotected anal intercourse (UAI) involves at least two partners. We examined the associations between insertive or receptive UAI and perceived HIV seroconcordance and partnership type in self-perceived HIV-negative and self-perceived HIV-positive men who have sex with men (MSM). MSM (age ≥ 18 years) were recruited for a cross-sectional survey at the sexually transmitted infections clinic in Amsterdam, the Netherlands, in 2008-2009. Participants completed a questionnaire concerning partnerships in the preceding 6 months. Associations were quantified via multinomial logistic regression models using generalized estimating equations. The outcomes were 'no, or safe anal intercourse', 'insertive UAI', and 'receptive UAI'. We included 5,456 partnerships from 1,890 self-perceived HIV-negative men and 1,861 partnerships from 558 self-perceived HIV-positive men. Within the partnerships, perceived HIV status of the partner was an important determinant of UAI (p < 0.001). Among HIV-negative men, perceived HIV discordance was negatively associated with receptive UAI compared with no or safe UAI (OR 0.57; 95 % CI 0.36-0.92); when the partners were more familiar with each other, the risk of receptive UAI was increased relative to no or safe anal intercourse. Among HIV-positive men, perceived HIV discordance was negatively associated with insertive UAI (OR 0.05; 95 % CI 0.03-0.08). Within partnerships, perceived HIV status of the partner was one of the strongest determinants of UAI among self-perceived HIV-negative and HIV-positive MSM, and discordant serostatus was negatively associated with UAI. The findings suggest that serosorting is one of the main strategies when engaging in UAI.