Abstract

Most mathematical models used to examine the role of different stages of human immunodeficiency virus infection unrealistically assume that HIV is transmitted through one-off contacts or that transmission rates are the same between males and females. We sought to examine whether inferences from previous models are robust to the relaxation of those unrealistic assumptions. We developed a model of HIV transmissions through sexual partnerships assuming that sexual partnerships have variable duration, sexual partnerships are concurrent, and the male-to-female transmission rate is higher than the female-to-male transmission rate, with a focus on the third assumption. Assuming a higher rate for male-to-female than female-to-male transmissions decreases the overall transmission of HIV but increases the equilibrium fraction of transmissions during primary HIV infection in long-term partnerships, compared to the case where transmission rates are assumed to be symmetric between males an females. Previous modeling studies that assume symmetric transmission rates between males and females may have overestimated the overall spread of HIV, but underestimated the relative contribution of PHI. To make robust inferences on the role of different stages of HIV infection in the sexual spread of HIV, models should take into account that transmission rates may be asymmetric by sex