Secondary HIV transmission rates in a mixed-gender sample

Information about the sexual behaviour of HIV-infected individuals is needed to predict the course of the sexually transmitted HIV epidemic in the US. The present study provides model-based estimates of the secondary transmission rate (i.e. the number of infections expected among the sex partners of already infected individuals) for a sample of HIV-positive persons in Atlanta. A mathematical model was used to estimate the secondary transmission rate of HIV infection for a sample of HIV-positive men and women in Atlanta, based on their self-reported sexual behaviour, extrapolated over a 15-year horizon. Separate rates were calculated for different transmission routes, including: from women to men-who-have-sex-with women (MSW) and from men-who-have-sex-with-men (MSM) to other MSM. Sensitivity analyses were conducted to assess the impact of different parametric and modelling assumptions. Restricted to the sub-sample that reported transmission risk behaviours, the mean number of secondary infections was 0.14 for transmission from women to MSW; 0.31 for transmission from MSW to women; and 0.84 for MSM to MSM transmission. Bisexual men were at especially high transmission risk, with 1.59 and 0.54 secondary infections expected among their male and female partners, respectively. The main analysis indicates that, in this sample, each current infection will lead to fewer than one future infection for all groups other than bisexual men, which suggests that the epidemic is contracting in this community, although this analysis cannot rule out the possibility of a growing epidemic among MSM. This method can be used to identify groups at high risk for HIV transmission and thereby to better target HIV prevention resources.