Sexual Contact Patterns between Men and Women and the Spread of HIV-1 in Urban Centres in Africa

Abstract

A simple model of the transmission of HIV-1 by heterosexual contact and from mother to unborn infant is developed to assess the influence of patterns of mixing between low and high sexual activity classes of the two sexes on the pattern of spread of the virus and the demographic impact of AIDS. Numerical studies of model behaviour are based, where possible, on parameter estimates derived from epidemiological studies of HIV-1 spread in Africa. Analyses reveal that the assumed pattern of mixing, ranging from assortative (like with like) through random (proportional) to disassortative (like with unlike), has a very major impact on the predicted spread of the virus and the concomitant demographic impact of AIDS. Patterns of strong assortative mixing are predicted to generate the least spread and demographic impact, by comparison with proportional or disassortative mixing. Analyses also reveal that the rules governing behaviour changes, once AIDS-induced mortality changes the structure of the population (i.e. the numbers in the low and high sexual activity classes of the two sexes), have a very significant influence on the course of the epidemic. Where possible, predicted patterns are compared with observed trends in Africa.

AIDS, mixing patterns, Africa, mathamatical models, sexual behaviour, transmission