Monitoring the course of the HIV-1 epidemic: The influence of patterns of fertility on HIV-1 prevalence estimates

An age structured model of heterosexual HIV-1 transmission is used to explore the impact of observed patterns of fertility on measures of HIV prevalence derived from child bearing women. Observed reductions in fertility caused by women being outside sexual unions before the age of 20 years, the influence of bacterial sexually transmitted infections and the influence of HIV associated morbidity are all included in the model. We illustrate how the biases in prevalence estimates for a localised epidemic can change with time since the start of the epidemic. As the average age of HIV infected women increases, the over-estimate of prevalence from antenatal clinic samples first increases and then declines. This works in opposition to the influence of HIV-1 on fertility, which causes HIV-1 prevalence to be under-estimated initially. Additionally a reduction in fertility associated with bacterial infection in the highest sexual activity classes causes a substantial under-estimate of HIV prevalence initially, but with the greater HIV associated mortality in this population the bias reduces rapidly.