OBJECTIVES: To examine new evidence from studies on the estimates of the fertility rate ratio comparing HIV-infected and uninfected women, of the population change in total fertility attributable to HIV, and to review the evidence of changes in fertility in HIV-uninfected women. DESIGN: A review and analysis of data from the many individual studies that have examined the associations between HIV/AIDS and fertility. METHODS: Data from sub-Saharan Africa were collected from published studies, personal communications and the Demographic and Health Surveys. A mathematical model was used to demonstrate the impact of the HIV/AIDS epidemic on the number of births in Uganda. RESULTS: Fertility was lower among HIV-infected women than HIV-uninfected women, with the exception of those aged 15-19 years, in whom the selective pressure of sexual debut on pregnancy and HIV infection led to higher fertility rates among the HIV infected. This fertility differential resulted in a population-attributable decline in total fertility of 0.37% (95% confidence interval 0.30%, 0.44%) for each percentage point of HIV prevalence. The evidence for fertility changes in HIV-uninfected women was ambiguous. An estimated reduction of 700 000 births occurred in Uganda, as a result of the reduced fertility in HIV-infected women and premature mortality among reproductive age women. CONCLUSION: Large fertility differentials existed between HIV-infected and uninfected women, with substantial variation by age. The extent to which these could be attributed to the direct impact of the epidemic on both infected and uninfected women, as opposed to pre-existing differences in their fertility, merits further study.