Prevalence and incidence of pregnancy, HIV and STIs among eligible women attending at baseline (n = 1020).

1. Participants were tested for each incident endpoint at baseline, and at three-monthly intervals during follow-up. Incidence was calculated as the number of new cases divided by the total time in years that participants remained in the study without the incident endpoint of interest. Following standard practice, participants were considered censored at their last recorded study visit.2. Pregnancy and HIV seropositive status were initial study exclusion criteria. By definition, HIV prevalence was therefore zero among 1020 eligible women at baseline.3. Incident cases were defined as those who tested positive for both TPPA and RPR for the first time during follow-up and exclude women with active syphilis (TPPA+ and RPR+) at baseline.