|| Checking for direct PDF access through Ovid
Previous studies on the operational effectiveness of programmes to reduce transmission of HIV from mother-to-child (PMTCT) in Africa have generally been hospital-based pilot studies with short follow-up periods.Prospective cohort study to evaluate the routine operational effectiveness of the South African National PMTCT Programme, primarily measured by HIV-free survival at 36 weeks post-delivery. Three of eighteen pilot sites participating in the programme were selected as they reflected differences in circumstances, such as HIV prevalence, socioeconomic status and rural–urban location. A total of 665 HIV-positive mothers and their infants were followed.HIV-free survival at 36 weeks varied significantly across sites with 84% in Paarl, 74% in Umlazi and 65% in Rietvlei (P = 0.0003). Maternal viral load was the single most important factor associated with HIV transmission or death [hazard ratio (HR), 1.54; 95% confidence interval (CI), 1.21–1.95]. Adjusting for health system variables (fewer than four antenatal visits and no antenatal syphilis test) explained the difference between Rietvlei and Paarl (crude HR, 2.27; 95% CI, 1.36–3.77; adjusted HR, 1.81; 95% CI, 0.93–3.50). Exposure to breastmilk feeding explained the difference between Umlazi and Paarl (crude HR, 1.74; 95% CI, 1.06–2.84; adjusted HR, 1.41; 95% CI, 0.81–2.48).Ever breastfeeding and underlying inequities in healthcare quality within South Africa are predictors of PMTCT programme performance and will need to be addressed to optimize PMTCT effectiveness.