Decision analysis to guide choice of interventions to reduce mother-to-child transmission of HIV.

Introduction: Antiretroviral prophylaxis, avoidance of breastfeeding, and early weaning are candidates to prevent mother-to-child transmission (MTCT) of HIV worldwide. Methods: We developed a model to help guide population-level decisions about MTCT intervention strategies. We estimated the numbers of early childhood deaths prevented by (1) prenatal short-course zidovudine, (2) intrapartum and neonatal short-course nevirapine, (3) avoidance of breastfeeding, and (4) early weaning (age 6 months); four combinations of these; and one possible future strategy (postnatal antiretroviral prophylaxis) in a scenario typical of a developing country. We evaluated the effectiveness of the interventions for a range of R, the relative risk of mortality for children exposed to breastfeeding interventions compared with breastfed children (independent of HIV infection). We also estimated the reduction in breastfeeding transmission needed for a postnatal antiretroviral intervention to prevent more early childhood deaths than do currently available interventions. Results: Where R ≤ 1.5, strategies combining antiretroviral prophylaxis with breastfeeding interventions prevent the most early childhood deaths. However, strategies that include early weaning and avoidance of breastfeeding, respectively, can result in more deaths than with no intervention when R > 1.5 and R > 1.9, respectively. The relative effectiveness of a postnatal antiretroviral intervention compared with avoidance of breastfeeding varies with R; such that an intervention would be more effective than early weaning as a single intervention, at any R, if it reduced HIV transmission through breastfeeding by 25%. Conclusion: This spreadsheet model is a simple, locally adaptable tool to allow decision-makers to explore key questions about intervention strategies to prevent MTCT of HIV.