A key gap in prevention of mother-to-child HIV transmission (PMTCT) implementation science is the lack of validated approaches to measure program effectiveness. Although this need has been recognized for years, it has come to the forefront with the 2011 launch of the "eliminating mother-to-child HIV transmission" (EMTCT) campaign by UNAIDS, PEPFAR, and other global partners. Alongside ambitious goals to dramatically reverse the pediatric AIDS epidemic, EMTCT calls for new measures to define program success. Traditional metrics such as program coverage have been replaced with direct health outcomes such as HIV-free survival, an important paradigm shift that will require new and innovative evaluation approaches. The World Health Organization is currently developing guidelines for PMTCT evaluation and among their recommended methodologies is a community survey. Strengths of the survey approach are its population-level sampling, an ability to measure HIV-free survival, a cross-sectional design with retrospective data collection, and the ease with which it could be incorporated into ongoing nationwide initiatives (e.g., Demographic and Health Surveys). However, potential ascertainment and reporting biases could lead to inaccurate results. If this methodology is to be recommended for broader implementation, a formal validation is urgently needed. To address this important knowledge gap, a two-part study is proposed. First, a community survey methodology will be validated against a "gold standard" community cohort design. In support of this aim, retrospective community survey data from 10,500 households across 42 communities (250 per community) involved in the ongoing Better Health Outcomes through Mentorship and Assessment (BHOMA) study will be used to calculate HIV- free survival in HIV-exposed children at 18 months of age. In parallel, 1,680 HIV-exposed newborns will be enrolled in cohorts across the same communities (40 per community) and followed prospectively for 18 months. Measurements of HIV-free survival from the two methodologies will be compared, and the validity of the survey approach determined. Second, program characteristics and practices associated with improved PMTCT outcomes will be identified by collecting facility-level data-including direct observations of patient care, assessment of patient encounter times, and supply inventory levels-and correlating them with community- level HIV-free survival from both the survey and cohort methodologies. In this manner, the study will identify "best practices" for PMTCT and potentially generate new scientific questions for interventional research, a key function of implementation science within PEPFAR's framework. The results from the study are expected to have a direct impact on PMTCT program evaluation in Zambia and immediately inform monitoring and evaluation policy recommendations globally.
Programs to prevent mother-to-child HIV transmission (PMTCT) have successfully and rapidly expanded across sub-Saharan Africa;however, approaches for measuring their impact lack consensus and standardization. We seek to validate a community survey methodology for estimating HIV-free survival among HIV-exposed infants at a population level, comparing our measurement against that of a "gold standard" cohort design. Our results will have an immediate impact on regional and international policy recommendations by helping to establish a rigorous evaluation framework for pediatric HIV elimination.