Background Ambient air pollution is worsening in sub-Saharan Africa (SSA) due to increased urbanization and rapid population growth. Vulnerable groups such as pregnant women and children are disproportionately affected, with increased risk of adverse health outcomes. In the background of air quality data scarcity in Africa, the proposed study aims to develop and validate new air pollution proxies and quantify the impact of air pollution on Maternal, Newborn, and Child Health (MNCH) outcomes in SSA. Methods The project enhances local data science capacity and establishes a sustainable research and data resource hub to generate and analyze data on the impact of air pollution on MNCH. Air quality proxy indicators will be validated using personal exposure air pollution and health outcomes data from Kenya, The Gambia and Mozambique. Statistical modelling, machine learning, and geospatial techniques will be employed to estimate air pollution effects on MNCH from cohort and clinical trial data involving 33 countries in SSA. Stakeholder engagement, including policy makers, will be enhanced throughout the study, to inform analysis and priorities for evidence translation. Discussion The study will fill a critical knowledge gap on the impacts of air pollution on MNCH in SSA by developing and validating scalable air quality proxy indicators. The project applies advanced data science and machine learning methods to generate evidence on impact on health to inform policy. Conclusions Integration of data science and machine learning will strengthen analytical capacity in SSA and generate policy-relevant evidence on the impacts of air pollution on MNCH.
Munthali et al. (Thu,) studied this question.