Spatio-temporal variations and determinants of antenatal care utilization among adolescents in Bulawayo metropolitan area, Zimbabwe: an analysis of routine data, 2019-2024

Adolescent pregnancy remains a major public health chal- lenge in low- and middle-income countries, contributing to mater- nal and neonatal morbidity and mortality. Antenatal Care (ANC) mitigates pregnancy-related risks through timely screening, edu- cation, and skilled care. However, adolescent ANC utilization remains low - even in urban areas with numerous health service providers. While national demographic and health surveys are used to estimate ANC utilization rates in urban areas, they lack the spatial detail needed to reveal intra-urban disparities for local level health planning. We modelled spatial and temporal varia- tions for at least one visit with a skilled provider (ANC1+) utiliza- tion among pregnant adolescents (10-19 years) within Bulawayo metropolitan province, Zimbabwe, 2019-2024. We extracted ANC utilization records from the District Health Information System and linked the data to a geocoded list of health facilities. Adolescent population denominators (pregnancies) were derived from three independent sources: WorldPop, national statistics agency and the US Census Bureau International Database (IDB). Health Facility Catchment Areas (HFCA) were estimated based on Thiessen polygons and linked with ANC use, pregnancies by population source and geospatial covariates (travel time to facili- ties, urbanization, maternal education, household wealth index, family planning, and vaccine coverage). A Bayesian spatial-tem- poral model was used to estimate ANC1+ coverage per HFCA by year and population. Provincial ANC1+ coverage ranged from 60.4% (WorldPop) to 70.6% (IDB) based on the population source. There was a high spatial heterogeneity in coverage across catchment areas, ranging from below 25% to over 80%. HFCAs located within core urban areas had higher coverage relative to the periphery. No clear temporal trend was observed. Higher wealth index and shorter travel time were significantly associated with ANC1+ utilization. The results are useful for local targeting of resources.