Geospatial Variation and Determinants of Time to Pregnancy Loss Among Reproductive‐Aged Women in East Africa: Bayesian Spatial Frailty Model

Background Pregnancy loss is the terminated pregnancy before the completed pregnancy time. Spatial location can significantly influence pregnancy outcomes, yet geographic disparities in the risk of pregnancy loss remain poorly studied in East Africa. This study is aimed at examining the spatial variation and determinants of time to pregnancy loss among women of reproductive age in the region. Methods A Bayesian spatial survival model with an intrinsic conditional autoregressive approach was utilized to identify factors related to time to pregnancy loss across 169 regions in 9 East African countries using secondary data from recent demographic and health surveys (2015–2023). A spike‐and‐slab prior was used for variable selection. Model comparison utilized deviance information criteria, Watanabe–Akaike information criterion, and log pseudo marginal likelihood, with diagnostics from Cox and Snell residuals. Results Pregnancy loss showed significant spatial clustering. Earlier occurrence was observed among women aged 25–34 years ( ϕ = 0.712; 95% CrI: 0.680–0.744) and ≥ 35 years ( ϕ = 0.326; 95% CrI: 0.307–0.344), those living with their husbands ( ϕ = 0.893; 95% CrI: 0.856–0.932), with primary education ( ϕ = 0.939; 95% CrI: 0.892–0.987), regular media exposure ( ϕ = 0.854; 95% CrI: 0.819–0.889), employed ( ϕ = 0.818; 95% CrI: 0.788–0.850), and attending antenatal care ( ϕ = 0.964; 95% CrI: 0.933–0.996). In contrast, college‐educated women ( ϕ = 1.109; 95% CrI: 1.024–1.203), rural residents ( ϕ = 1.144; 95% CrI: 1.102–1.190), low parity ( ϕ = 14.93; 95% CrI: 13.853–16.132), and grand multiparity ( ϕ = 22.088; 95% CrI: 20.131–24.247) were associated with longer time to pregnancy loss. After adjusting for individual‐level factors, residual variation in the hazard of pregnancy loss remained. Conclusions Pregnancy loss remains a significant public health challenge in East Africa, with significant geographic variation. Spatial analysis can guide region‐specific healthcare strategies and targeted interventions.