Spatial and hierarchical Bayesian analysis to identify factors associated with caesarean delivery use in Ethiopia: Evidence from national population and health facility data.

Abstract

Caesarean section has a significant role in reducing maternal and neonatal mortality. A linked analysis of population and health facility data is valuable to map and identify caesarean section use and associated factors. This study aimed to identify geographic variation and associated factors of caesarean delivery in Ethiopia. Linked data analysis of the 2016 Ethiopia Demographic and Health Survey (EDHS) and the 2014 Ethiopian Service Provision Assessment Plus (ESPA+) survey was performed. Spatial analysis was conducted to identify geographic variations and factors associated with caesarean delivery. Hierarchical Bayesian analysis was also performed to identify factors associated with caesarean delivery using the SAS MCMC procedure. Women's age and education, household wealth, parity, antenatal care (ANC) visits, and distance to caesarean section facility were associated with caesarean delivery use. Women who had ≥4 ANC visits were 4.67 (95% Credible Interval (CrI): 2.17, 9.43) times more likely to have caesarean delivery compared to those who had no ANC visits. Women who had education and were from rich households were also 2.80 (95% CrI: 1.83, 4.19) and 1.80 (95% CrI: 1.08, 2.84) times more likely to have caesarean deliveries relative to women who had no education and were from poor households, respectively. A one-kilometer increase in distance to a caesarean section facility was associated with an 88% reduction in the odds of caesarean delivery (Adjusted Odds Ratio (AOR) = 0.12, 95% CrI: 0.01, 0.78). Hotspots of high caesarean section rates were observed in Addis Ababa, Dire Dawa, and the Harari region. In addition, women's age at first childbirth and ≥4 ANC visits showed significant spatially varying relations between caesarean delivery use across Ethiopia. Caesarean section is a lifesaving procedure, and it is essential to narrow disparities to reduce maternal and neonatal mortality and avoid unnecessary procedures.