Spatial variation and determinant of home delivery in Ethiopia: Spatial and mixed effect multilevel analysis based on the Ethiopian mini demographic and health survey 2019

Background Health facility delivery is vital in reducing maternal mortality however, the percentage of deliveries taking place in health facilities were remained below 50%. Therefore, this study was aimed to assess spatial variation and determinant factors of home delivery in Ethiopia. Methods We used cross-sectional survey data from Ethiopian Mini Demographics and health 2019. A total of 5,527 reproductive-age women were included in this study. ArcGIS 10.7 was applied to explore the spatial distribution of home delivery and Sat scan 9.6.1 version software was used for spatial cluster analysis. A mixed effect multilevel binary logistic regression model was fitted for determinant factors due to the hierarchical nature of the data. Finally, an adjusted odds ratio (AOR) with 95% confidence level was used to declare significant determinants of home delivery. Result According to EMDHS 2019, the spatial variation of home delivery was non-random across the country. Somali, Afar, SNNPR, and part of Amhara were hot spot areas, where some parts of Benishangul, central Oromia, Addis Ababa, Dire Dawa, and Harari were identified as cold spot areas. The odds of women who had primary, secondary and higher education was decreased by 50% (AOR = 0.50; 95% CI: 0.42-0.61), 72% (AOR = 0.28; 95% CI: 0.19-0.40) and 90% (AOR = 0.10; 95% CI: 0.05-0.19) as compared to women no-education respectively. Mothers who had ANC visits were 70% (AOR = 0.30; 95% CI: 0.26-0.36) less likely to have a home birth as compared to women who had no ANC visit. The odds of having home birth among rural residents were 5.2 times (AOR = 5.2; 95% CI: 3.11-8.55) more likely higher compared to the counterpart. Conclusion The prevalence of home delivery in Ethiopia was still more than half percent. The spatial distribution was varied across the region. Maternal age, religion, wealth status, had ANC visit, birth order, region, and residence were significant factors with home delivery. Therefore, improving maternal educational status, interventional design in hotspot region, and inspire the mother to take antenatal care is essential to reduce the prevalence of home delivery.