Giving birth in Ethiopia: a spatial and multilevel analysis to determine availability and factors associated with healthcare facility births

Objective: To assess spatial variations in the use of healthcare facilities for birth and to identify associated factors. Design: Cross-sectional analysis of population- and healthcare facility-based data. Setting: Ethiopia Demographic and Health Survey (EDHS 2016) linked to Service Provision Assessment data (SPA 2014). Population: A sample of 6954 women who gave birth in the 5 years preceding EDHS 2016 and 717 healthcare facilities providing delivery care. Methods: Secondary data analysis of linked population and health facility data was conducted. Multilevel and spatial analyses were conducted to identify key determinants of women’s use of health facilities for birth and to assess spatial clustering of facility births. Main outcome measure: Health facility birth. Results: A one-unit increase in the mean score of the readiness of health facilities to provide basic emergency obstetric care (EmOC) was associated with a two-fold increase in the odds of facility birthing (adjusted odds ratio, aOR, 2.094, 95% CI 1.187–3.694). A woman’s attendance for at least four antenatal care visits was significantly associated with facility birth (aOR 8.863, 95% CI 6.748–11.640). Distance to a healthcare facility was inversely related to a woman’s use of facility birthing (aOR 0.967, 95% CI 0.944–0.991). Women in the richest wealth quintile were also more likely to have facility births (aOR 2.892, 95% CI 2.199–3.803). Conclusions: There were geographic variations in facility births in Ethiopia, revealing critical gaps in service availability and readiness. It is important to ensure that health facilities are in a state of readiness to provide EmOC. Tweetable abstract: Failure to ensure health facility readiness is associated with failure to give birth at a healthcare facility.