Urban-rural disparities and change in postnatal care use from 2016 to 2019 in Ethiopia: Multivariate decomposition analysis

Abstract

Background Postnatal care (PNC) is essential for early identification and management of life-threatening obstetric complications. Despite efforts by the Ethiopian government to improve maternal and child health service use, PNC service has remained low, and disparity across geographic locations is a major public health problem. This study aimed to investigate the change and contributing factors in PNC service use across geographical locations (rural-urban) and over time (2016 to 2019) in Ethiopia. Methods We analyzed data on women who gave birth from the 2016 and 2019 Ethiopian Demographic and Health Surveys. A total of 6,413 weighted samples (4,308 in 2016 and 2,105 in 2019) were included in the analysis. A multivariate decomposition analysis technique was used to determine the change and identify factors that contributed to the change across geographical locations and over time. Statistical significance was defined at a 95% confidence interval with a p-value of less than 0.05. Results The prevalence of PNC use was higher among urban residents, and the urban-rural disparity reduced from 32.59% in 2016 to 19.08% in 2019. The difference in the composition of explanatory variables was the only statistically significant for the urban-rural disparity in PNC use in both surveys. Specifically, female household heads (4.51%), delivery at a health facility (83.45%), and birth order of two to three (5.53%) and four or more (-12.24%) in 2016 significantly contributed to the urban-rural gap. However, in 2019, middle wealth index (-14.66%), Muslim religion (3.84%), four or more antennal care contacts (18.29%), and delivery at a health facility (80.66%) significantly contributed to the urban-rural gap. PNC use increased from 16.61% in 2016 to 33.86% in 2019. About 60% of the explained change was due to the difference in the composition of explanatory variables. Particularly, urban residence (-5.79%), a rich wealth index (2.31%), Muslim (3.42%), and other (-2.76%) religions, having radio or television (1.49%), 1–3 (-1.13%), and 4 or more (11.09%) antenatal care contacts, and delivery at a health facility (47.98%) were statistically significant contributors to the observed change. The remaining 40% of the overall change was due to the difference in unknown behaviors (coefficient) of the population towards PNC. Conclusions There was a significant change in PNC service use by residence location and over time in Ethiopia, with urban women in both surveys being more likely to use PNC service. The urban-rural disparity in PNC uptake was due to the difference in the composition of explanatory variables, whereas the change over time was due to the change in both the composition of explanatory variables and population behavior towards PNC. Increased antenatal care contacts and delivery at a health facility played a major role in explaining the gap in PNC services across residences and over time in Ethiopia, highlighting the importance of stepping up efforts to enhance their uptake in rural settings.