Individual and community-level factors influencing optimal breastfeeding: A multilevel analysis from a national survey study of Ethiopia

Background Optimal breastfeeding is critical for healthy growth of the child. Globally, 820,000 children and 20,000 women lost due to in appropriate breastfeeding each year. In Ethiopia, 50,000 children lost related to malnutrition with 18% were due to poor breastfeeding habit. Little is known on the determinants of breastfeeding with hierarchical level. Therefore, this study aimed to identify factors influencing optimal breastfeeding among children under six month in Ethiopia using a multilevel analysis. Methods The data of this study were obtained from Ethiopian Demographic and health survey conducted from January to June 2016. A total 1,087 children aged 0-5 months were selected using two stage stratified sampling technique. Multilevel logistic regression analysis was done to identify significant explanatory variables. Akaike information criteria were used to select the best model fit. Fixed effect was done to estimate the association between the outcome and explanatory variable and also random effect to measure the variation explained by the higher level. Result Among the total of 1,087 children, 45.4% were optimally breastfeed. Children from the richest wealth index (AOR = 2.87; 95% CI: 1.53-5.43) was positively associated with optimal breastfeeding but, children aged 4-5 months (AOR = 0.19; 95%CI: 0.12-0.27), children born through cesarean section (AOR = 0.18; 955 CI: 0.07-0.51) and residing in Afar region (AOR = 0.13; 95%CI: 0.02-0.92) were found inversely associated with optimal breastfeeding. The random-effects showed that the variation between communities was statistically significant. Conclusion Individual and community level factors play a significant role in shaping optimal breastfeeding. Future strategies and health interventions should be strengthen to target individual and community level factors that enhance optimal breastfeeding.