Multilevel analysis of the predictors of completion of the continuum of maternity care in Ethiopia; using the recent 2019 Ethiopia mini demographic and health survey

Abstract

Background Despite the significant benefit of the continuum of care to avert maternal and neonatal mortality and morbidity, still the dropout from the continuum of care remains high and continued to become a challenge in Ethiopia. Therefore, this study aimed to assess the level of completion along the continuum of maternity care and its predictors among reproductive-age women in Ethiopia. Methods A secondary data analysis was done using the 2019 mini Ethiopian demographic health survey. A total weighted sample of 2,905 women aged 15–49 years who gave birth in the last five years preceding the survey and who had antenatal care visits was included. A multilevel mixed-effects logistic regression model was used to examine the predictors that affect the completion of the continuum of maternity care services. Finally, statistical significance was declared at a p-value < 0.05. Results In this study, the overall prevalence of completion along the continuum of maternity care was 12.9% (95%CI: 11.1 – 14.9%). Attending higher education (AOR = 2.03: 95%CI; 1.14 - 3.61), belonged to medium wealth status (AOR = 1.69: 95%CI; 1.07 - 2.66), belonged to rich wealth status (AOR = 2.05: 95%CI; 1.32, 3.17), and informed about danger signs during pregnancy (AOR = 2.23: 95%CI; 1.61, 3.10) were positively associated with the completion of the maternity continuum of care. However, late initiaton of first antenatal care visits (AOR = 0.66: 95%CI; 0.49, 0.89), being rural resident (AOR = 0.67: 95%CI; 0.42 - 0.93), lived in the Afar (AOR = 0.36: 95%CI; 0.12 – 0.83) and Gambella (AOR = 0.52: 95%CI; 0.19 – 0.95) regional states were negatively associated with the completion of the continuum of maternity care. Conclusion Despite most of the women using at least one of the maternity services, the level of completion along the continuum of care after antenatal care booking remains low in Ethiopia. Therefore, enhancing female education and economic transitions with special consideration given to rural, Afar, and Gambella regional state residents. Counseling towards the danger signs of pregnancy and its complications during antenatal care follow-upshould be strengthened. . Furthermore, the identified predictors should be considered when designing new policies or updating policies and strategies on maternity services uptake to step-up its full utilization, which in turn helps in the achievement of the sustainable development goals of ending preventable causes of maternal, neonatal, and child death by 2030.

Multilevel analysis of dropout from maternal continuum of care and its associated factors: Evidence from 2022 Tanzania Demographic and Health Survey

“…Conversely, women in the 40–44 age group had lower odds of dropout from institutional delivery compared to their younger counterparts (15–19 years). This pattern is corroborated by studies in Ethiopia [ 29 ] and Mozambique [ 21 ]. This report could be associated with an increase in knowledge and decision-making abilities as women age [ 30 ].…”

“…A study in Colombia showed 95% of women were dropouts. Similarly, in Ethiopia, the proportion of women experiencing CoC dropout was reported at 87.1% [ 21 ]. Only about 10.7% Ugandan women received complete CoC [ 16 ].…”

Multilevel and subnational analysis of the predictors of maternity continuum of care completion in Nigeria: a cross-sectional survey

“…Multilevel model was fitted at each model stage (model-I, model-II model-III and model-IV) to evaluate enumeration area cluster dependencies in prediction of maternity continuum of care. Model-I is the empty model i.e., intercept only terms, model-II is the individual only term, model-III comprises of the community term while model-IV combined all the terms as a full model 48 , 49 . The individuals (postpartum women) within households are nested in communities’ enumeration area cluster.…”

“…Thus, the ICC measured the degree of clustering based on heterogeneity between clusters i.e., the proportion of total variation in maternity CoC attributable to the community level random-effect component which can be expressed as: where is the between cluster variance (community random-component) and is the within community cluster variance from assumed normally distributed population 49 , 50 .…”

“…The explained variance is computed from the proportional change in variance (PCV) which estimates the difference in variance between the null and the model with at least a factor term (i.e., including covariates) as a ratio of the null model variance. where is the variance of the null model, is the variance of observed factor for i is the number of factor or covariates terms in the models i.e., model II, III and IV 21 , 49 , 50 .…”