This study employs structural equation modelling to explore the inter-relationships among optimal antenatal care (ANC), health facility delivery and early postnatal care (EPNC) in Ethiopia. By identifying both direct and indirect influencing factors, the study offers valuable insights to support integrated maternal health strategies and guide informed decision-making by policymakers and women alike.

The secondary analysis of the Ethiopian Demographic and Health Survey 2016 was performed to investigate inter-relationships between optimal ANC, health facility delivery and postnatal care (PNC) among women in Ethiopia. Data were analysed with R software V.4.3.2. The study used binary logistic regression to examine differences in optimal ANC, health facility delivery and EPNC, focusing on variables with a p value of 0.1 or less. Selected variables were incorporated into a generalised structural equation model (GSEM) using the LAVAAN package to explore both direct and indirect effects. The GSEM method assessed the impact of exogenous variables on endogenous variables, all binary, using a logistic link and binomial family. Missing data were handled with the multiple imputation by chained equations package, and sampling weights were applied to ensure national and regional representativeness.

The source population comprised all women of reproductive age (15–49 years) who gave birth in the 5 years preceding the survey. From 16 650 interviewed households (98% response rate), we identified 7590 eligible women with recent births. Finally, we included 2415 women who had attended four or more ANC visits.

Media exposure significantly boosts the likelihood of using ANC (OR=1.8, 95% CI (1.04 to 3.23), p=0.04), health facility delivery (OR=1.7, 95% CI (1.23 to 2.45), p=0.05) and PNC (OR=2.0, 95% CI (1.6 to 4.01), p=0.01). Urban residence and secondary education also enhance ANC (OR=1.2, 95% CI (1.01 to 2.88), p=0.022; OR=1.3, 95% CI (1.20 to 3.01), p=0.018), health facility delivery (OR=1.1, 95% CI (1.01 to 3.24), p=0.035; OR=1.5, 95% CI (1.22 to 3.45), p=0.03) and PNC (OR=1.6, 95% CI (1.01 to 4.32), p=0.03). ANC directly affects health facility delivery (OR=1.4, 95% CI (1.28 to 3.09), p=0.01) and PNC (OR=1.6, 95% CI (1.01 to 3.80), p=0.03). Additionally, women aged 20–34 years and those from male-headed households positively impact health facility delivery (OR=1.5, 95% CI (1.20 to 4.80), p=0.01; OR=1.3, 95% CI (1.07 to 3.45), p=0.014) and PNC (OR=1.4, 95% CI (1.10 to 2.90), p=0.01; OR=1.2, 95% CI (1.07 to 3.08), p=0.025).

Optimal ANC is vital for encouraging health facility delivery and EPNC. To enhance maternal and neonatal health, policies should integrate these services. Key predictors include being aged 20–34, having secondary and higher education, media exposure, male-headed households and living in urban areas. Improving education and media exposure can boost maternal healthcare service use.

Non-adherence to antidiabetic medication remains a major barrier to achieve optimal health outcomes among individuals with diabetes, particularly in developing countries. This issue exacerbates poor health outcomes and leads to the wastage of limited healthcare resources.

This study aimed to assess the prevalence of non-adherence to antidiabetic medications and identify associated factors among adult type 2 diabetes mellitus (DM) patients in the North Wollo zone.

An institutional-based cross-sectional study.

The study was conducted in three randomly selected public hospitals in the North Wollo zone: Woldia Comprehensive Specialized Hospital, Lalibela General Hospital and Mersa Primary Hospital.

A total of 327 adult type 2 DM patients receiving follow-up care were included. Participants were selected proportionally from each hospital using consecutive sampling. Inclusion criteria included individuals aged ≥18 years, on antidiabetic treatment for at least 6 months and actively on follow-up care during the study period. Patients with hearing impairment, severe illness or incomplete medical records were excluded.

Adherence was assessed using the Morisky Medication Adherence Scale-8, a validated eight-item, self-reported questionnaire. Scores ranged from 0 to 8, with adherence levels classified as high (≥8), medium (6–7.75) and low (

Data were analysed using SPSS V.27. Descriptive statistics were used to summarise the data, and multivariable logistic regression analysis was performed to identify factors associated with non-adherence. A p value ≤0.05 was considered statistically significant.

The overall prevalence of medication non-adherence was 24.5%. Factors significantly associated with non-adherence included living with diabetes for less than 3 years (adjusted OR (AOR) 3.37, 95% CI 1.91 to 5.95), residing in rural areas (AOR 2.67, 95% CI 1.49 to 4.79), having comorbidities (AOR 2.99, 95% CI 1.67 to 5.34) and having no formal education (AOR 3.26, 95% CI 1.49 to 7.00).

The prevalence of non-adherence to antidiabetic medications (24.5%) exceeded the widely accepted benchmark of ≤20%. Key factors such as rural residence, comorbidities, lower education levels and shorter duration since diagnosis were significantly associated with non-adherence. These findings underscore the need for targeted interventions, including patient education, improved rural healthcare access and integrated care models, to enhance adherence and diabetes management outcomes.