Implementation status of maternal death surveillance and response system in Ethiopia: Evidence from a national-level system evaluation

Abstract

Background In Ethiopia, Maternal Death Surveillance and Response (MDSR) was integrated into the existing Integrated Disease Surveillance and Response (IDSR) system in 2014. Despite providing valuable evidence to inform policies and actions, system implementation has not been evaluated. Thus, a national-level evaluation was conducted to assess the level and status of system implementation. Methods A national cross-sectional study was conducted using a multi-stage sampling approach in 2020. A total of 629 health facilities were included in the study. A modified tool, adapted from the World Health Organization (WHO) and the Centers for Disease Control and Prevention (CDC), was employed to assess each functional component of the system, encompassing structure, core, supportive, and system attributes. The score for each component was based on Ethiopian Public Health Institute’s mid-term evaluation metrics. To objectively evaluate the implementation status, a composite score of the Maternal Death Surveillance and Response Performance Index (MDSRPI) was calculated based on five performance indicators. Descriptive statistics, independent t-tests, and one-way analysis of variance (ANOVA) with Bonferroni correction were used to examine the variations in scores among the different characteristics. Results Of the total sample size, 82.5% (519/629) of health facilities were assessed. Among the assessed health facilities, 77.0% (400/519) fulfilled the criteria for final analysis. Accordingly, the overall readiness score was 44.9% (95% CI: 43.9% to 45.9%), which is rated as less functional. The structures of the system were rated at 51.7% (95% CI: 49.9% to 53.4%), and the system attributes were rated at 69.6% (95% CI: 68.0% to 71.2%), which were considered fairly functional. In contrast, the core functions were rated at 20.0% (95% CI: 18.9% to 21.1%), and the supportive functions were rated at 38.4% (95% CI: 36.4% to 40.4%), which were categorized as not functioning and less functional, respectively. Regionally, Tigray’s overall readiness score (54.8%, 95% CI: 50.4–59.1%) was significantly higher than Oromia (41.6%, 95% CI: 40.2–43.0%, P = 0.0001), Amhara (47.7%, 95% CI: 43.9–45.9%, P = 0.05), and SNNPR (42.3%, 95% CI: 39.3–45.3, P = 0.0001). Additionally, Amhara’s score was significantly higher than Oromia and SNNPR. Secondary-level healthcare facilities (49.6%, 95% CI: 45.7–53.7, P = 0.029) had a significantly higher readiness score compared to primary health facilities (44.6%, 95% CI: 43.5–45.6). The overall score for the Maternal Death Surveillance and Response Performance Index (MDSPI) was 33.9%. Conclusion Despite the noticeable regional variation, the overall system readiness and status to implement MDSR were suboptimal, characterized by low representativeness, completeness, and community engagement. Efforts should be directed toward improving community surveillance and enhancing all components of the system to address regional variations and improve overall performance through triangulation and integration with various data sources.