Skip to content

Abstract Search

Global Health

Reduction of Child Mortality Rates following 10 years of a health system strengthening program in rural Madagascar: results from the Ifanadiana Health Outcomes and Prosperity longitudinal Evaluation (IHOPE) Ann C. Miller* Ann Miller Hobisoa Razanandranaivo Benedicte Razafinjato Sarah-Anne Barriault Marius Randriamanambintsoa Samuel Andrianambinina Andres Garchitorena Rado Rakotonahary Luc Rakotonirina Laura Cordier Karen Finnegan Matthew Bonds Andres Garchitorena

Background:Since 2014, the NGO Pivot and the Madagascar Ministry of Public Health collaborated on a health system strengthening(HSS) intervention in a low-resource District in eastern Madagascar (pop. 200,000). This study assessed HSS impact on child mortality at the population level over 10 years.

Methods: This population-representative longitudinal cohort study used a 2-stage cluster sample. 1600 households were randomly sampled from 80 randomly-selected clusters, half per each of 2 strata:Pivot’s initial catchment(IC) and the rest of district(ROD). Professional data collectors conducted face-to-face interviews with women ages 15-49 repeatedly between 2014 and 2023. We provide 5-year weighted estimates of population-level mortality using synthetic life-table methods. We estimated absolute change from 2014 baseline for neonatal, infant and under-5 mortality(probability of death before 1, 12, and 60 months respectively per 1000 live births) comparing IC to ROD for each of the 5 waves of the cohort (2014, 2016, 2018, 2021, 2023), using 95% confidence intervals.

Results: 4380 children were included. Neonatal mortality dropped from 47.3/1000 [95%CI 26.9-67.7]live births to 30.7/1000 [4.5-56.8] in IC vs rising from 34.5/1000 [95%CI 18.4-50.6] to 64.4/1000 in ROD [37.5-91.3]. Infant mortality dropped from 66.1/1000 [40.4-91.9] in 2014 to 45.3/1000 [16.2-74.4] in 2023 (IC), while rising in ROD (64.3/1000 [45.6-82.9]to 88.2/1000 [60.4-116.4]). Under-5 mortality dropped for both strata from baseline; IC experienced greater reductions 104.1 [72.6-135.5] to 71.3/1000[36.1-106.5] vs ROD 168.6 [128.2-208.9]to153.3/1000[121.29-185.3]. Under-5 mortality was significantly different between strata in 2023.

Conclusions: In an area where health and economic indicators are not otherwise improving, these results provide evidence that strengthening health systems for universal care and access in low-resource settings can save lives.