A Multilevel Analysis to Uncover the Drivers of Infant Mortality in Indonesia

Abstract

Introduction: Infant mortality remains a critical concern in Indonesia due to its large population and regional health disparities. This study investigates the factors associated with infant mortality in Indonesia by considering the hierarchical data structure, in which individuals are nested within provinces. This study addresses a gap in previous research that has overlooked regional-level variation.

Method: This study utilized data from the 2017 Indonesian Demographic and Health Survey (IDHS). A multilevel mixed-effects logistic regression model was employed to account for both individual- and provincial-level variations in infant mortality. The multilevel approach allows for more accurate estimation by capturing unobserved heterogeneity across provinces. Adjusted Odds Ratios (AORs) with 95% Confidence Intervals (CIs) were reported to assess the strength and significance of associations between explanatory variables and infant mortality risk.

Result: The results indicate that mother's age at first childbirth (Adjusted Odds Ratio (AOR) = 1.262; 95% CI 1.180 - 1.350), marital status (AOR = 0.688; 95% CI 0.549 - 0.852), health insurance (AOR = 0.982; 95% CI 0.972 - 0.991), maternal occupation (AOR = 0.982; 95% CI 0.973 - 0.993), maternal education level (AOR = 1.147; 95% CI 1.103 – 1.192) household head’s age (AOR=0.713; 95% CI 0.654 – 0.777), and the birth order (AOR = 1.171; 95% CI 1.117 – 1.227) have significant impact on the risk of infant mortality.

Conclusion: Both maternal and household characteristics play an important role in driving infant mortality risks in Indonesia. The findings highlight the importance of considering regional heterogeneity in policy design. It is recommended that the government and relevant institutions prioritize efforts aimed at reducing infant mortality.