Decomposition Analysis of Socioeconomic Inequalities in Vaccination Dropout in Remote and Underserved Settings in Ethiopia

Abstract

ABSTRACT. Despite increments in immunization coverage over the past decades, substantial inequality due to wealth status has persisted in Ethiopia. This study aimed to decompose the concentration index into the contributions of individual factors to socioeconomic inequalities of childhood vaccination dropout in remote and underserved settings in Ethiopia by using a decomposition approach. A wealth index was developed by reducing 41 variables related to women’s household living standards into nine factors by using principal component analysis. The components were further totaled into a composite score and divided into five quintiles (poorest, poorer, middle, richer, and richest). Vaccination dropout was calculated as the proportion of children who did not get the pentavalent-3 vaccine among those who received the pentavalent-1 vaccine. The concentration index was used to estimate socioeconomic inequalities in childhood vaccination dropout, which was then decomposed to examine the factors contributing to socioeconomic inequalities in vaccination dropout. The overall concentration index was −0.179 (P <0.01), confirming the concentration of vaccination dropout among the lowest wealth strata. The decomposition analyses showed that wealth index significantly contributed to inequalities in vaccination dropout (49.7%). Place of residence also explained −16.2% of the inequality. Skilled birth attendance and availability of a health facility in the kebele (the lowest administrative government structure) also significantly contributed (33.6% and 12.6%, respectively) to inequalities in vaccination dropout. Wealth index, place of residence, skilled birth attendance, and availability of a health facility in the kebele largely contributed to the concentration of vaccination dropout among the lowest wealth strata. Policymakers should address vaccination inequality by designing more effective strategies.