Delivery channels and socioeconomic inequalities in coverage of reproductive, maternal, newborn, and child health interventions: analysis of 36 cross-sectional surveys in low-income and middle-income countries.

backgroundGlobal reports have described inequalities in coverage of reproductive, maternal, newborn, and child health (RMNCH) interventions, but little is known about how socioeconomic inequality in intervention coverage varies across multiple low-income and middle-income countries (LMICs). We aimed to assess the association between wealth-related inequalities in coverage of RMNCH interventions.

methodsIn this cross-sectional study, we identified publicly available Demographic Health Surveys and Multiple Indicator Cluster Surveys from LMICs containing information on household characteristics, reproductive health, women's and children's health, nutrition, and mortality. We identified the most recent survey from the period 2010-19 for 36 countries that contained data for our preselected set of 18 RMNCH interventions. 21 countries also had information on two common malaria interventions. We classified interventions into four groups according to their predominant delivery channels: health facility based, community based, environmental, and culturally driven (including breastfeeding practices). Within each country, we derived wealth quintiles from information on household asset indices. We studied two summary measures of within-country wealth-related inequality: absolute inequalities (akin to coverage differences among children from wealthy and poor households) using the slope index of inequality (SII), and relative inequalities (akin to the ratio of coverage levels for wealthy and poor children) using the concentration index (CIX). Pro-poor inequalities are present when intervention coverage decreased with increasing household wealth, and pro-rich inequalities are present when intervention coverage increased as household wealth increased.

findingsAcross the 36 LMICs included in our analyses, coverage of most interventions had pro-rich patterns in most countries, except for two breastfeeding indicators that mostly had higher coverage among poor women, children and households than wealthy women, children, and households. Environmental interventions were the most unequal, particularly use of clean fuels, which had median levels of SII of 48·8 (8·6-85·7) and CIX of 67·0 (45·0-85·8). Interventions primarily delivered in health facilities-namely institutional childbirth (median SII 46·7 [23·1-63·3] and CIX 11·4 [4·5-23·4]) and antenatal care (median SII 26·7 [17·0-47·2] and CIX 10·0 [4·2-17·1])-also usually had pro-rich patterns. By comparison, primarily community-based interventions, including those against malaria, were more equitably distributed-eg, oral rehydration therapy (median SII 9·4 [2·9-19·0] and CIX 3·4 [1·3-25·0]) and polio immunisation (SII 12·1 [2·3-25·0] and CIX 3·1 [0·5-7·1]). Differences across the four types of delivery channels in terms of both inequality indices were significant (SII p=0·0052; CIX p=0·0048).

interpretationInterventions that are often delivered at community level are usually more equitably distributed than those primarily delivered in fixed facilities or those that require changes in the home environment. Policy makers need to learn from community delivery channels to promote more equitable access to all RMNCH interventions.

fundingBill & Melinda Gates Foundation and Wellcome Trust. TRANSLATIONS: For the French, Portuguese and Spanish translations of the abstract see Supplementary Materials section.