The associations between housing quality and acute lower respiratory infection among under-5 children: a multi-country analysis across sub-Saharan Africa

Abstract

Background/objectives: Acute lower respiratory infection (ALRI) among children under five years old (under-5) is a leading risk factor for disease burden in sub-Saharan Africa (SSA), but its association with housing quality is not fully explored. We aim to investigate their spatial and temporal associations within and across 34 SSA countries. Methods: Data were extracted from the Demographic and Health Surveys (DHS) conducted 1990-2018 that collected data on both housing and ALRI. Housing quality were derived as a binary variable: 1) finished and unfinished built based on materials used for wall, roof and floor; 2) improved and unimproved house based on data on built materials as well as access to safe water and sanitation. ALRI was defined as having cough in combination with short/rapid breathing, or difficult breathing at any time in the two weeks preceding the surveys. Conditional logistic regression was ran to model the associations in each country and then the odds ratios (OR) were pooled via meta-analysis, controlling for cluster-level effects and adjusting for child’s age, sex, breastfeeding, size at birth, maternal education and type of cooking fuel. Results: Preliminary analysis currently included the most recent DHS surveys from 12 SSA countries, consisting of 56,519 under-5 children. The percentage of urban residence ranged from 16% to 61%; majority of mothers were educated up to primary-school level and most households did not have access to clean cooking fuel. Prevalence of unfinished built ranged from 43% to 95% across countries whilst using the stricter definition, over 90% of houses in each country were classified as unimproved. Pooling all country-specific ORs, we observed a higher ALRI risk in association with housing with unfinished built (OR: 1.07, 95%CI: 0.99-1.16). Conclusion: Poor housing quality is associated with higher ALRI prevalence in SSA and necessary interventions should be considered to reduce the ALRI burden.