The mediating role of behavioural and socio-structural factors on the association between household wealth and childhood malaria in Ghana.

Background: Children under five continue to bear a disproportionate burden of malaria morbidity and mortality in endemic countries. While the link between socioeconomic position (SEP) and malaria is well established, the causal pathways remain poorly understood, hindering the design and implementation of more targeted structural interventions. This study examines the association between SEP and malaria among children in Ghana and explores the potential mediating role of behavioural and socio-structural factors.

Methods: Data from the Ghana Demographic and Health Survey (DHS) 2022 were analysed. As part of the survey, children were tested for malaria using a rapid diagnostic test (RDT), and SEP was measured using a household asset-based wealth index. Mediation analysis (MA) using a regression-based approach was performed to assess mediated effects between SEP and malaria in children under five in Ghana through housing quality, educational attainment (EA), long-lasting insecticidal net (LLIN) use, indoor residual spraying (IRS), and healthcare-seeking behaviour (HSB). Reported are the total natural indirect effects (TNIEs) and the proportion mediated (PM).

Results: Of the 3,884 children included in the survey, 19.4% (757) had malaria. Belonging to a household with high SEP was associated with a 43% lower risk of malaria (Prevalence Ratio, PR = 0.57; 95% Confidence Interval, CI 0.46-0.71). Regarding indirect (mediated) effects, maternal EA of secondary school or higher (OR = 0.68; 95% CI 0.60-0.77; PM = 17.5%), improved housing (OR = 0.80; 95% CI 0.68-0.91, PM = 9.2%), LLIN use (OR = 0.95; 95% CI 0.90-0.99, PM = 2.1%) partially mediated the association between SEP and malaria. The combined effect of all three mediators was higher than those in a single mediator or two sequential mediators (with EA as the initial mediator) (OR = 0.58; 95% CI 0.51-0.68, PM = 25.7%). No evidence of mediation was observed for HSB and IRS.

Conclusion: We found evidence of mediation by EA, housing, LLIN use and IRS, suggesting that current biomedical and behavioural malaria control efforts could be complemented with structural interventions, such as improved housing and education. Future studies that test the effect of different or joint effects of multiple mediators based on prospective designs are recommended to strengthen the evidence.