Multilevel Analysis of Diarrhea and Its Determinants Among Children Under Five in Mozambique, 2022–2023

Background and Purpose: Childhood diarrhea remains one of the leading causes of illness and death among children under five in low- and middle-income countries, including Mozambique. It is also one of the most preventable and treatable childhood illnesses, making its persistence a pressing concern. Guided by the socioecological model, this study examined risk and protective factors at individual, household, and community levels, with a focus on how place—including environmental, cultural, and infrastructural dimensions—shapes health outcomes. Particular attention was given to water, sanitation, and hygiene (WASH) conditions and how they intersect with geography and caregiving practices across diverse regions.

Methods: Data were drawn from the 2022–2023 Mozambique Demographic and Health Survey (DHS), with a weighted analytic sample of 4,890 children under age five. Multilevel logistic regression was used to account for children nested within communities (primary sampling units), allowing for the estimation of both individual- and community-level influences. Predictor variables included child age and sex, maternal age and education, household wealth, sanitation, presence of observed handwashing facilities, urban/rural status, and region. A cross-level interaction between handwashing and urban residence was tested. Model fit was evaluated using deviance, Akaike Information Criterion (AIC), and Bayesian Information Criterion (BIC).

Results: Approximately 10.6% of children had diarrhea in the past two weeks. Children aged 7–12 months had over three times the odds of diarrhea compared to those 0–6 months (OR = 3.02, 95% CI [2.11, 4.32], p < .001), and those aged 13–24 months also had elevated odds (OR = 2.30, 95% CI [1.65, 3.20], p < .001). Observed handwashing facilities were protective (OR = 0.54, 95% CI [0.38, 0.77], p = .001), as was maternal age over 25 years. While Maputo Province—a coastal, urban, and more resourced region—showed reduced risk, the inland, rural province of Manica, situated along the Zimbabwean border over 1,000 kilometers from the capital, also demonstrated significantly lower odds (OR = 0.44, 95% CI [0.25, 0.79], p = .006). Although the handwashing × urban interaction was not statistically significant, its inclusion improved model fit. The intraclass correlation coefficient (ICC = .12) suggested meaningful variation across communities.

Conclusions and Implications: This analysis illustrates that health is not only shaped by access to infrastructure but also by the conditions and customs of place. Many WASH interventions focus on delivering physical resources—latrines, water taps, soap—while overlooking how families use, interpret, or adapt them within their environment. Mozambique’s diverse landscape—coastal and inland, urban and rural—reflects the complex intersections of its political history, cultural knowledge, and ecological systems. Progress toward SDG 6 (clean water and sanitation) and SDG 3 (health for all) requires regionally responsive strategies. As the World Health Organization emphasizes, reducing childhood deaths from diarrhea depends on integrated solutions—safe water, sanitation, and local caregiving practices—rooted in the contextual realities of communities.