Structural equation modelling (SEM) for malaria prevalence and risk factors in Uganda

Malaria remains a leading cause of morbidity and mortality among children under five years in sub-Saharan Africa, despite extensive public health efforts. Its transmission is influenced by a complex interaction of socioeconomic, environmental, maternal, and child health factors. Traditional analytic approaches often fail to capture these multifaceted relationships. This study employs Structural Equation Modelling (SEM) to explore the latent and observed predictors of child malaria prevalence, offering a comprehensive understanding of the underlying pathways.

Utilizing secondary data from the 2018-2019 Uganda Malaria Indicator Survey (MIS), a SEM framework was constructed comprising four latent constructs: Socioeconomic Status (SES), Environment, Maternal Health, and Child Health. Each construct was defined by multiple observed variables, and the outcome of interest was the malaria status of the child. Factor loadings and regression coefficients were estimated using standardized model parameters. Model fit was assessed using indices including the Comparative Fit Index (CFI), Tucker-Lewis Index (TLI), Root Mean Square Error of Approximation (RMSEA), and Standardized Root Mean Square Residual (SRMR).

The SEM analysis demonstrated significant pathways from latent variables to child malaria prevalence. Child Health had the strongest positive association with malaria status (β = 0.22, p < 0.001), followed by a marginal negative association with the Environment construct (β = - 0.36, p = 0.056). Socioeconomic Status and Maternal Health were not statistically significant predictors. Model fit indices suggested a moderately acceptable fit (CFI = 0.786, TLI = 0.630, RMSEA = 0.064, SRMR = 0.053), indicating that the conceptual framework captured meaningful relationships.

This study highlights the value of SEM in disentangling the intricate network of factors influencing child malaria. The findings underscore the importance of child-level health conditions and environmental factors, while pointing to the limited direct effect of socioeconomic and maternal health variables. These insights can inform integrated intervention strategies that simultaneously address child health and environmental exposures to reduce the burden of malaria.