Determinants of Infant Diarrhoea and Respiratory Illness in a Low‐Resource Community in South Africa: A Structural Equation Modelling

BACKGROUND: Diarrhoea and acute respiratory infections continue to cause significant morbidity in children under 5 years, influenced by multiple factors. This secondary data analysis of the Eggcel randomised controlled study aimed to explore the association of anthropometric status, food insecurity, socioeconomic and environmental factors with diarrhoea and respiratory morbidity in children aged 6-9 months from a low-resource community in South Africa. METHODS: We utilised data from 216 infants in the control group who completed the Eggcel-growth randomised controlled trial in Jouberton, South Africa. Baseline data on maternal and infant demographics, household socioeconomic and environmental conditions were collected using a questionnaire. Food insecurity was assessed through the Food Insecurity Experience Scale (FIES). Anthropometric measurements, including weight, height and head circumference, taken at baseline were used. Morbidity symptoms were recorded daily by mothers using a visual wellness diary. The exposure latent variables included anthropometric status, environmental conditions, food insecurity and socioeconomic status. Outcome variables were diarrhoea or respiratory episodes over the 6-month follow-up period. Structural equation modelling (SEM) was performed using R version 4.5.0. RESULTS: Infants with diarrhoea had higher stunting and underweight rates at 28/75 (37.3%) and 9/75 (12.0%), respectively, and the highest access to safe water at 74/75 (98.7%). Higher anthropometric status was associated with a 31% reduction in diarrhoea (IRR = 0.69; 95% CI: 0.50, 0.95; p = 0.022). Respiratory episodes were not associated with any of the latent factors. CONCLUSION: Nutritional status was an important determinant of diarrhoeal morbidity among infants in a low socioeconomic community but was not associated with respiratory morbidity. These findings suggest context-specific pathways through which nutritional, food insecurity and environmental factors may be associated with childhood morbidity.