A framework for combining morbidity and mortality to identify determinants of child health.

Objective: The objective of this study is to combine morbidity in terms of various levels of growth faltering or malnutrition among surviving children with mortality to better understand the determinants of infant and child health. Method: This study uses data from the latest National Family Health Survey (NFHS)-5 of India conducted during 2019-21 to provide a framework that integrates both mortality and morbidity as a measure of infant and child health to investigate their association with maternal and other socio-economic determinants of health. The outcome variable representing health status of a child is defined as a categorical variable with four progressively worsening levels- no malnutrition, moderate malnutrition, severe malnutrition, and not alive. The levels of morbidity in terms of growth faltering or malnutrition are defined in line with the World Health Organization (WHO) guidelines. Cumulative Link Mixed Model (CLMM) with partial proportional odds framework has been used to assess the association of risk factors with the outcome variable. Multivariate Imputations by Chained Equation has been employed to address high amount of missingness in three risk factors. Parameter estimates from CLMMs fitted to each of the 20 multiply imputed datasets were combined using Rubin's rules to obtain pooled estimates. Results: Among the examined determinants, maternal education and household wealth index demonstrated the strongest associations with child health outcomes. Notably, low birth weight and non-institutional delivery are consistently strong risk factors across all thresholds of child health status, from moderate malnutrition till mortality. Antenatal care incompleteness does not associate with whether a child enters moderate malnutrition, but it strongly associates (14-19% higher odds) with severe malnutrition and with death, confirming its role in preventing the worse child health outcomes. Conclusion: Combining child mortality and morbidity allows better indication of child health status with added sensitivity through progressive health status levels, ranging from best scenario (no malnutrition) to worst scenario (death). Consequently, it enables a more comprehensive assessment of maternal and socioeconomic determinants of child health. In addition, using such combined outcome ensures complete utilization of information available in a survey dataset, irrespective of the mortality status of the child.