A Multinomial method to analysis the patterns of childhood co-morbidity in India

Terminal illness in children in developing countries is frequently characterized by the co-occurrence of more than one disease- a phenomenon referred to as Co-morbidity. Using self-reported illness data from the 2005-06 India Demographic and Health survey, we applied a multinomial model to access the risk factors of childhood Co-morbidity of fever, diarrhoea and acute respiratory infection (ARI). Various models were fitted and compared using the Akaike information criterion (AIC) and Bayesian information criterion (BIC). Specifically, risk factors associated with child Co-morbidity included age of the child, birth order, and vitamin A, sex of children, breastfeeding, household water supply, toilet facility, rural-urban place of residence and six regions of India. The result show that age of child, sex, vitamin A, place of residence (urban/rural) and regions are important predictor not only for co-morbidity but also for single disease. The results emphasis the need for interventions targeted at this group, and may include micro-nutrient supplements, e.g. vitamin A, and use of insecticides treated nets, and combined intervention could be cost-effective to implement. Rural areas are particularly vulnerable and deserve attention when scaling up-interventions.