A machine learning logistic classifier approach for identifying the determinants of Under-5 child morbidity in Bangladesh

Background: Numerous biological, maternal, family, and socio-economic factors are regarded as influencers of child morbidity. The current study focuses on determining the factors that regulate the incidence of preventable outbreaks of disease or symptoms among under 5 children in Bangladesh. Methods: This study used data from the most recent nationally representable cross-sectional Bangladesh Multiple Indicator Cluster Survey (MICS) data which is conducted in 2019. In this study, the Machine Learning algorithm of Logistic Classifier has been applied to the information of 23,099 children age below 5 years. Result: In general, younger children, male children (OR = 1.087, 95% CI (1.012-1.167)), and children of the young mother are more likely to suffer from the disease than their counterparts. Children belonging to the household which have improved toilet (OR = 0.88, 95% CI (0.79-0.98)) and use salt that contains above 15 PPM (OR = 0.926, 95% CI (0.846-1.014)) are faced a lower risk of illness than those households which did not have improved toilet and used salt that contain lower than the 15 PPM iodine. Conclusion: Besides those household factors wealth quintile, household size, materials used for hand wash were found with a significant impact on morbidity. However, it is observed that water treatment practice alone had no impact on child morbidity. Accelerated initiatives that encourage access to and use of hygienic sanitation, the use of salt with adequate iodine, and a healthier household climate can be effective in the reduction of childhood morbidity in Bangladesh.