Factors affecting External causes of deaths among adults (15–59 years) in South Indian states: A study using Bayesian model on National Family Health Survey-4 (2015–16) data

Abstract

The study aimed at finding risk factors associated with adult mortality (15–59 years) due to external causes (accidents, suicide, poisoning, homicide, and violence) of the South Indian states by using the fourth round of National Family Health Survey data. Despite having a higher literacy rate and comparatively being better developed than other Indian states, the number of external causes of deaths in the South Indian states is alarming. The study used Bayesian Logistic Regression method for analysis. Three Chain of Markov Chain and Monte Carlo (MCMC) model was used. The sex of the deceased was found to be an important factor in all 5 states (Tamil Nadu, Karnataka, Kerala, Tamil Nadu, Telangana and Andhra Pradesh). Place of residence was an important factor for all states except Karnataka. The prevalence of external adult deaths was high among female-headed households irrespective of all the south Indian states. The family members in Karnataka and Andhra Pradesh, wealth index in Telangana, drinking habit in Kerala, car access in Kerala and Tamil Nadu and mass-media access in Tamil Nadu were found significant factors (the model converged for the factors) that affecting external causes of deaths. Some holistic efforts should be undertaken by the Government for adult deaths by spreading awareness among adult, setting up specialized teams for policy-making and timely measures to curtail the underlying causes of such unnatural deaths and to ameliorate the overall situation.