Prediction of Hospital Visits for Respiratory Morbidity Due to Air Pollutants in Lucknow

Various epidemiological and toxicological studies have shown an association between air pollutants and their risk of respiratory morbidity and mortality. Only a few studies have been conducted in India, which evaluates the impact of seasonal air pollution on data on respiratory morbidity. Machine learning models like random forest regression are employed in the present context to predict the change in number of hospital visits for respiratory morbidity associated with the change in concentration of various air pollutants in the atmosphere and to study the effect of potential confounders like temperature and humidity, and also seasonal effect in Lucknow, India, for period of 2017–18. The results of the model revealed that a decrease of 16 patients daily is predicted if there is a reduction in the ambient concentration of PM2.5 to National Ambient Air Quality Standards (NAAQS) in the city of Lucknow in one government hospital. SO2 increases the number of respiratory patients as its ambient air concentration increases. It is observed that with 2 µg/m3 increase from 18 to 20 µg/m3 increased to nearly four patients. The synergistic effect of PM2.5 and NO2 is the most harmful for the citizens of Lucknow City. This study provides evidence that respiratory morbidity increases with an increase in the concentration of air pollutants in Lucknow. The post-monsoon season is considered as the most polluted period of the year with a higher number of hospital visits.