Back

Demography, hygiene and previous disease prevalence as plausible risk factors associated with Covid-19 deaths across Indian states

Chatterjee, B.; Mande, S. C.

2023-06-29 epidemiology
10.1101/2023.06.22.23291740 medRxiv
Show abstract

Severity of Covid-19 diseases has been disproportionate with higher case-fatality ratio affecting developed nations. In India, states with higher income have reported more number of deaths compared to lower income states. The global burden of diseases India 2019 and the National Health Profile 2019 data was used to draw correlations with Covid-19 mortality at two different dates of peak Covid-19 cases in India. We explored correlation of mortality in different states of India with prevalence of different diseases, demography, development, sanitation etc. The study found a positive correlation with known demographic parameters such as percentage of elderly population(spearman correlation coefficient(rho) =0.44 and 0.46 with 1st and 2nd peak respectively). Similarly, percentage urbanization was seen to correlate well with mortality(rho=0.71 and 0.57) suggesting Covid-19 to be a predominantly urban disease. Prevalence of Autoimmune diseases, and Cancer show higher correlation with deaths. A surprising positive correlation emerged between improved sanitation parameters, such as closed drainage and indoor toilets, with COVID-19 deaths. Overall the multivariate regression model achieved by combining demography, sanitation, autoimmune diseases and cancer gave us the best prediction for Covid-19 mortality(adjusted R square value of 0.71 with peak 1 and 0.85 with peak 2). Analysis of the Covid-19 related data seems to indicate that as the wealth of a state increases, the states urban landscape changes often leading to better sanitation facilities. The lifestyle and prevalence to autoimmune diseases as well as cancer also increases. However, this may affect the states ability to fight pandemics due to lower exposure to pathogens and immune training.

Matching journals

The top 5 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.