Survival Analysis and Predictive Modelling Approach for COVID-19

Abstract

This research article focuses on studying the lives infected by the virus. A comparison is made
between the probability of dying due to Covid-19 in India and world, and among the different
countries for the period 30th January 2020 to 11th May 2020 i.e. where there are 50000+ cases
registered by 11th May 2020. A survival function (in two cases: survival rate (SR) and survival score
(SC)) pattern of total cases of Covid-19 infected patients in different countries is also studied. The
Kaplan-Meier Analysis has been carried out to achieve survival function. The mortality rate pattern
of India is also discussed in the article. Also a SEIR-D Model is used to estimate the future spread of
Covid-19 in India and China.
Findings:
Regarding probability of dying, it is found that UK is under high risk with 0.44921. But China among
all the countries is performing best with 0.020603. Also China performing best survival function in
both the cases with SR of 0.99153 and SC of 92.69268, India stands at 7th position with SR of
0.920469 and 4th with SC of 66.49281. Italy has the lowest SR of 0.0751 and Turkey is at last place
with SC of 39.78197. It is also observed that after the age of 80, there is a difference of 14.80% in the
probability of dying of a healthy and a Covid-19 infected person. In India, SEIR-D Model predicts
that 0% change in deaths can be seen by September.