A survival analysis approach for identifying the risk factors in time to recovery of COVID-19 patients using Cox proportional hazard model

The coronavirus pandemic was a global health crisis taking away millions of lives worldwide. People diseased by the virus, differ in the extent of severity of the infection. While it turns out to be fatal for some, for several others the extent of severity is as ordinary as common cold. These people are reported to have recovered from the disease without hospitalization and consuming some relevant medicine and home remedies. But people who have comorbidity like geriatric, high blood pressure, heart and lung problems, diabetes, cancer etc. are at high risk of developing serious illness from the infection. This study is an application of the Cox proportional hazard model with an aim to identify the risk factors that affect the recovery time of the COVID-19 patients. The model is an advanced regression technique that can be utilized to evaluate simultaneously the effect of several factors on the possibility of instantaneous failure in patients. The paper also uses the Mantel-Haenszel test (Log-Rank test) to compare if the probability of survival of different treatment procedures or different groups of patients differ significantly. The information is collected from 129 respondents of Assam, India. The study identifies that the significant risk factors that prolong the recovery time from COVID-19 are pre-disease, location, and food habits.