Analysing the Effect of Gender on Mortality of COVID-19 Patients through Cox-Proportional Hazard Model

Arora, S.; Kumar, A.; Sambhav, S.

One of the major achievements and methodological advancements of biostatistics is the ability to capture censoring time to event data. Fitting distributions to the data is not always possible and not every time do data follow any of the distributions. Cox Proportional Hazard Model is a semi-parametric approach to compare the hazards or relative force of mortality of two lives (or two homogeneous groups of lives) without any information about the distribution of the data. This study is an attempt to analyze the effect of covariate: gender on the death rate of the Covid-19 patients by drawing a detailed comparison of the hazards of male lives and female lives. Later by noticing the pattern and shape of the graph of the hazard function, exponential distribution has been fitted to the data and the model is converted from semiparametric to fully parametric model by estimating the baseline hazard based on the distribution. And then the overall hazard of the two groups of lives. i.e., males and females have been compared. The effect of the covariate (gender) is tested statistically. An insight into the hazard of groups of lives can tell which lives are more exposed to the risk and timely actions and preventions can be taken by those who are at higher risk, to minimize the hazard rate. © 2021 IEEE.