An extended Cox Proportional Hazards model and it s application in COVID-19 patients

The Cox Proportional Hazards (CPH) model is one of the most popular regression techniques in survival analysis, which is used to relate several risk factors or exposures, considered simultaneously, to survival time. An essential assumption in this model is that the hazard ratio over time must be constant. There are some solutions when this assumption are not met. In this paper, we proposed an extended CPH model and applied it to medical record data of some COVID-19 patients that is hospitalized from 2019 to 2021. The results indicate that there is a linear relationship between therapy via three drugs including Azithromycin, Favipiravir, and Remdesivir and survival time of hospitalized patients. This method can be used for study the treatment e ect of other drugs.