Survival Analysis

A common objective in many medical studies is to investigate the survival time of an individual after being diagnosed with a particular disease or health related condition. In most survival analysis studies the analysis is based on modeling the probability of survival. One of the goals in a survival analysis is usually to model the survival function. This chapter presents three different approaches for modeling a survival function. They are the Kaplan–Meier method of modeling a survival function, the Cox proportional hazards model for a survivor function, and the use of logistic regression for modeling a binary survival response variable. It is important to note that the proportional hazards model is based on the actual survival times and the explanatory variables, a proportional hazards model will provide more information about the survival probabilities than will either the Kaplan–Meier model or a logistic regression model.