Regression on cause-specific hazard in Competing Risks for Survival analysis

Abstract

In the study of  survival analysis, Kaplan–Meier (KM) estimates of survival curves and Cox proportional hazard models were widely used to describe survival trends and identify significant prognostic factors. All these statistical analyses deal with only one type of event like death, independently of its cause. Most part of the analysis are all  based on the hazards. In a competing risks setting, There are two types of hazard namely, the cause-specific hazard and the sub distribution hazard. The cause-specific hazard is the instantaneous risk of progression to a specific type of event, conditionally on being at risk for experiencing that event, generally hazards are not the probabilities, they are the rates., and used to quantify the probability of the crude risk or cause-specific cumulative incidence to progress to a specific cause by time t, When estimating the cause-specific hazard, the individuals experienced a competing event leaving behind the risk set. The cause-specific regression model was applied on to the real data sets and  try to shed light on causal mechanisms and prediction at the population level.