Proportional Subdistribution Hazards Model for Competing Risks in Case-Cohort Studies

Competing risks refer to the situation where there are multiple causes of failure and occurrence one type event prohibits other types or alters chance observe them. In large cohort studies with long-term follow-up, often competing risks. When events rare, information on certain risk factors is difficult costly measure for full cohort, a case-cohort study design can be desirable approach. this paper, we consider semiparametric proportional subdistribution hazards model in presence studies. The function, unlike cause-specific gives advantage outlining marginal probability particular event. We propose estimating equations based inverse weighting techniques estimation parameters. methods, considered weighted availability indicator properly account sampling scheme. also proposed Breslow-type estimator cumulative baseline hazard function. resulting estimators shown, using empirical processes martingale properties, consistent asymptotically normally distributed. performance methods finite samples examined through simulation by considering different levels censoring interest percentages. results from scenarios suggest that parameter estimates reasonably close true values respective parameters model. Finally, applied sample Sister Study, which illustrated studying association between selected CpGs invasive breast cancer ductal carcinoma situ as risk.