Thesis (M.Sc.) - Cairo University - Institute of Statistical Studies and Research - Department of Statistics and Econometrics

A topic that has received attention in both the statistical and medical literature is the estimation of the probability of failure for endpoints that are subject to competing risks. Methods of standard survival analysis such as the complement of Kaplan-Meier (KM) for estimation of cumulative incidence curves (1-KM), the log-rank test for comparison of these curves, and the standard Cox model for the assessment of covariates, are still used by medical researchers, however lead to biased and un interpretable results when an analysis focuses time-to-first event in scenarios where there are competing events. In thesis these standard survival methods are compared with other alternative methods as the cumulative incidence competing risks (CICR) curves, the Gray's test for comparing these curves, and the competing risks regression model (CRR).The data used included 348 patients with primary cancer of the oral cavity diagnosed at Tanta cancer center (TCC) during the period 2003-2007, data were analyzed using R program. The main aim is to attract the attention of medical researchers to use the appropriate statistical technique according to the nature of the available taking into consideration the presence of the competing risks. The Results showed that the incidence of each main outcome was overestimated by the standard methods

Issued also as CD