Statistical Analysis of Competing Risks: Overall Survival in a Group of Chronic Myeloid Leukemia Patients

Abstrakt

Outcomes of specific methods designed for competing risks analysis are compared with the those obtained using standard methods of survival analysis. The effect of Sokal score classification is found ambiguous. While the score should identify high- and low-risk CML patients it appears to be predictive only for the failure due to other causes than CML. The article stresses the inportance of careful consideration of the censoring process as well as the use of appropriate methods for analysing censored data.