Overestimation of the probability of death on peritoneal dialysis by the Kaplan-Meier method: advantages of a competing risks approach

摘要: Background: In survival analysis, patients on peritoneal dialysis are confronted with three different outcomes: transfer to hemodialysis, renal transplantation, or death. The Kaplan-Meier method takes into account one event only, so whether it adequately considers these risks is questionable. more recent competing has been shown be appropriate in analyzing such situations. Methods: We compared the estimations obtained by and (namely Kalbfleisch Prentice approach), 383 consecutive incident patients. By means of simulations, we then two virtual centers where had exactly same probability only difference between was transplantation available not. Results: At five years, 107 (27.9%) died, 109 (28.4%) transferred 91 (23.8%) transplanted, 37 (9.7%) were still alive dialysis; before 39 (10.2%) censored dialysis. five-year probabilities estimated methods respectively: death: 50% versus 30%; hemodialysis: 59% 32%; transplantation: 39% 26%; event-free survival: 12% 12%. sum exceeded 100%, implying that could experience than event, death for example, which impossible. increased as increased, although actually remained constant. Conclusion: appears estimating events context univariable analysis.