COMPARISON OF PARAMETRIC AND NONPARAMETRIC SURVIVAL METHODS USING SIMULATED CLINICAL-DATA

We derived three parametric survival models (the log-normal, log legit , and Weibull) from the clinical data of chemotherapy trials for stage II breast cancer. We then used these models to generate simulated sur vival data, which we analysed using both parametric (log-normal) and n on-parametric (logrank, Gray-Tsiatis and Laska-Meisner) methods. With limited follow-up (5 years), the non-parametric tests had greater powe r than the log-normal model. This advantage diminished, however, with extended follow-up (15 years). Furthermore, only the log-normal model could distinguish reliably a survival advantage due to an increase in cured fraction from an advantage due to an increase in time to failure . (C) 1997 by John Wiley & Sons, Ltd.