A New Testing Approach for Comparing the Overall Homogeneity of Survival Curves

The assessment of overall homogeneity of time-to-event curves is a key element in survival analysis. The currently commonly used methods, e.g., log-rank and Wilcoxon tests, may have a significant loss of statistical testing power under certain circumstances. In this paper a new statistical testing approach is developed to compare the overall homogeneity of survival curves. The proposed new method has greater power than the commonly used tests to detect overall differences between crossing survival curves. The small-sample performance of the new test is investigated under a variety of situations by means of Monte Carlo simulations. Furthermore, the applicability of the proposed testing approach is illustrated by a real data example from a kidney dialysis trial. (© 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim)