STATISTICAL POWER OF AN ALTERNATIVE TEST FOR COMPARISON OF SURVIVAL DISTRIBUTIONS WITH CROSSED SURVIVAL CURVES IN THE PRESENCE OF CENSORING: A SIMULATION STUDY

In this paper, we developed a new method that provides an alternative to some existing methods of assessing the overall homogeneity of Kaplan-Meier survival curves under low and high censoring rates. The performance of the test statistics was measured based on the statistical powers of the tests. Further, a Monte Carlo simulation study was conducted to assess the relative performance of the existing methods. Also, we numerically compared its performance with Logrank, Wilcoxon, Tarone-Ware, Peto-Peto, Modified Peto-Peto and the Fleming-Harrington (1, 1) tests. Simulation was used to evaluate and compare the performance of our new method to the existing ones. The result showed that the new method has robust performance compared to the six popular tests with relatively high power to detect differences when the censoring rates in the groups are low and high. The statistical power of the new method was comparable to the existing ones therefore, it is a worthy alternative to some of the existing methods.