DESIGN AND ANALYSIS FOR SURVIVAL-DATA UNDER ORDER RESTRICTIONS WITH AMODIFIED LOGRANK TEST

We previously proposed a class of ordered weighted logrank tests for a nalysing censored survival data under order restrictions, However, the power of these tests is asymmetrical with respect to possible alterna tive configurations, While it is superior in most cases, the power can be inferior to the non-ordered logrank test in extreme cases. We prop ose a modified ordered logrank test which performs uniformly better th an the non-ordered test. The power of the modified test is equivalent to the generalized Jonckheere's test but its computation is much simpl er, We give sample size requirements for sufficient power to reject th e global null hypothesis at specified hazard ratios between the contro l group and the best group. Following Fisher's least significant diffe rence (LSD) strategy for multiple comparisons, power investigations in dicate that the nominal power for the global test carries over to the control versus best comparison during pairwise testing. The power for detecting intermediate survival differences is inadequate but the samp le sizes required to detect such differences may be impractical in mos t applications, (C) 1998 John Wiley & Sons, Ltd.