Survival Analysis Using SAS®: A Practical Guide

Testing for Differences in Survivor Functions

If an experimental treatment has been applied to one group but not another, the obvious question to ask is “Did the treatment make a difference in the survival experience of the two groups?” Since the survivor function gives a complete accounting of the survival experience of each group, a natural approach to answering this question is to test the null hypothesis that the survivor functions are the same in the two groups; that is, S₁(t) = S₂(t) for all t, where the subscripts distinguish the two groups. PROC LIFETEST calculates two alternative statistics for testing this null hypothesis: the log-rank test (also known as the Mantel-Haenszel test) and the Wilcoxon test. A third test, the likelihood-ratio ...