Application of new two‐sample tests to data from a randomized placebo‐controlled heart‐failure trial

Abstract

In a randomized placebo-controlled double-blind trial of 230 congestive heart failure patients, four treatments were evaluated for efficacy, with exercise tolerance time (ETT) as the primary outcome. Various two-sample tests were applied to the analysis of ETT data. It is shown in this paper that the conventional two-sample tests (t and rank-sum) are insensitive to situations where the effect of the experimental therapy is not consistent across a patient population. Tests recommended by O'Brien are more appropriate for these data. It is also shown that the application of the O'Brien tests led to the identification of sub-groups where the observed effect of the experimental therapy was most pronounced.

On the use of the generalized t and generalized rank‐sum statistics in medical research

“…A test of p1 in (5) is, of course, equivalent to the usual t test; hence the term generalized t test for model (4). Throughout this paper the term conditional generalized test will be applied to the procedure recommended by O'Brien.…”

Generalized treatment effects for clinical trials

“…We do note that the di!erence could become statistically signi"cant with larger sample sizes than the 40}60 per treatment group in this analysis. We also note that there may be some asymmetry in the distribution of ETT changes for the milrinone group [21], suggesting a non-parametric approach may be worth considering here. (The p-value for M versus D by the rank sum test is 0.76 [21].…”

“…Tandon et al [21] compared O'Brien's methods to standard methods in the analysis of a randomized clinical trial of four treatments for congestive heart failure. The design was a parallel comparison of three experimental treatments versus a control (digoxin).…”