[Correspondence] Percutaneous coronary intervention for stable angina in ORBITA
We applaud ORBITA1 by Rasha Al-Lamee and colleagues as a well executed first placebo-controlled trial of the incremental impact of percutaneous coronary intervention (PCI) on exercise time (ET) in stable angina. However, given the study’s importance, we were surprised by the statistics. By use of an unpaired T test, the authors compared change in ET between treatment groups and found no difference (16·6 s, 95% CI −8·9 to 42·0, p=0·20). However, in parallel randomised trials, mixed effects models such as ANCOVA are indicated because they are more precise than the unpaired T test, correct for baseline imbalances that predispose to regression to the mean, and confer increased statistical power to detect small but clinically meaningful differences.