[Correspondence] Percutaneous coronary intervention for stable angina in ORBITA
The ORBITA trial1 by Rasha Al-Lamee and colleagues has a major statistical problem in the analysis of the primary endpoint. The mean pre-randomisation exercise time has a substantial treatment imbalance—it is 38·0 s higher in the percutaneous coronary intervention (PCI) group than in the placebo procedure group. Patients with lower baseline exercise times could have greater increases in exercise time 6 weeks later. This concept (called regression to the mean) is routinely adjusted for by doing an ANCOVA.