Negative selection bias for women inclusion in a clinical trial

Introduction Despite the growing awareness towards the importance of adequate representation of women in clinical trials among patients treated with percutaneous coronary intervention (PCI), available evidence continues to demonstrate a skewed distribution of study populations in favour of men. Methods and results In this pre-specified analysis from the MASTER DAPT screening log and trial, we aimed to investigate the existence of a negative selection bias for women inclusion in a randomized clinical trial. A total of 2847 consecutive patients who underwent coronary revascularization across 65 participating sites, during a median of 14 days, were entered in the screening log, including 1749 (61.4%) non-high bleeding risk (HBR) and 1098 (38.6%) HBR patients, of whom 109 (9.9%) consented for trial participation. Female patients were less represented in consented versus non-consented HBR patients (22% versus 30%, absolute standardized difference: 0.18) and among non-consented eligible versus consented eligible patients (absolute standardized difference 0.14). The observed sex gap was primarily due investigators' choice not to offer study participation to females because deemed at very high risk of bleeding and/or ischemic complications, and only marginally to a slightly higher propensity of females compared to males to refuse study participation. Conclusions Female HBR patients undergoing PCI are less prevalent, but also less likely to participate in the trial than male patients, mainly due to investigators' preference.