'It depends': The complexity of allowing residents to fail from the perspective of clinical supervisors.

Clinical supervisors acknowledge that they sometimes allow trainees to fail for educational purposes. What remains unknown is how supervisors decide whether to allow failure in a specific instance. Given the high stakes nature of these decisions, such knowledge is necessary to inform conversations about this educationally powerful and clinically delicate phenomenon. 19 supervisors participated in semi-structured interviews to explore how they view their decision to allow failure in clinical training. Following constructivist grounded theory methodology, the iteratively collected data and analysis were informed by theoretical sampling. Recalling instances when they considered allowing residents to fail for educational purposes, supervisors characterized these as intuitive, in-the-moment decisions. In their post hoc reflections, they could articulate four factors that they believed influenced these decisions: patient, supervisor, trainee, and environmental factors. While patient factors were reported as primary, the factors appear to interact in dynamic and nonlinear ways, such that supervisory decisions about allowing failure may not be predictable from one situation to the next. Clinical supervisors make many decisions in the moment, and allowing resident failure appears to be one of them. Upon reflection, supervisors understand their decisions to be shaped by recurring factors in the clinical training environment. The complex interplay among these factors renders predicting such decisions difficult, if not impossible. However, having a language for these dynamic factors can support clinical educators to have meaningful discussions about this high-stakes educational strategy. [ABSTRACT FROM AUTHOR]