Surgery Residents Who Perform Best May Get the Least Feedback, Study Finds

When a senior surgeon watches a resident perform an operation, a quiet calculation takes place: how much of this task can this trainee be trusted to do alone? That judgment, known in medical education as entrustment, has become the backbone of how general surgery residents in the United States are now assessed. A new multi-institution study of more than a thousand workplace assessments suggests that the residents who earn the highest levels of trust are also the ones least likely to receive written narrative feedback — a paradox that could leave the most advanced trainees flying blind precisely when their skills are being fine-tuned.

The study, published in Global Surgical Education, the journal of the Association for Surgical Education, analyzed 1,071 Entrustable Professional Activity assessments completed by faculty for general surgery trainees between July 2023 and June 2025. Entrustable Professional Activities, or EPAs, are defined units of professional work — such as performing an operation, managing a critically ill patient, or coordinating pre-operative care — that supervisors rate on an entrustment scale ranging from watching the resident to granting full independence. The framework was adopted nationally in general surgery as part of the profession’s shift toward competency-based medical education, replacing time-based measures with direct observations of what a resident can actually be trusted to do.

The research team, led by Gabrielle Moore of the University of Utah Spencer Fox Eccles School of Medicine, applied mixed-effects ordinal regression models to the assessment data. This statistical approach is well suited to the structure of the dataset because it accounts for the fact that multiple assessments are nested within the same residents and the same faculty evaluators, and because entrustment ratings are ordered categories rather than continuous numbers. The models examined how the odds of receiving a higher entrustment rating changed with the presence of narrative feedback, post-graduate year, case complexity, the phase of care being assessed, and the demographics of both trainees and evaluators.

The headline finding was stark. Of the 1,071 assessments, 896 — 83.7 percent — contained narrative feedback, while 175, or 16.3 percent, had none at all. But residents rated at higher levels of entrustment for a given activity were significantly less likely to have received written comments: the odds ratio was 0.44, meaning the odds of receiving narrative feedback were less than half for those at higher entrustment levels compared with those rated lower, a difference that was highly statistically significant. In other words, the trainees who most impressed their supervisors got the least written explanation of why.

The pattern has a plausible psychological explanation, though the study design cannot prove causation. Faculty may feel that a resident performing at an expected or advanced level needs no justification, while a struggling trainee prompts a written note. Alternatively, supervisors may simply be more motivated to document concerns than to document excellence. Either way, the result runs against one of the central promises of the EPA framework — that every entrustment decision should be accompanied by actionable, individualized feedback that tells the trainee what to do next. Without narrative comments, a high entrustment rating is a verdict without a rationale, and a resident cannot easily convert it into improvement.

The analysis also mapped how entrustment varies across the arc of training and the texture of clinical work, and here the results were more reassuring. The likelihood of a higher entrustment rating increased significantly with post-graduate year, peaking in the senior stages of residency — evidence that the assessment system tracks genuine progression rather than assigning ratings at random. Trainees operating on straightforward cases were nearly twice as likely to earn higher entrustment than those in complex cases, with an odds ratio of 1.94. That gradient makes sense: a routine procedure offers fewer opportunities for things to go wrong and fewer moments where a supervisor must intervene, but it also raises the question of whether the rating reflects the resident or the operation.

Phase of care mattered even more dramatically. Assessments conducted before and after operations were far more likely to carry high entrustment ratings than those conducted intraoperatively: the odds ratio was 4.94 for pre-operative assessments and 3.04 for post-operative assessments, both statistically significant. The operating room, it turns out, is the hardest place to earn trust — unsurprising, given that intraoperative performance unfolds in real time under pressure, with the attending surgeon literally scrubbed in and sharing responsibility for the patient on the table. But the finding also hints that faculty may find it easier to delegate cognitive and coordinative tasks, such as planning and post-operative management, than to hand over the scalpel.

One of the most consequential results concerns fairness. Previous research in surgical education has documented demographic disparities in operative autonomy and assessment, including gender differences in entrustment and in the language faculty use to describe trainees. This study, however, found no difference in entrustment based on trainee or faculty demographics once post-graduate year and case complexity were controlled for statistically. That is an important methodological caveat: raw differences between groups can disappear when the model properly accounts for how much training a resident has had and how difficult their cases are. The finding offers cautious reassurance that the EPA system, at least in this sample, is rating the work rather than the person — though the authors and the broader literature agree that continued surveillance is warranted as implementation matures.

The study arrives at a moment when EPAs are moving from pilot projects to routine practice in American surgical training. A national pilot published in Annals of Surgery in 2023 documented the early implementation experience, and subsequent work has examined the validity of EPA ratings, the quality of the narrative comments attached to them, and the gaps between what faculty think they are communicating and what residents actually hear. Earlier studies have found that intraoperative EPA feedback often focuses narrowly on technical performance and provides limited guidance, and that the quality of narrative comments varies widely across evaluators. The new findings add a structural dimension to that critique: it is not only what faculty write that matters, but whether they write anything at all — and who ends up on the receiving end of the silence.

The authors’ conclusion is direct: efforts should be made to ensure that all residents receive actionable feedback along the continuum of their training. In practical terms, that could mean requiring a narrative comment for every EPA assessment regardless of entrustment level, building prompts into the mobile assessment platform that ask faculty to justify high ratings, or using faculty development workshops — an approach already shown to improve the utility of narrative evaluations in other specialties — to train supervisors in giving feedback that is specific, growth-oriented, and useful to trainees at every stage. The mobile platform through which the American Board of Surgery delivers EPA assessments could, in principle, make a comment field mandatory rather than optional.

For the surgical profession, the stakes extend beyond individual trainees. Entrustment decisions ultimately feed into certification and into the public promise that a newly independent surgeon can be trusted with patients’ lives. If the highest-rated residents are the least documented, programs lose a rich source of validity evidence — the written reasoning behind the judgment — at exactly the point where it is most valuable. The study’s dataset of 1,071 assessments across multiple institutions provides a solid foundation for that concern, and its statistical controls for training level and case complexity make the feedback gap difficult to dismiss as an artifact. The next step, the authors suggest, is ensuring that trust, once earned, is always explained.

Subject of Research: The relationship between entrustment levels and assessment factors in Entrustable Professional Activity evaluations of general surgery residents

Article Title: The relationship between entrustment and EPA assessment factors in general surgery residency training

Article References: Moore, G., Hegeholz, D., Sun, T., Weaver, M. L., Ward, E. P., Mabes, E. S., & Jaiswal, K. (2026). The relationship between entrustment and EPA assessment factors in general surgery residency training. Global Surgical Education – Journal of the Association for Surgical Education, 5(1), Article 157. https://doi.org/10.1007/s44186-026-00559-7

Image Credits: AI Generated

DOI: 10.1007/s44186-026-00559-7

Keywords: entrustable professional activities, general surgery residency, entrustment, narrative feedback, surgical education, competency-based medical education, operative assessment, post-graduate year, case complexity, faculty evaluation, resident training, assessment validity