Medical Training’s Mistreatment Problem Demands a System-Wide Reboot

Mistreatment in medical education has haunted the profession for decades, and a new editorial in the Journal of General Internal Medicine argues that the field’s response has been fundamentally mismatched to the scale of the problem. Writing in response to a scoping review by Gonzalez and colleagues, Kathryn E. Miller and Lia S. Logio of Case Western Reserve University School of Medicine contend that medicine does not lack awareness of trainee mistreatment; it lacks alignment between how widespread the problem is and how modest the interventions against it have been. Their argument, published as the profession confronts record levels of clinician burnout, is essentially a call to redesign the clinical learning environment from the ground up rather than continue patching it with one-off training sessions.

The Gonzalez review, which the editorial frames as a timely analysis of a wicked problem, examined the published literature on structures and interventions aimed at mitigating mistreatment within graduate medical education, the training phase that begins after medical school graduation. The authors identified 71 articles and 30 abstracts published between January 2015 and July 2025, describing 71 distinct structures. Of these, 56 were educational or training interventions and 15 were programs or policies. The policy-level approaches included instruments to measure mistreatment and institutional or programmatic strategies to address or prevent it, some of which targeted efforts at specific groups such as minoritized trainees. On the surface, that volume of activity might suggest a field taking the problem seriously. The details tell a different story.

The most striking finding is how thin most of the effort has been. The majority of the training interventions cataloged in the review were single sessions, 44 of them, and most targeted residents as participants, 34 in total. In other words, the dominant model for combating a deeply entrenched cultural problem has been a brief lecture or workshop delivered to one layer of the hierarchy. Miller and Logio argue that this approach exposes a critical structural limitation: mistreatment is treated as a learner issue, something to be managed through trainee education, when it is in fact a workforce, quality, safety, retention, and equity issue that implicates everyone who shapes the clinical environment.

The data on prevalence help explain why the editorial’s authors are impatient. Mistreatment in undergraduate medical education has been tracked annually since 1999 through the American Association of Medical Colleges Graduation Questionnaire, and the most recent data show that 38 to 40 percent of medical students report experiencing at least one incident of mistreatment during medical school. Mistreatment is defined by the AAMC as any behavior that shows disrespect for the dignity of others and unreasonably interferes with the learning process. Nearly four in ten students encountering such behavior suggests that disrespect is not an aberration in clinical training but a routine feature of it.

The picture in graduate medical education looks different, and the difference matters for how interventions should be designed. Since 2020, the Accreditation Council for Graduate Medical Education has included professionalism items on its annual resident survey, asking whether residents personally experienced or witnessed abuse, harassment, mistreatment, discrimination, or coercion. National rates of personally experienced mistreatment have ranged from 5 to 7 percent, and witnessed mistreatment from 6 to 8 percent. Those percentages translate into an average of nearly 9,500 trainees who personally experience mistreatment and more than 10,700 who witness it each year. The editorial argues that these figures reveal the persistence of an unsafe culture in which bad behavior is written off as a professional rite of passage.

Why do medical students report mistreatment at rates several times higher than residents? The editorial points to structural differences between the two populations. Medical students are novice learners who receive grades during short-term clinical rotations, making them uniquely vulnerable to the power dynamics of the wards. Residents, by contrast, are embedded in a specialty as employed junior colleagues who often work with the same faculty over years, and they are joining a discipline-specific guild in which success depends heavily on a strong network of mentors. Graduate medical education programs are also housed within hospitals where institutional factors may shape the culture of reporting. All of this makes the true scope of mistreatment difficult to measure, let alone eliminate, and it means that findings from one level of training cannot simply be transferred to the other.

Perhaps the most consequential evidence cited in the editorial comes from a study by Leitman and colleagues at the Icahn School of Medicine, which used an online reporting system to track unprofessional behavior. Of 173 reported incidents, 104 were attributed to faculty. Just twenty faculty members, 0.7 percent of the faculty, accounted for 52 of the reports, a full half of the total, while 97 percent of faculty members had no reports at all. Miller and Logio endorse the study’s conclusion that longitudinal skills programs focused on the small fraction of repeat offenders, roughly 3 percent, are more likely to move the needle than broad-based training aimed at everyone. The concentration of harm in so few hands suggests that targeting resources at the source, rather than diffusing them across entire trainee populations, could be the single highest-yield intervention available.

The editorial also insists that any reboot of the clinical learning environment must center the marginalized populations who bear a disproportionate burden of mistreatment, bias, and harassment. A 2026 AAMC report on sexual harassment among medical school faculty found overall rates of 11.3 percent among men and 29.3 percent among women. Among LGB+ men the rate was 17.2 percent, and among LGB+ women it reached 40.9 percent. Critically, metrics around engagement were significantly lower among those who had been harassed compared with their peers. That erosion of engagement, the authors argue, undermines both learning and patient care, and it demonstrates that rigorous standards of competence and a genuine sense of belonging are not in tension. Psychological safety, they suggest, is the foundation on which both rest.

So what would a systems approach look like? The editorial draws on the literature on healthy clinical learning environments, citing Stuart Slavin, the ACGME Vice President of Well-Being, who advocates for organizational design, strong leadership, and balancing workplace demands with available resources. Eliminating mistreatment, on this view, requires every member of the clinical learning environment to focus on workplace well-being, backed by leaders who buy into necessary changes and reinforce the message clearly and frequently. The authors also point to relationship-centered care, an approach that fosters a strong web of interpersonal connections among care team members in addition to strong clinician-patient relationships. They raise an uncomfortable possibility: the large health systems created through hospital mergers, acquisitions, and private equity buyouts may have eroded something foundational to the learning environment itself.

The destination the editorial describes comes from organizational psychologist Amy Edmondson, whose work identifies high psychological safety combined with high accountability as the hallmark of optimal conditions for learning and performance. Institutions, the authors write, need robust reporting systems that practice transparency and accountability, with particular attention to repeat offenders, and clinical learning environments should embrace frameworks that allow strong interpersonal relationships to form and grow. Future research should follow the gaps the review exposes: longitudinal, multi-level programs that perhaps track trainees from medical school into residency, interventions that measure whether behavior actually changes rather than merely whether awareness increases, and comparative studies across undergraduate and graduate settings where reporting cultures and power dynamics differ substantially. The editorial’s closing argument is blunt. A clinical learning environment shaped by fear, silence, and preventable disrespect cannot produce the belonging, trust, and accountability that learning and patient care require, and rebooting it will require leaders to redesign systems, relationships, and norms so that mistreatment is prevented at its source and no longer mistaken for part of becoming a physician.

Subject of Research: Mistreatment and interventions in the medical education clinical learning environment

Article Title: Reboot the Clinical Learning Environment

Article References: Miller, K. E., & Logio, L. S. (2026). Reboot the Clinical Learning Environment. Journal of General Internal Medicine. https://doi.org/10.1007/s11606-026-10884-8

Image Credits: AI Generated

DOI: 10.1007/s11606-026-10884-8

Keywords: medical education, graduate medical education, mistreatment, clinical learning environment, resident wellness, psychological safety, sexual harassment, ACGME, AAMC, scoping review, physician burnout, unprofessional behavior