Mentorship has long been recognized as one of the most powerful predictors of a physician’s professional trajectory, shaping everything from career choice and research productivity to job satisfaction and resilience in the face of burnout. Yet for residents from backgrounds underrepresented in medicine, the path to meaningful mentorship is frequently obstructed by structural barriers, scarcity of shared identity role models, and the diffuse, unstructured nature of the mentoring arrangements most training programs rely on. A new study published in the Journal of General Internal Medicine describes how Brigham and Women’s Hospital (BWH), a major teaching affiliate of Harvard Medical School in Boston, tackled this problem head-on with a deliberately engineered, multidimensional mentorship program for underrepresented in medicine (URiM) residents, and the early evaluation results suggest the approach is delivering strikingly better outcomes than the conventional model.
The program, developed within the BWH Internal Medicine Residency, was designed and implemented by a team of physician-educators led by Dr. Zainab Jaji, with Drs. Nora Y. Osman, Barbara Gottlieb, and Valerie E. Stone as co-authors, all affiliated with Harvard Medical School and, in several cases, the Division of General Internal Medicine at Brigham and Women’s Hospital. Rather than treating mentorship as an informal, ad hoc relationship that residents are expected to cultivate on their own, the program treats it as a designed system with distinct, complementary components. The result is a layered architecture of support that moves residents from simple recruitment into the profession toward what the authors describe as a genuine sense of belonging.
At the foundation of the program lies the personalized dyadic pairing, in which each URiM resident is matched one-on-one with a faculty mentor. This is not a random assignment but a deliberate match informed by the resident’s needs, career interests, and experiences, as well as the mentor’s capacity to support them, including across differences in background. The dyad is the anchor of the program, providing continuity and a stable, trusted relationship that can span the entire arc of residency. Around that anchor, the program adds mentorship communities, small groups that create peer-to-peer and resident-faculty networks so that no individual mentor-mentee relationship bears the full weight of a resident’s developmental needs. This reflects a growing consensus in the medical education literature that mentorship is best conceived as a network rather than a single relationship, since no one mentor can supply guidance on research, clinical development, work-life integration, and identity all at once.
The third structural element, speed mentoring, borrows the rapid-rotation format familiar from professional networking events and applies it to academic medicine. In concentrated sessions, residents meet sequentially with multiple faculty members, allowing them to sample perspectives, gather targeted advice on specific questions, and identify potential additional mentors with whom they feel natural rapport. The format lowers the threshold for initiating new professional relationships, a hurdle that can be especially high for trainees who may feel isolated or uncertain about whether senior faculty will be receptive to them. Finally, the program invests in mentorship trainings, with particular emphasis on mentoring across differences. These training sessions prepare faculty to navigate conversations about race, identity, discrimination, and the additional, often invisible burdens that URiM trainees can carry, including what researchers have called the minority tax, the unpaid and unrecruited labor of diversity work that falls disproportionately on underrepresented faculty and trainees alike.
The conceptual scaffolding for the program draws on well-established frameworks. The authors invoke Urie Bronfenbrenner’s ecological model of human development, which situates individual growth within nested layers of environment, from immediate relationships to institutional culture, and which argues that supportive development requires intentional design across all of those layers. They also draw on the concept of community cultural wealth, articulated by education scholar Tara Yosso, which reframes students and trainees from underrepresented backgrounds not as lacking assets but as bringing distinctive forms of capital, including navigational, social, and resistant forms of knowledge, that programs should recognize and build upon. In practical terms, this means the program does not merely aim to remediate deficits; it aims to connect URiM residents to networks, sponsors, and institutional knowledge that historically have been less accessible to them.
The urgency of this kind of intervention is well documented. Systematic reviews have shown that mentoring in academic medicine is inconsistently provided, often left to chance, and less accessible to URiM trainees and faculty. Studies of minority resident physicians have documented experiences of discrimination and isolation during training, and national surveys across surgical and internal medicine residencies have revealed disparities in mentorship access that correlate with worse wellness and career preparation. Prior work has also shown that residents who receive strong mentorship are more likely to pursue academic careers and report higher satisfaction with their training, making mentorship a plausible lever for improving both individual trajectories and the retention of a diverse physician workforce. Despite this evidence, few programs have published detailed descriptions of structured, multidimensional mentoring interventions specifically built for URiM residents, which is precisely the gap the BWH team set out to fill.
To evaluate whether the program was actually working, the investigators administered a residency-wide mentorship survey in 2024 and compared the experiences of residents enrolled in the URiM Mentorship Program with those of residents in the Usual Mentorship Program offered to everyone else in the residency. The comparison yielded numbers that are difficult to dismiss. Among the 49 survey respondents, 31 were in the Usual Program and 18 in the URiM Program. In the URiM cohort, 94 percent of residents reported viewing their assigned mentor as a primary mentor or as part of their mentoring network. In the Usual Program, that figure was just 39 percent. In other words, URiM residents in the structured program were more than twice as likely to have converted an administrative assignment into a real, functioning mentoring relationship.
Satisfaction with faculty mentor assignments told a similar story. In the URiM Mentorship Program, 83 percent of residents expressed satisfaction with the faculty mentor they were assigned. In the Usual Program, satisfaction stood at 48 percent. The near forty-percentage-point gap in mentor integration and the thirty-five-point gap in assignment satisfaction together suggest that the difference lies not in the willingness of individual faculty to mentor but in the structure surrounding the relationship. Unstructured mentoring, the data imply, leaves too much to chance: residents must identify potential mentors, initiate contact, negotiate expectations, and sustain momentum largely on their own, a sequence that disproportionately disadvantages those who lack pre-existing social capital within academic medicine. Structured programs, by contrast, front-load the matchmaking, provide scaffolding through communities and events, and train mentors so that the relationship starts on firmer ground.
The authors are careful to frame the work as a program description and evaluation rather than a definitive trial, and the survey sample, 49 respondents at a single institution, limits generalizability. The program itself was also deemed an educational initiative rather than human subjects research, and no external funding supported the work. Still, the internal comparison is unusually clean: both groups existed within the same institution, the same residency, and the same broader culture, so the large differences in mentor integration and satisfaction point directly to the program’s design choices as the operative variable. The findings were previously shared as a workshop at the Society of General Internal Medicine Annual Meeting in Boston in May 2024, and their formal publication now offers a replicable template for other residency programs grappling with similar disparities.
What makes the BWH model notable is its multidimensionality. Each component addresses a distinct failure mode of traditional mentoring. Dyadic pairing solves the problem of access and initiation. Mentorship communities solve the problem of over-reliance on a single mentor. Speed mentoring solves the problem of exposure to diverse perspectives and low-cost relationship building. Cross-difference training solves the problem of mentors feeling unprepared for conversations about identity and discrimination, which can otherwise cause well-intentioned relationships to stall. Together, the components form a system in which a resident’s mentoring needs are met through redundancy and layering rather than through a single fragile connection, echoing the mentorship-as-network model that has gained traction across academic medicine.
The implications extend beyond the residents directly served. A physician workforce that reflects the diversity of the patient population it serves is associated with better communication, better patient outcomes, and reduced health care disparities, and residency is a decisive bottleneck through which diverse talent either flourishes or leaks out of the pipeline. If structured mentorship can more than double the rate at which residents form genuine mentoring relationships, programs that adopt similar architectures may see compounding returns: stronger mentorship fuels career advancement, advancement produces more senior URiM role models, and those role models in turn enrich the mentoring ecosystem for the next generation of trainees. The BWH authors describe their program’s trajectory as moving from recruitment to belonging, and the survey data suggest that belonging, once engineered deliberately rather than left to chance, is measurable, achievable, and transformative for the residents who experience it.
Subject of Research: A structured, multidimensional mentorship program for underrepresented in medicine (URiM) residents in an internal medicine residency, compared with usual unstructured mentorship
Subject of Research: Medicine
Article Title: From Recruitment to Belonging: A Multidimensional Mentoring Program for Underrepresented in Medicine (URiM) Residents
Article References: Jaji, Z., Osman, N. Y., Gottlieb, B., & Stone, V. E. (2026). From Recruitment to Belonging: A Multidimensional Mentoring Program for Underrepresented in Medicine (URiM) Residents. Journal of General Internal Medicine. https://doi.org/10.1007/s11606-026-10713-y
Image Credits: AI Generated
DOI: 10.1007/s11606-026-10713-y
Keywords: mentorship, URiM, residency, professional development, medical education, underrepresented in medicine, Brigham and Women’s Hospital, dyadic mentoring, mentoring across differences, internal medicine, mentorship networks, diversity in medicine
Cite Scienmag News
APA MLA Chicago
Ophelia Keating. (September 11, 2026). Building Belonging: A Multidimensional Mentoring Program for URiM Residents. Scienmag. https://scienmag.com/building-belonging-a-multidimensional-mentoring-program-for-urim-residents/
Ophelia Keating. “Building Belonging: A Multidimensional Mentoring Program for URiM Residents.” Scienmag, 11 September 2026, https://scienmag.com/building-belonging-a-multidimensional-mentoring-program-for-urim-residents/. Accessed 11 September 2026.
Ophelia Keating. “Building Belonging: A Multidimensional Mentoring Program for URiM Residents.” Scienmag. September 11, 2026. https://scienmag.com/building-belonging-a-multidimensional-mentoring-program-for-urim-residents/
Copy citation Download RIS
Tags: addressing burnout in medical residentsaddressing diversity in medical educationburnout prevention through mentorshipearly evaluation of mentorship interventionsevaluation of mentorship program outcomesfostering belonging among underrepresented medical traineesHarvard Medical School mentorship initiativesHarvard-affiliated residency programshospital-based mentorship programsimpact of mentorship on resident career developmentimproving diversity and inclusion in medical traininginnovative mentorship models in internal medicinementorship impact on physician career developmentmultidimensional mentoring programs in medical trainingmultidimensional mentoring programs in medicinephysician educator-led mentorship initiativesrole models for URiM residentsshared identity role models in healthcarestructural barriers in medical mentorshipstructural barriers in medical trainingstructured mentorship models in internal medicineunderrepresented in medicine physician supportURiM resident mentorshipURiM residents mentorship
