I believe in holistic, competency-based education. I have believed in it for most of my career. I have also provided many assessments that were neither holistic nor competency-based.
Here is one. I was covering for a colleague over a holiday weekend, on a service that was busier than expected. A student and I shared a single patient, a teenager admitted the night before. He presented her to me at the bedside, we talked through the plan with her and her mother, and I moved on to my next patient. He caught up with me in the hallway, asked for feedback, and handed me a short evaluation form to fill out and return to the program.
So I gave him feedback. His presentation had been well organized, and he had spoken to that fifteen-year-old like she was a person with an opinion about her own care, which is not a small thing. He thanked me, and we both went back to a busy day. On my way out I scribbled something on the form and dropped it in the box outside the clerkship director's office.
Everything in that account is accurate, and it may well have been useful to him in the moment. I have no idea whether any of it changed his performance in the long run. I did not know what he had already improved, or what he had been told before. I did not know whether communicating with an adolescent was a strength someone had specifically identified for him already, or whether what he actually needed that day was help building a stronger management plan. I saw him perform once, chose what to focus on, and moved on.
Think about how we train an elite pianist. A student spends years with a teacher who knows exactly what she has been working on, what she was doing six months ago, and what typically comes next. The performances change. The teacher persists. Then a jury hears her at intervals and judges her against standards everyone agrees on.
Clinical education runs the other way. The performances persist, because the patients keep coming, and the teacher changes every few weeks. We inherited the most demanding part of apprenticeship without the system that makes it work.
For most of the history of medical education, the arrangement was essentially: learn your 'theory,' work alongside a lot of different physicians, and we will examine you at the end. The curriculum was knowledge. The clinical years were exposure, and what you became out of that exposure was largely a matter of who you happened to work with.
Competencies changed that. We now have agreement on what a graduating physician must be able to do, and with the December 2024 release of shared Foundational Competencies for UME — developed jointly by the AAMC, AACOM, and ACGME, that agreement is becoming genuinely shared across schools for the first time.1 Every student is headed for the same destination. The route each one takes is necessarily different, because clinical education is built out of whichever patients walk through the door, but what counts as a passing performance is no longer a matter of local judgment.
Competencies and performance indicators, or milestones, are what make holistic, individualized assessment possible rather than aspirational. They are what let us say that two students who saw entirely different patients, in different settings, with different teachers, are both becoming equally skilled clinicians — and to show how we know. This is a system designed to ensure that each learner gets to the same place, by whatever route their clinical experience actually gave them.
Competencies have arrived in undergraduate medical education as a thoughtful piece of educational design, yet risk landing as yet another documentation burden. The debut of the residency milestones themselves have a mixed history in this regard.2,3 We are right to question how this will be different, particularly for those already assessing residents daily. The honest answer is that nothing about extending this model into UME guarantees it will work, and assuming that what works in residency will work in medical school is its own mistake. Success depends on two things that look different in UME than they do in GME.
Competency-based teaching requires an educator to observe a performance, interpret it against a shared framework, judge where this particular learner sits in their development, decide what they need next, and record that in a way the next teacher can use. Most of our clinical faculty have been trained to do this with residents, and the competency framework has been part of their working life for two decades.
But residency is a conservatory. The faculty group is small, the specialty is defined, and the same teachers see the same trainee over three years. The picture builds itself. What we are asking of faculty working with medical students is a different job: assess a learner you will see for a few days or even hours, at a lower and far more variable level of competence, and produce something meaningful out of a snapshot. Our backwards apprenticeship now comes with a competency framework attached.
Like a piano teacher, a residency program director already has an operating system. In fact, she is the operating system. She holds the whole history of one resident in her head, or at least at her fingertips, because she is there for all of it. Medical school has no such person. Almost no teacher sees a learner for more than one rotation, and sometimes for no more than a few hours or days, so no one is present for enough of it to hold the history. We have to build the thing that does the remembering.
Think about what a functioning one would have to do:
Record a learner's performance against the competencies in the moment, in a minute rather than an evening.
Integrate that observation into an emerging picture of the individual learner.
Make that picture usable right then, by both the learner and the teacher.
And carry it forward, to the next teacher, and eventually to the jury.
We have the first job, done poorly. A form in a hallway, dropped in a mail slot, a box clicked in the LMS. We do not have the other three at all.
This is why the opportunism of clinical education has felt like a problem to be managed. What gets assessed today depends on which patients came through the door, and we have spent years trying to discipline that with checklists, which is a losing fight, because our clinical day does not cooperate. But opportunism is only chaos if you cannot see what a learner needs. If you can, it becomes an advantage. The teacher is not working down a list that ignores the patients in front of her.
She is choosing, from what the day actually offered, the thing this learner needs today and can leverage tomorrow.
Which is what a good teacher does with whatever the student brings to the lesson.
Until UME can produce competency-based evidence that residency trusts, and the LCME supports the shift, the grading floor will hold. And as long as the grading floor holds, our learners will keep optimizing for it.
Residency program directors recruit against a finished product. A program director in Ohio has to trust an MSPE from anywhere, and without UME competencies in place, standardized testing remains the cheapest guarantee of a floor. So the real apprenticeship gets left to residency, where the conditions favor it: the specialty is defined, the trainees are licensed to care for patients, and something real is certified at the end. The message our learners receive is unmistakable. Show me you can clear the bar, and I will teach you the craft in three years. They hear it, and they respond accordingly. Until this changes, they will read competency as binary — have it or do not, an A or an F — rather than as a developmental path through school that belongs to them.
And the LCME has doubled down on uniformity as the goal, which keeps every school pointed at the same floor. That is a bigger argument for another day.
Both the training and the operating system are now within reach in a way they were not five years ago. We can build the thing that does the remembering. We can deliver consistent, individualized, and scalable training and provide faculty support to teachers who need it in the middle of a clinical week rather than at a retreat in June.
The rush to the common floor eases, because what we can now give program directors is what they have wanted all along: assessment grounded in real clinical performance, aligned to competencies they recognize. Not a ranking. Evidence. If program directors trust it, learners will trust it, and the binary reading starts to dissolve — not because we explained competency better, but because the system stopped rewarding the floor.
And the learner gets what we have never given him: a view of his own development, in time to do something about it. He arrives at a new rotation able to say where he is and what he needs. He sits in his own portfolio review as a contributor rather than a subject.
I think about that student and the form often. Not because I got it wrong — I gave him an honest account of the one thing I was in a position to see. But nobody ever heard him perform twice. Each of us met him briefly and lost him by the end of the month, and each of us told him something true, with no context for what we saw and nowhere useful to put it.
He deserved a teacher who knew what he had been working on, and where he was going. We could not give him one. We can now give him the next best thing, which is every teacher he meets knowing what he needs now, and next.
What those teachers say to him at that moment is the whole of it, and it is where this series goes next.
At Aquifer, we are approaching this challenge by connecting learning across disciplines, so that a change in rotation does not mean starting over in understanding a learner’s development. Built by our Consortium of health professions educators, our comprehensive clinical learning resources give faculty a consistent pedagogical approach they can integrate across the curriculum, with repeated opportunities for learners to practice clinical reasoning and assess their understanding in different clinical contexts. Bringing information from that work together helps learners and faculty recognize strengths, identify needs, and make more informed decisions about what comes next. Our aim is to help each learning experience contribute to a picture of the learner that grows over time and informs the next teaching conversation.
1 Association of American Medical Colleges, American Association of Colleges of Osteopathic Medicine, Accreditation Council for Graduate Medical Education. Foundational Competencies for Undergraduate Medical Education. AAMC. Accessed October 15, 2026. engage.aamc.org/UME-Competencies-AAMC-ACGME-AACOM
2 Malik MU, Diaz Voss Varela DA, Stewart CM, Laeeq K, Yenokyan G, Francis HW, Bhatti NI. Barriers to implementing the ACGME Outcome Project: a systematic review of program director surveys. J Grad Med Educ. 2012;4(4):425-433. doi:10.4300/JGME-D-11-00222.1
3 Hariton E, Bortoletto P, Barnes KL, Kaimal AJ, Stagg AR. Resident and Program Director's Perceptions of Milestone-Based Feedback in Obstetrics and Gynecology. J Med Educ Curric Dev. 2018 May 20;5:2382120518774794. doi: 10.1177/2382120518774794. PMID: 29845118; PMCID: PMC5964859.