Are We Training Physicians or Just Staffing Hospitals? The Service vs. Education Problem Nobody Wants to Name
Every July 1st, a new academic year begins in graduate medical education.
New residents arrive. Schedules get distributed. Orientation happens. And then in most programs, something quieter and more consequential takes place: everyone figures it out as they go.
Rounds happen when they happen.
Discharges happen when they happen.
Attendings show up with varying degrees of clarity about what their role in the educational enterprise actually requires of them. Residents learn by absorption, by proximity, by trial and error in environments that were never explicitly designed for learning.
This not a failure of commitment. Most people in academic medicine care deeply about education. It is a failure of structure. And it is one of the most persistent and least discussed problems in graduate medical education.
What the Goals and Objectives Actually Say
Every accredited residency program has goals and objectives. ACGME requires them. They live in program documents, in New Innovations, in the materials handed to residents at the start of each rotation.
Most of them describe what residents are expected to learn. Some include a list of recommended readings or core references. A few gesture toward the competencies that will be assessed at the end of the rotation.
Almost none of them describe what the rotation will actually look like.
Not the anticipated schedule. Not how many clinic sessions versus inpatient sessions versus teaching sessions a resident should expect. Not who they will work with or what those attendings are specifically responsible for teaching. Not what a typical day looks like, what time rounds start, what the expectation is around discharges, or how the educational experience is structured within the reality of patient care demands.
The result is a document that describes outcomes without describing the environment designed to produce them. And an environment without explicit structure is not a learning environment. It is a service environment with educational language attached to it.
The Service vs. Education Imbalance
This is the tension that academic medicine has never fully resolved and rarely discusses honestly: residency programs exist simultaneously as educational programs and as workforce solutions. Residents are learners and they are also, in very practical terms, essential to the functioning of the clinical service.
That dual role is not inherently problematic. Learning happens in the context of real patient care, and that is as it should be. The problem arises when the service demands consistently crowd out the educational design. When the rotation exists primarily to staff the floor and the educational goals exist primarily to satisfy accreditation requirements.
When that happens, a few things follow predictably.
Accountability becomes diffuse. If no one has clearly defined what the educational experience is supposed to look like, no one can be held responsible when it falls short. The resident who isn’t learning efficient rounding practices has no framework to measure against. The attending who isn’t providing structured teaching has never been told explicitly that structured teaching is part of their role on this rotation. The program director who receives end-of rotation evaluations full of vague feedback has no baseline to interrogate them against.
Expectations become implicit rather than explicit. Rounds happen at 8am in some programs and 10am in others, not because of any educational rationale but because that is what has always happened. Discharges happen whenever they happen because no one ever built a structure around when they should happen. Residents figure out the unwritten rules by watching, by asking quietly, by making mistakes in environments where the rules were never written down.
And the free-for-all becomes normalized. After enough years, the absence of structure stops feeling like a gap and starts feeling like the culture. This is just how it works here. Everyone adapted. No one questions it.
The Ripple Effect Nobody Talks About
The impact of unstructured educational environments does not stay inside the educational relationship between attending and resident. It moved outward, into every corner of the hospital, and it touches everyone.
Nursing staff absorb the uncertainty first. When rounds happen at unpredictable times, when discharge orders arrive without warning or not at all, when the team composition changes without communication, it is the nurses who field the questions, manage the gaps, and hold the clinical environment together while the educational structure figures itself out. That is not their job. But in programs without explicit structure, it becomes their reality.
Patients feel it too. The patient waiting for a discharge that was supposed to happen at noon but hasn’t been written because the attending hasn’t rounded yet, or because the resident wasn’t sure whether that was their responsibility, or because no one ever built a structure around when discharges should happen. That patient is not a bystander to an educational problem; they are experiencing it directly.
And the institution absorbs it systemically. Length of stay metrics. Throughput data. Patient satisfaction scores. Nursing retention. These are not separate from the educational structure of residency programs. They are downstream consequences of it. A hospital where the educational enterprise runs without explicit structure is a hospital where the clinical enterprise runs with predictable inefficiency, not because anyone is negligent, but because the systems that should create accountability and clarity were never built.
This is the argument that most discussions of the service versus education balance never fully make: the cost of educational disorganization is not borne only by residents and programs. It is distributed across the entire institution, and some of the people bearing it most directly have the least ability to name what is causing it.
What Structured Educational Design Actually Looks Like
The programs that have genuinely resolved this tension share something in common: they made the implicit explicit. They took the time to design the educational experience as deliberately as they designed the clinical service, and they documented that design in ways that create accountability at every level.
At the rotation level this means goals and objectives that go beyond learning outcomes to describe the actual structure of the educational experience. What does a typical week look like? How many clinic sessions, inpatient days, teaching conferences, and procedural opportunities should a resident except? Who is the primary attending for each component and what is the attending’s specific educational responsibility? What time do rounds start and why? What is the expectation around discharge planning and how does that connect to the educational goals of the rotation?
There are not bureaucratic questions. They are the foundational questions that determine whether a rotation produces the learning it was designed to produce or simply produces clinical work with education as a byproduct.
At the attending level this means faculty who understand their role in explicit terms. How many sessions per week does this attending have dedicated to resident teaching? What does their productivity expectation look like and how does the program account for the time investment that genuine teaching requires? Has anyone sat with them and discussed what effective supervision and feedback look like in the context of this rotation?
Faculty development in most programs treats teaching skill as something attendings either have or don’t have. The more foundational question is whether attendings have been given a clear picture of what their educational responsibilities are and what the program expects them to deliver.
At the program level this means someone is looking at the schedule every year, not just distributing last year’s version with updated names. How has the clinical volume changed? How has the resident complement changed? Are the educational goals still aligned with what the rotation actually produces? When did we last ask residents whether the structure we designed is the structure they are experiencing?
The Follow-Through Problem
Here is the part of the conversation that most articles leave out.
Building explicit educational structure is not a one-time exercise. It is an ongoing commitment that requires someone to follow up and follow through, consistently and over time. And progress is rarely immediate. The program that redesigns its rotation goals and objectives in July will not see dramatically different evaluations scores by September. The faculty development initiative that launches in the fall will not transform teaching culture by spring.
This is where most structural improvement efforts stall. Leadership invests in the design work, the launch, the announcement. And then, when results don’t appear quickly, momentum fades. The new structure gets absorbed into the old culture. The explicit expectations become implicit again. The fee-for-all reasserts itself because no one stayed with the change long enough to let it take hold.
Sustainable improvement in educational structures requires someone with institutional authority and institutional patience to hold the line. To ask, in the annual program evaluation, whether the rotation looks the way it was designed to look. To follow up with faculty who were given explicit expectations and check whether those expectations are being met. To name, differently and without softening, when the structure that was built on paper is not the structure that residents are experiencing.
This is not the comfortable work. It requires a willingness to have difficult conversations with faculty who have practiced a certain way for years. It requires program directors to hold attendings accountable for educational responsibilities that may never have been clearly defined before. It requires institutional leadership to support that accountability rather than treating it is an administrative inconvenience.
But it is the work. And without it, every structural improvement eventually reverts to the mean.
A Practical Starting Point
If you are a program director or GME leader reading this and recognizing your program in what I have described, the path forward does not require a complete overhaul of everything at once.
Start with one rotation. Pick the one where the educational experience is most unclear, most variable, or most frequently the subject of vague or frustrated feedback. Sit with the residents who just completed it and ask them what the rotation actually looked like, not what the goals and objectives said it would look like.
Then build the structure that closes the gap between those two things. Document it specifically. Share it with the attending faculty. Make the expectations explicit on both sides. And then, three months later, go back and ask whether the rotation looks like what you desgined.
That cycle, repeated deliberately across rotations and over time, is how educational structure actually improves. Not through a single redesign effort but through the accumulated discipline of asking whether what we said we would do is what we are actually doing.
Progress will be slow. There will be rotations that improve and then slide back when a key faculty member leaves or a clinical volume spike changes the rhythm of the service. That is not failure. That is the nature of building something in a living, changing institution. What matters is that someone is paying attention, following through, and wiling to name when the structure needs to be rebuilt.
Closing Reflection
Graduate medical education exists to train physicians. Not to staff hospitals, not to move patient through the system, not to generate clinical revenue, though all of those things happen inside it. Its primary purpose is educational.
When the structure of a residency program cannot answer basic questions about what a resident’s week looks like, who is responsible for their learning on any given rotation, or what the explicit expectations are for the faculty supervising them, that purpose has been subordinated to other priorities. And the consequences of that subordination do not stay inside the educational relationship. They move through the entire institution, touching nursing staff, patients, and operational metrics in ways that are rarely connected back to their source.
The new academic year starts July 1st. Every program in the country is preparing for new reisdents to arrive.
The question worth asking before they do is not whether the orientation is planned or the schedules are distributed. It is whether the educational structure those residents are walking into was designed, deliberately, explicitly, and with someone committed to following through, to actually produce the learning it promises.
Lay the groundwork. Rise to the occasion.
The residents arriving this July deserve an environment that was built for them. So do the nurses working alongside them, the patients in their care, and the institution responsible for all of it.