The morning schedule showed two names side by side in the eight o’clock slots. One belonged to a clinician who had been out of school for seven months, still learning to trust the examination findings that matched the textbook. The other belonged to a clinician who had completed a three-year fellowship, published on clinical reasoning, and spent a decade learning to see what the textbook never described. They shared a hallway. They shared a treatment room divider. They shared a scheduling template that assigned patients to the next available provider.
At 8:15, both were treating a shoulder. The new graduate was working through a supraspinatus protocol — painful arc, positive Hawkins-Kennedy, resisted external rotation reproducing concordant pain. The examination was competent. The treatment selection was defensible. Across the divider, the fellowship-trained clinician was asking a different set of questions entirely. The patient’s shoulder was the referral diagnosis, but the presentation included eighteen months of progressive sleep disruption, a medication change six weeks prior, and a functional decline that did not map to any single structure. The fellowship-trained clinician was not treating a shoulder. The fellowship-trained clinician was managing a person whose shoulder happened to be the thing that finally brought them through the door.
Both clinicians billed the same evaluation code. Both received the same reimbursement. If you read the schedule that afternoon, you could not tell the difference between the two sessions. The system could not tell the difference either. It was not designed to.
This is the profession’s central structural fact, and it is worth stating plainly: physical therapy is functionally flat. Credentials exist — board certifications, residency completions, and fellowship designations. Training pathways exist that develop clinicians well beyond entry level. But these credentials do not determine who treats whom. They do not trigger referral pathways. They do not create triage mechanisms based on case demands. A new graduate and a fellowship-trained clinician see the same patients, bill the same codes, and receive the same reimbursement. The credential exists, but it determines nothing about how care is organized.
What is missing is the entire architecture. Medicine built a developmental continuum decades ago. A medical student becomes a resident. A resident becomes a fellow. A fellow becomes an attending. Each transition carries scope — referral pathways route patients toward clinicians whose training matches the case demands. Complexity drives escalation. Training level determines scope. A primary care physician who recognizes a cardiac presentation beyond their scope of practice refers to a cardiologist. An electrophysiology problem goes to an electrophysiologist. The structure is imperfect. It is also real. It exists.
Physical therapy has no equivalent. The profession produces clinicians through entry-level doctoral programs, hands them a license, and offers no systematic mechanism for continued development (ABPTRFE, 2023). Fewer than five percent of new graduates complete a residency. That leaves ninety-five percent to develop clinical expertise through unstructured practice experience — through volume, through repetition, through the slow accumulation of pattern recognition that may or may not reach the level patients actually need. The profession’s flat career structure offers no built-in pathway from entry-level generalist to specialist clinician. What exists is a series of voluntary, uncoordinated options — continuing education courses, certifications, specialty board examinations — without a deliberate developmental architecture connecting them to how patients are assigned, how clinicians are deployed, or how care is delivered.
And yet, the infrastructure is not unimaginable. At least one fellowship program has built what the profession has not: 33 competencies, 10 entrustable professional activities, 111 mapping relationships, a complexity framework, an entrustment scale, and behavioral descriptors for calibrating the people who make the judgments. It serves a fraction of the profession. The profession as a whole has no equivalent development infrastructure at any level — not for new graduates, not for mid-career clinicians, not for clinical leaders. The infrastructure can be built. It has been. It simply has not been built at scale.
This is not an argument against the clinicians inside the system. The new graduate in that eight o’clock slot was doing exactly what the system asked — treating competently, billing accurately, and moving to the next patient. The problem is that the system asked the same question of the fellowship-trained clinician on the other side of the divider. The system made no distinction because it has no mechanism for distinction. It is not that clinicians at different developmental levels exist, and the system ignores them. It is that the system was never designed to see them.
The consequence plays out in every clinic, every morning, in every state. Patients whose complexity exceeds the clinician assigned to them cannot be matched with someone whose framework fits. They get treatment that addresses the findings in front of the clinician, and when they do not improve, the system offers two explanations: the clinician was not good enough, or the patient was not amenable to conservative care. Neither explanation captures the mismatch. The system produced the outcome. The system does not recognize that it produced the outcome.
We talk about this as though it is inevitable — as though a profession without a development continuum is simply how things are. But Medicare reimbursement for physical therapy, adjusted for inflation, is roughly half what it was thirty years ago, and a new graduate bills the same codes at the same rates as a fellowship-trained clinician (MedPAC, 2023). The reimbursement system pays the same rate regardless of training level. The regulatory structure does not differentiate scope by credential. The professional infrastructure that could create escalation pathways has not done so. The building blocks exist. They have not been assembled.
The two clinicians from that morning are still sharing the hallway. The schedule still assigns patients to whoever has the next open slot. The new graduate is still treating competently. The fellowship-trained clinician is still managing complexity that the schedule does not recognize. And somewhere between the two of them — in the gap the system does not see — a patient is receiving care that is defensible, evidence-based, and not enough.
Stay up to date on The Development Gap and subscribe to us on Susbtack!
