The resident had logged everything. Fourteen months of clinical hours documented to the half-day. One hundred and sixty hours of direct mentoring. Three hundred hours of didactic coursework. Every evaluation completed on schedule, every competency checklist initialed, every supervision form signed and filed. The program director reviewed the portfolio before the final clinical performance examination and confirmed that every standard had been met.
The patient who walked in for that examination was a fifty-eight-year-old letter carrier with bilateral shoulder pain, a twelve-year history of type 2 diabetes, a recent medication change that had disrupted sleep for three weeks, and a Workers’ Compensation claim that had been denied twice. The resident performed a thorough bilateral shoulder examination, including range of motion, strength, and special tests, all documented with appropriate precision. The treatment plan was mechanically sound. It addressed the shoulders.
It did not address anything else.
The mentor observing the examination did not mark the resident as failing. The clinical reasoning was not wrong. The examination technique was proficient. But the treatment plan revealed something the hours could not: the resident had not yet developed the capacity to recognize when the presenting complaint is not the clinical problem. The hours said ready. The patient said otherwise.
This is the gap that competency-based education was designed to close.
For decades, the dominant model of clinical training in physical therapy has operated on a deceptively simple assumption: accumulate enough supervised hours and competence will follow. The logic feels intuitive. More time with patients, more exposure to clinical complexity, more supervision — these should produce better clinicians. The profession’s accreditation standards, residency requirements, and continuing education mandates are all built on this foundation. Time is the currency. Hours are the unit of measurement.
The evidence does not support the assumption. A systematic review examining the relationship between supervision quantity and clinician outcomes found no correlation (Simpson et al., 2021). More hours of direct supervision did not predict better patient outcomes. The finding is not an indictment of supervision itself — it is an indictment of the metric. Measuring how much time a clinician spends being supervised tells us nothing about what happened during that time, what was learned, or whether the learning transferred to independent practice.
This is not a new insight in health professions education. The shift from time-based to competency-based frameworks has been underway in medicine for over two decades. Frank and colleagues synthesized 173 published definitions and identified the core premise: competency-based education organizes curricula around outcomes rather than time, asking not how long a clinician has trained but what they can be trusted to do (Frank et al., 2010). The evidence base is now substantial enough that the question is no longer whether competency-based models work, but why so many training systems have not yet adopted them.
The answer, in physical therapy, is structural. The profession’s training infrastructure was not designed for competency-based assessment. Continuing education is delivered in disconnected units — a weekend course here, an online module there — with no mechanism for tracking whether the learning changed clinical practice. Residency programs that have adopted competency-based frameworks represent a small fraction of the profession’s training capacity. The pipeline remains thin: fewer than twelve percent of new graduates even apply for residency training. The barriers are not motivational — they are economic. Student debt already exceeds a hundred thousand dollars at the median. Geographic constraints. The absence of any reimbursement incentive that would justify the investment for either the clinician or the employer.
And the economic structure offers no incentive to build more of what works: a clinician who completes a competency-based residency bills the same codes and receives the same reimbursement as one who completed a time-based program, or no program at all. Medicare reimbursement for physical therapy, adjusted for inflation, is roughly half of what it was thirty years ago (MedPAC, 2023). The payment system does not recognize the output of the training system. It cannot distinguish between a clinician who has been assessed against demonstrated competencies and one who has simply logged the requisite hours. When the signal the market receives is identical regardless of training quality, the market underinvests in training quality.
What competency-based education requires is a fundamental inversion. Instead of tracking inputs — hours logged, courses completed, forms signed — it tracks outputs: what the clinician can actually do, at what level of complexity, with what degree of independence. The credential becomes a specific claim about demonstrated capacity rather than a record of time served. In orthopedic surgery, this inversion shortened training time for some residents while maintaining or improving outcomes over an eight-year evaluation period (Nousiainen et al., 2018). In family medicine, it generated 24,000 data points across 150 residents in three years — enough information to identify performance outliers early and intervene before those gaps reached patients (Schultz & Griffiths, 2016).
The data from proficiency-based progression training are even more striking: a 60% reduction in performance errors compared to conventional approaches (Mazzone et al., 2020). These are not marginal improvements. There are differences between a system that tracks whether learning occurred and one that assumes it did. And they emerge from a model structurally different from what most clinicians experience — not more of the same training delivered more intensively, but a fundamentally different relationship among the learner, the assessment, and the standard.
The hardest element of this model is not the assessment architecture. It is the willingness to say “not yet.”
Genuine competency-based education means some clinicians complete training in less time, and some require more. It means the system must be willing to hold a clinician at a given level — not as punishment, but as an honest acknowledgment that the evidence does not yet support advancement. This requires assessment tools that can make graduated trust decisions rather than binary pass-fail judgments. It requires mentors trained to calibrate those decisions. It requires a professional culture willing to accept that time served is not the same thing as readiness.
The profession has faced this implementation lag before. The shift from structure-based to outcome-based education in medicine took three decades from initiation to meaningful adoption — driven not by lack of evidence but by the absence of assessment tools adequate to the task (Carraccio et al., 2002). Physical therapy is tracing the same arc, decades behind. The evidence exists. The frameworks exist. What does not yet exist is the economic and institutional infrastructure to make competency-based training the norm rather than the exception. The profession remains caught between a training model that has been shown to work and a payment model that cannot see the difference.
The resident from the opening examination is still practicing. The clinical reasoning that was proficient but incomplete at fourteen months continued to develop — not because more hours were logged, but because a mentor identified the specific gap and the program had the assessment infrastructure to name it. The resident was not told they had failed. They were told: here is what you can do, here is what you cannot yet do, and here is what comes next.
That is not a system built on hours. That is a system built on trust — calibrated, evidence-informed, and honest enough to say what the hours alone could not.
