PROMIS adaptive testing fits routine physical therapy without adding much friction
A mixed-methods study in Dutch primary care physical therapy found that PROMIS computerized adaptive tests, or CATs, can be woven into routine musculoskeletal care with relatively little disruption. Therapists generally viewed the tool as easy to use, and patients said the questionnaires were short enough to complete and relevant enough to their symptoms and function. The practical appeal is straightforward: when outcome measurement becomes shorter and more targeted, it is easier to sustain in the flow of an ordinary consultation rather than treating it as a separate administrative task.
The study focused on patients with musculoskeletal disorders of the spine and upper extremity across 14 physical therapy practices. PROMIS CATs were integrated into the electronic health record system and used during routine care, which matters because feasibility in a controlled setting often looks very different from feasibility in a busy clinic. Here, the central question was not whether the instrument could produce scores, but whether therapists and patients would actually accept it as part of care. Their responses suggest that the digital format did not feel abstract or detached from the encounter; it was experienced as a usable clinical tool.
One therapist-facing barrier stood out: uncertainty about how to interpret CAT scores.
Therapists wanted clearer score interpretation and fewer parallel questionnaires
Twenty-five physical therapists completed the survey, and 10 participated in interviews. Most described the PROMIS CATs as easy to use and helpful for steering conversations toward symptoms and functioning rather than relying only on memory or vague impressions. In practice, that is a meaningful shift. A patient who answers a brief adaptive questionnaire before a visit gives the clinician a structured snapshot that can sharpen assessment, especially when pain, disability, and recovery are changing from week to week.
Yet ease of administration did not erase all friction. Therapists wanted more confidence in what the scores meant and how to translate them into clinical decisions. In a setting where physical therapy decisions often depend on gradual changes in function, uncertainty around score interpretation can blunt the value of even a well-designed measure. A questionnaire that is quick to administer still needs a clear clinical language attached to it, or else it risks becoming another data point with no obvious bedside consequence.
Patients also described a different but equally practical burden: they were asked to complete legacy patient-reported outcome measures at the same time as the newer PROMIS CATs. Dual measurement increased respondent burden and diluted the advantage of the adaptive format. The study’s findings point to a simple operational principle: if a clinic wants shorter, smarter measurement, it should stop asking patients to carry older instruments in parallel unless there is a specific reason to do so.
Patients accepted the technology when it fed back into the visit
Nine patients were interviewed, and their comments made the implementation story more concrete. They described the CATs as quick to complete and relevant to their condition, and they were willing to keep using patient-reported outcomes when the results were actually discussed in consultation. That last point matters because patient-reported outcomes can feel perfunctory when they disappear into a chart. When a score informs the encounter, it becomes part of the therapeutic relationship rather than a bureaucratic requirement.
Some patients did find individual items less relevant to their own experience, which is a reminder that adaptive testing does not eliminate the problem of item fit entirely. It can reduce burden by tailoring the questionnaire, but the instrument still needs to ask enough of the right questions to feel clinically credible. A desire to review individual item responses also emerged, suggesting that transparency may improve trust and help patients understand how the score was assembled. For a therapy population that often wants to track recovery in plain terms, item-level visibility may make the tool feel less opaque.
The study also identified barriers tied to the online portal. If the digital pathway is clumsy, even a well-validated instrument will lose its advantage. In outpatient rehabilitation, where time is limited and the visit is already packed with assessment, exercise instruction, and progress review, any extra technological step can become the difference between smooth adoption and abandonment.
Implementation will depend on workflow, reimbursement, and guideline support
The investigators framed PROMIS CATs as feasible, acceptable, efficient, and clinically useful in physical therapy practice. Those are strong implementation signals, but they do not automatically guarantee widespread uptake. A measurement system has to survive three tests in the real world: it must be easy enough for patients, understandable enough for clinicians, and worth the effort financially and administratively. The study explicitly points to reimbursement as one of the implementation hurdles, alongside the need to embed PROMIS in clinical guidelines.
The economic detail is modest but relevant: the use of PROMIS CATs in this study was supported financially, with approximately €0.15 excluding VAT per CAT administration. In high-volume primary care, even small per-use costs can influence whether a clinic adopts a tool consistently. When a measure is perceived as clinically worthwhile but financially unsupported, adoption often remains patchy and dependent on local enthusiasm rather than becoming standard practice.
From a clinical perspective, PROMIS CATs fit a broader shift in musculoskeletal care toward capturing patient-centered outcomes directly from the patient rather than inferring progress solely from observation or therapist judgment. Pain intensity, physical function, and symptom burden are not always visible in the exam room, especially when patients compensate well or describe fluctuating symptoms. Adaptive patient-reported testing offers a way to capture those domains efficiently, but only if the workflow is built to use the information in the moment.
The next step is not simply more deployment. The system needs better score literacy among therapists, less duplication with older instruments, and broader PROMIS content for primary care physical therapy, where the current domains may not cover everything clinicians want to measure. Until those pieces are in place, the promise of PROMIS CATs will depend as much on implementation design as on measurement science.