PROMIS CATs: AI Qualitative Analysis of Patient Interviews

According to the PLOS One study on PROMIS CAT feasibility published on November 13, 2026, PROMIS CATs were feasible in primary care physical therapy, but successful implementation depended on reducing duplicate questionnaires, improving score interpretation, and embedding the tools into clinical workflows.

AI qualitative analysis is useful for this type of mixed-methods research because it can organize interview transcripts and survey comments around feasibility domains, then compare where patient and clinician accounts converge or diverge.

• The November 13, 2026 PLOS One study covered 14 physical therapy practices, 25 surveyed physical therapists, 10 interviewed physical therapists, and 9 interviewed patients.
• The study reported that 100% of surveyed physical therapists found PROMIS CATs easy to administer, while 72% rated them easy to use.
• The study found that 80% of surveyed physical therapists were interested in using PROMIS CATs, but only 28% reported regular daily use.
• The study reported that 56% of surveyed physical therapists wanted to see individual item responses, showing demand for more granular qualitative and quantitative feedback.
• The study's interviews showed that patients generally considered the CATs quick and relevant when results were used during consultations.

The study measured PROMIS CAT feasibility through acceptability, demand, practicality, and implementation, using survey data from physical therapists and semi-structured interviews with physical therapists and patients.

According to the November 13, 2026 PLOS One article, the research team integrated Dutch-Flemish PROMIS CATs into electronic health record workflows in 14 Dutch primary care physical therapy practices. The CATs selected questions adaptively, administering a minimum of 4 and a maximum of 12 items, with a precision-based stopping rule.

The researchers conducted interviews from March 21 through November 23, 2022, and interview sessions lasted between 32 and 62 minutes. Two researchers independently performed open coding, resolved disagreements through consensus meetings, and mapped inductively developed themes onto Bowen's feasibility domains.

The qualitative method matters for AI-enabled research because the study did not rely on frequency counts alone. Thematic analysis explained why survey results sometimes differed from participant accounts, particularly around patient satisfaction and time efficiency.

AI qualitative analysis can clarify that PROMIS CAT adoption was constrained less by the adaptive questionnaire itself than by surrounding workflow, interpretation, and policy conditions.

According to the PLOS One researchers, patients generally valued PROMIS CATs when the purpose was clear and results were discussed, while physical therapists preferred the CATs as replacements for legacy PROMs rather than as additional questionnaires. One physical therapist described the burden directly: "Only about 10–20% fill them out beforehand; people get questionnaire fatigue" (PT#8).

The interviews also exposed a difference between stakeholder perceptions. A patient said, "It zoomed in on what mattered to me, instead of asking irrelevant things" (Patient#4), while the survey showed that only 24% of physical therapists believed patients were satisfied with PROMIS CATs. The PLOS One authors interpreted this divergence as a difference between therapists' perceptions and patients' direct accounts.

The study identified score interpretation as another implementation theme. A physical therapist warned, "If patients see the scores without explanation, they may misinterpret the numbers and develop unhelpful thoughts" (PT#6). AI-assisted coding can surface such concerns across transcripts, but researchers must retain responsibility for context, interpretation, and ethical review.

• For health researchers, code transcripts by feasibility domain first, then compare themes by stakeholder group.
• For implementation teams, distinguish barriers caused by PROMIS CATs from barriers caused by portals, duplicate measures, reimbursement, and training.
• For mixed-methods analysts, use AI to locate supporting excerpts for survey findings, then manually verify every quotation and interpretation.
• For patient-centered research, preserve item-level comments and negative cases instead of reporting only the most frequent theme.

The November 13, 2026 PLOS One study shows that PROMIS CATs can support focused, patient-centered conversations, but adoption depends on workflow integration, training, replacement of redundant measures, and clear interpretation.

AI qualitative analysis gives healthcare researchers a practical way to connect survey patterns with interview evidence, compare patient and clinician perspectives, and identify implementation barriers at scale.

Research teams should use AI to accelerate organization and retrieval while keeping human researchers responsible for methodological decisions, quotation accuracy, privacy, and clinical context.

Teams working with patient interviews, open-ended surveys, or implementation evidence can Try Evidano for free.