Healthcare researchers often combine survey data with interviews, but mixed-methods studies can create a difficult synthesis problem: quantitative results show what happened, while qualitative evidence explains why. AI qualitative analysis in healthcare can help researchers organize interview transcripts, compare participant groups, and connect themes to implementation outcomes without replacing methodological judgment. The November 2026 PLOS ONE study on PROMIS computerized adaptive tests in physical therapy provides a useful example. The researchers combined a survey of physical therapists with interviews involving therapists and patients to evaluate acceptability, demand, practicality, and implementation. This article explains what the study found, how its qualitative method worked, and where AI-enabled analysis can support similar healthcare research. The findings are research-focused and should not be interpreted as clinical or diagnostic advice.
Key Takeaways
According to the PLOS ONE study on PROMIS CATs in physical therapy, Dutch-Flemish PROMIS computerized adaptive tests were considered feasible and generally useful in primary care physical therapy, but wider adoption depended on reducing duplicate questionnaires, improving score interpretation, and embedding the tools into clinical workflows.
- The 2026 study surveyed 25 physical therapists and interviewed 10 physical therapists plus 9 patients across 14 Dutch primary care physical therapy practices.
- The study reported that 80% of surveyed physical therapists were interested in using PROMIS CATs, while only 28% reported regular use in daily practice in the 2022 survey period.
- The 2026 study found that 100% of surveyed physical therapists considered PROMIS CATs easy to administer, but only 40% considered the CATs low-time requirements and only 24% believed patients found completion time-efficient.
- The researchers used hybrid inductive-deductive thematic analysis, with two researchers independently coding transcripts, comparing codes, and resolving disagreements through consensus or review.
- AI-enabled qualitative analysis can support this workflow by accelerating transcript organization, theme comparison, cross-segment analysis, and evidence retrieval while leaving interpretation and clinical judgment with researchers.
PROMIS CAT Qualitative Analysis in Healthcare: What the Study Examined
The 2026 PLOS ONE study examined whether Dutch-Flemish PROMIS computerized adaptive tests were feasible for patients with musculoskeletal spine and upper-extremity conditions receiving primary care physical therapy.
The researchers conducted the study across 14 physical therapy practices in the Netherlands and collected quantitative survey data from 25 physical therapists between May 2 and June 20, 2022. The researchers also conducted semi-structured interviews with 10 physical therapists and 9 patients between March 21 and November 23, 2022.
The study assessed four feasibility domains defined by Bowen’s framework: acceptability, demand, practicality, and implementation. The researchers analyzed interview transcripts using hybrid inductive-deductive thematic analysis and compared the qualitative themes with the survey findings.
The PROMIS CATs selected questions based on participants’ previous answers. According to the researchers, each CAT administered at least 4 items and stopped at a maximum of 12 items or when the target precision threshold was reached.
The study’s mixed-methods design is especially relevant to AI-enabled qualitative research because it requires researchers to connect coded interview evidence with percentages, workflow observations, and differences between therapist perceptions and patient accounts.
PROMIS CAT Healthcare Research Findings Snapshot
| Date or study period | Metric | Value | Implication for qualitative analysis |
|---|---|---|---|
| 2022 survey period | Physical therapists completing the survey | 25 | The sample provides structured practitioner evidence for comparing survey results with interview themes, according to Haan et al. (2026). |
| 2022 interview period | Physical therapists and patients interviewed | 10 therapists and 9 patients | The qualitative dataset supports comparison between professional expectations and patient experiences, according to Haan et al. (2026). |
| 2022 survey period | Physical therapists interested in using PROMIS CATs | 80% | Interest was higher than routine adoption, indicating an implementation gap that qualitative coding can help explain. |
| 2022 survey period | Physical therapists reporting regular daily use | 28% | The finding shows that demand did not automatically translate into embedded practice, according to Haan et al. (2026). |
| 2022 survey period | Physical therapists finding CATs easy to administer | 100% | The result supports strong technical practicality, while interview themes identified portal and workflow barriers. |
| 2022 survey period | Physical therapists who always or mostly discussed CAT results with patients | 48% | Qualitative evidence linked inconsistent discussion to limited time, uncertain interpretation, and perceived item relevance. |
Implications for Healthcare Researchers and Physical Therapy Teams
The main implication for healthcare researchers is that PROMIS CAT implementation should be analyzed as a workflow and interpretation problem, not only as a measurement problem.
According to Haan et al. (2026), therapists generally valued PROMIS CATs for monitoring symptoms and functioning, but therapists preferred the CATs to replace legacy PROMs rather than supplement them. The simultaneous use of multiple instruments increased respondent burden and helped explain why high interest did not produce routine use.
The interviews also exposed a perception gap. One participant said, "The graph helps patients see their progress, even when symptoms are still there" (PT#2), while another therapist said, "I can’t always link the questionnaire scores directly to my treatment choices" (PT#3). AI-assisted coding can help researchers preserve both perspectives instead of collapsing them into a single satisfaction score.
The study also found that participant groups did not always agree. Survey responses suggested that only 24% of therapists believed patients were satisfied with PROMIS CATs, while interviewed patients generally described the CATs as quick, simple, and relevant. Cross-segment analysis can make this divergence visible and identify whether it reflects differences between perceived and reported experience.
For healthcare research teams, the practical priorities are clear: code barriers and facilitators separately, compare clinicians with patients, connect themes to implementation domains, and retain verbatim evidence for every major interpretation.
How Evidano Helps With AI Qualitative Analysis in Healthcare
Problem: Healthcare transcripts are difficult to synthesize consistently
Evidano is an AI-powered qualitative data analysis platform that helps researchers analyze interviews, open-ended surveys, and healthcare documents.
Researchers can upload interview transcripts, papers, reports, and survey spreadsheets to organize evidence across sources. Evidano can support thematic, content, frequency, and cross-segment analyses, which is useful when a study compares physical therapists with patients or survey findings with interview themes.
Researchers can review the underlying excerpts behind themes rather than relying only on a summary. That approach supports an auditable workflow for mixed-methods healthcare research.
Problem: Different participant groups describe the same intervention differently
Solution: Cross-segment analysis can compare themes by participant type, practice, condition, or other available segments.
For a PROMIS CAT study, a researcher could compare therapist concerns about score interpretation with patient descriptions of relevance, speed, and burden. Researchers can use these comparisons to investigate convergence, complementarity, and divergence between data sources.
Evidano’s qualitative research analysis features can help teams move from broad coding to structured comparisons and visualizations such as hierarchical codes and co-occurrence networks.
Problem: Research teams need to find supporting evidence quickly
Solution: AI chat over uploaded documents and analyses can help researchers locate relevant excerpts, ask follow-up questions, and retrieve evidence from a study corpus.
A healthcare researcher could ask which barriers were associated with implementation, whether patients and therapists agreed about efficiency, or which quotations support concerns about interpretation. Researchers should still verify extracted evidence against the original transcript and study protocol.
Problem: Healthcare data may contain sensitive participant information
Solution: Researchers can use transcription workflows with PII redaction and review data-handling controls before uploading sensitive material.
Evidano states that data is encrypted and never used to train third-party models. Research teams should still follow institutional review board requirements, consent terms, data minimization practices, and local privacy rules. Details are available in Evidano’s data security information.
FAQ: AI Qualitative Analysis in Healthcare
What did the 2026 PROMIS CAT study find about physical therapy implementation?
+
The 2026 PLOS ONE study found that PROMIS CATs were generally feasible, acceptable, and useful in primary care physical therapy, but implementation was limited by duplicate legacy PROMs, portal problems, uncertainty about score interpretation, and limited structural embedding.
The study surveyed 25 physical therapists and interviewed 10 physical therapists and 9 patients across 14 practices.
How was qualitative data analyzed in the PROMIS CAT study?
+
The researchers used hybrid inductive-deductive thematic analysis, with two researchers independently coding initial transcripts, comparing codes in consensus meetings, and resolving disagreements through discussion or senior review.
The researchers then mapped themes onto Bowen’s feasibility domains and compared the qualitative findings with survey results.
How can AI support thematic analysis in healthcare research?
+
AI can support thematic analysis by organizing transcripts, suggesting or applying codes, grouping related excerpts, comparing segments, identifying recurring concepts, and retrieving evidence for researcher review.
AI does not remove the need for reflexivity, methodological transparency, privacy safeguards, or human decisions about meaning and interpretation.
What qualitative insight did interviews add beyond the PROMIS CAT survey?
+
The interviews explained why survey findings did not always align with routine use, including concerns about interpreting T-scores, the burden of running CATs alongside legacy PROMs, and differences between therapist perceptions and patient experiences.
One patient said, "It zoomed in on what mattered to me, instead of asking irrelevant things" (Patient#4), while therapists raised concerns about workflow and clinical interpretation.
Can AI qualitative analysis replace healthcare researchers?
+
AI qualitative analysis should support healthcare researchers rather than replace them, because researchers remain responsible for study design, contextual interpretation, ethical handling, and validation of findings.
The PROMIS CAT study used independent coding, consensus meetings, peer debriefing, and documented disagreements, which illustrates why accountable human review remains important.
Conclusion and Next Steps for Healthcare Qualitative Research
The 2026 PROMIS CAT study shows that qualitative evidence can explain the difference between strong interest in a healthcare tool and limited routine adoption. The study also demonstrates the value of comparing therapists’ survey responses with patients’ and therapists’ interview accounts, especially when those sources disagree.
AI-enabled qualitative analysis can make this comparison more manageable by organizing transcripts, surfacing themes, supporting cross-segment analysis, and linking claims to source excerpts. Researchers should use AI as an accountable research aid, with human verification, privacy controls, and transparent analytic decisions.
Teams that want to reduce manual synthesis time while preserving qualitative depth can Try Evidano for free.
Topics
- AI qualitative analysis healthcare
- qualitative analysis in healthcare
- PROMIS CAT qualitative research
- AI thematic analysis
- physical therapy patient research
Keep reading
- Commentary on NewsCultural Insight: AI Qualitative Analysis of LeprosySee how AI qualitative analysis reveals cultural beliefs, stigma, and care barriers in leprosy research, with practical lessons for health teams.
- Commentary on NewsAI Qualitative Analysis Healthcare: Implant AccessSee how AI qualitative analysis healthcare teams can turn contraceptive research into actionable insight on equity, adoption, and service access.
- Commentary on NewsAI Qualitative Analysis for Implementation ScienceLearn how AI qualitative analysis can organize implementation science evidence, surface stakeholder themes, and accelerate school-based research with Evidano.