Lower Burden: AI Thematic Analysis in Healthcare

Healthcare researchers and physical therapy teams often collect patient-reported outcomes alongside interviews, surveys, and clinical workflow data, but the qualitative evidence can be difficult to synthesize consistently. AI thematic analysis in healthcare can help researchers identify recurring barriers, compare patient and clinician perspectives, and connect qualitative themes with implementation metrics. A November 2026 mixed-methods study in primary care physical therapy provides a useful example: researchers examined Dutch-Flemish PROMIS Computerized Adaptive Tests through a survey of physical therapists and interviews with therapists and patients. The study found that PROMIS CATs were generally acceptable and practical, while also revealing concerns about score interpretation, digital workflows, and the continued use of legacy questionnaires. This article explains what the study found, how its thematic analysis was structured, and how AI-assisted qualitative research can help teams move from transcripts to defensible insights without replacing human interpretation.

Key Takeaways

According to the November 2026 PLOS One study, PROMIS computerized adaptive tests were feasible and generally acceptable in primary care physical therapy, but implementation depended on reducing duplicate questionnaires, improving score interpretation, and aligning results with clinical workflows.

The November 2026 study included 25 physical therapists in a survey, plus interviews with 10 physical therapists and 9 patients across 14 Dutch physical therapy practices.
The study reported that 100% of surveyed physical therapists found PROMIS CATs easy to administer, while 72% rated the tools easy to use.
Only 48% of surveyed physical therapists reported always or mostly discussing CAT results with patients, according to the November 2026 study.
The study found a perception gap: 80% of physical therapists expressed interest in using PROMIS CATs, but only 28% reported regular use in daily practice.
AI thematic analysis can help healthcare researchers organize interview evidence around implementation domains such as acceptability, demand, practicality, and implementation, while keeping source quotations available for human review.

What the PROMIS CAT Study Measured

The November 2026 PLOS One study evaluated whether Dutch-Flemish PROMIS CATs could work in routine primary care physical therapy for adults with musculoskeletal conditions affecting the spine and upper extremities.

According to the study methods, researchers used a concurrent mixed-methods design: a survey measured physical therapists’ experiences, while semi-structured interviews explored the experiences of physical therapists and patients in greater depth. The researchers analyzed interview transcripts using hybrid inductive and deductive thematic analysis, with two researchers independently coding transcripts before consensus meetings.

The study organized feasibility around four domains from Bowen’s feasibility framework: acceptability, demand, practicality, and implementation. This structure is especially useful for AI-assisted qualitative research because researchers can use the domains as an initial coding framework while allowing new themes to emerge from participant language.

The PROMIS CATs selected items adaptively based on prior answers. According to the November 2026 study, each CAT administered at least 4 items and stopped at a maximum of 12 items, or earlier when the required precision threshold was reached.

The researchers reported that interviews lasted 32 to 62 minutes and took place between March 21 and November 23, 2022. The research team checked all transcripts for accuracy after transcription.

PROMIS CAT Findings Snapshot for Healthcare Researchers

Date or study detail	Metric	Value	Implication for qualitative research
November 2026	Physical therapy practices	14	The findings reflect implementation across multiple real-world practice settings.
November 2026	Physical therapists completing the survey	25	The survey provides descriptive context for the interview themes.
November 2026	Physical therapist interviews	10	Clinician interviews explain workflow and interpretation barriers.
November 2026	Patient interviews	9	Patient accounts provide a second perspective on relevance and burden.
November 2026	Physical therapists reporting CATs were easy to administer	100%	The quantitative result supports the qualitative theme of technical practicality.
November 2026	Physical therapists interested in using PROMIS CATs	80%	Interest exceeded routine adoption, indicating an implementation gap.
November 2026	Physical therapists reporting regular daily use	28%	Qualitative analysis is needed to explain why interest did not become routine behavior.

What AI Thematic Analysis Reveals About Patient and Clinician Feedback

Why did participants view PROMIS CATs as useful?

The November 2026 PLOS One study found that physical therapists and patients generally viewed PROMIS CATs as quick, relevant, and useful for monitoring symptoms and functioning.

A patient identified in the study as Patient #5 said, "Answering the questions made me reflect on my symptoms and what I want help with." A physical therapist identified as PT #2 said, "The graph helps patients see their progress, even when symptoms are still there." These quotations show why qualitative analysis adds context to usability percentages.

What barriers did the qualitative analysis identify?

The November 2026 study identified duplicate legacy PROMs, portal and login problems, uncertainty about interpreting CAT scores, and concerns about the relevance of some items as implementation barriers.

The study also found that participants wanted access to individual item responses, not only aggregate scores or graphs. A physical therapist identified as PT #3 said, "I can’t always link the questionnaire scores directly to my treatment choices, " illustrating a clinical interpretation problem rather than a purely technical problem.

What did the study reveal about mixed-methods integration?

The November 2026 study found convergence between survey and interview data on overall acceptability, but it also found divergence around patient satisfaction.

Only 24% of surveyed physical therapists believed patients were satisfied with PROMIS CATs, while interviewed patients generally described the CATs as quick, simple, and relevant. AI thematic analysis can help researchers locate these contradictions, compare the groups that produced them, and preserve the quotations needed to investigate why the results differ.

How Evidano Supports AI Thematic Analysis in Healthcare

Problem: Healthcare teams have too many interview transcripts to review manually

Evidano is an AI-powered qualitative data analysis platform that helps researchers analyze healthcare interviews, patient feedback, surveys, and research documents.

Evidano can organize transcripts with thematic, content, frequency, and cross-segment analyses. Researchers can use these analyses to compare patient and clinician perspectives, surface recurring implementation barriers, and identify themes that require closer human review. Teams can learn more about Evidano’s qualitative analysis features.

Problem: Researchers need evidence behind every theme

Evidano’s AI chat over documents and analyses can help researchers trace a theme back to relevant passages in source material instead of relying only on a summary.

For studies like the PROMIS CAT research, a researcher could examine whether concerns about score interpretation appear more often among physical therapists than patients and then review the underlying quotations before reporting the result.

Problem: Patient data requires careful handling

Healthcare qualitative research requires privacy-conscious workflows because interview transcripts may contain sensitive or identifying information.

Evidano encrypts data and does not use customer data to train third-party models. Researchers can also use transcription with PII redaction when preparing interview data for analysis. Teams should still follow their institutional review board, consent, retention, and access requirements.

Problem: Mixed-methods findings need clear comparisons

Evidano can support cross-segment analysis so researchers can compare themes across patient and clinician groups, demographic segments, study sites, or survey response categories.

The PROMIS CAT study demonstrates why this comparison matters: clinician perceptions of patient satisfaction differed from the accounts provided by interviewed patients. AI-assisted comparisons can make such divergences easier to detect, while researchers remain responsible for interpretation and methodological decisions.

FAQ: AI Thematic Analysis in Healthcare

What is AI thematic analysis in healthcare?

AI thematic analysis in healthcare is the use of artificial intelligence to help identify, group, compare, and summarize recurring ideas in clinical interviews, patient feedback, open-ended surveys, and research documents.

The November 2026 PLOS One study illustrates a suitable use case because researchers needed to connect interview themes about acceptability, workflow, and implementation with survey findings from physical therapists.

How did the PROMIS CAT study analyze interview data?

The November 2026 PLOS One study used hybrid inductive and deductive thematic analysis conducted by two researchers who independently coded transcripts and resolved differences through discussion.

The researchers mapped emerging themes onto Bowen’s four feasibility domains: acceptability, demand, practicality, and implementation. AI can accelerate similar coding and comparison tasks, but researchers should review outputs and retain responsibility for analytic judgment.

What were the main PROMIS CAT implementation barriers?

The main barriers reported by the November 2026 study were continued use of legacy PROMs, online portal difficulties, uncertainty about interpreting CAT scores, and concerns that some items were insufficiently relevant or specific.

The study also identified the need for training, financial reimbursement, professional guidance, and better access to item-level responses.

Can AI replace human qualitative researchers in healthcare?

AI can assist healthcare researchers with transcription, coding, theme comparison, and evidence retrieval, but AI should not independently replace human qualitative interpretation.

The November 2026 study used independent coding, consensus meetings, a disagreement log, and peer debriefing, showing the value of transparent human oversight. AI-assisted workflows should preserve those safeguards where appropriate.

How can researchers analyze patient and clinician perspectives separately?

Researchers can analyze patient and clinician data as separate segments, then compare theme frequency, language, and supporting quotations across groups.

The November 2026 study shows the value of this approach because physical therapists’ survey perceptions of patient satisfaction did not fully match the accounts of interviewed patients.

Conclusion: Use AI to Make Healthcare Qualitative Research More Traceable

The November 2026 PLOS One study shows that PROMIS CATs can be feasible in physical therapy, but successful implementation depends on workflow fit, understandable feedback, and reducing duplicate measurement burden.

The study also demonstrates why qualitative evidence matters: percentages showed interest and usability, while interviews explained the barriers behind limited routine adoption and exposed differences between clinician assumptions and patient experiences.

AI thematic analysis can help healthcare researchers process these mixed sources more efficiently, compare segments, and retrieve supporting evidence without removing human responsibility from interpretation.

Researchers who want to analyze interviews, patient feedback, and healthcare documents in one secure workspace can try Evidano for free.

Topics

AI thematic analysis in healthcare
qualitative analysis of patient feedback
PROMIS CAT qualitative research
AI healthcare research analysis
thematic analysis of interviews

Keep reading

Browse all articles