Smarter Care Insights: AI Thematic Analysis in Healthcare

According to the PLOS ONE SYNC-RF study protocol, AI thematic analysis in healthcare can support the organization of interview evidence within a broader mixed-methods design, but the protocol itself does not evaluate an AI analysis tool or report completed health outcomes.

• The SYNC-RF protocol, published on November 3, 2026, describes three linked phases: a systematic review, a district-level mixed-methods assessment, and development of an integrated referral and follow-up model.
• The protocol plans to assess 20 Primary Health Centres, including 10 rural and 10 urban facilities, to examine referral capacity and follow-up systems.
• The protocol plans to survey 600 adults aged 30 years and above who have two or more chronic conditions, with referral completion within 30 days as a secondary outcome.
• The protocol plans in-depth interviews with 10 patients, 10 caregivers, 10 healthcare providers, and 4 to 6 health officials, with the final qualitative sample guided by information power and data saturation.
• The authors state that the study will use reflexive thematic analysis and a joint display matrix to compare quantitative and qualitative findings before refining the SYNC-RF model.

The SYNC-RF protocol addresses fragmented referral and follow-up for people living with multiple long-term conditions in an Indian district health system. According to the authors, weak referral documentation, limited communication between healthcare levels, and inadequate follow-up can contribute to discontinuity of care.

The protocol uses a sequential, multistage mixed-methods design. In Phase 1, the researchers will synthesize global evidence through a systematic review. In Phase 2, the researchers will combine facility assessments, a patient survey, and in-depth interviews in Jodhpur district, Rajasthan. In Phase 3, the researchers will use the combined evidence and expert consultation to develop and refine the System for Navigation, Coordination, Referral, and Follow-up, abbreviated as SYNC-RF.

The protocol states that the researchers will use reflexive thematic analysis for interview transcripts. The researchers plan to read transcripts repeatedly, apply deductive and inductive codes, group codes into sub-themes and themes, and discuss theme development within the research team.

The authors describe the study as a way to combine structural and experiential evidence. Facility data can show how referral systems are organized, survey data can quantify referral experiences, and interviews can explain why patients, caregivers, providers, and policymakers encounter barriers or enabling conditions.

The authors summarize the intended progression as follows: "evidence generation, contextual assessment, and model development." This quotation describes the protocol's research logic, not a completed finding.

Healthcare researchers can use AI thematic analysis to accelerate organization of large qualitative datasets while retaining human responsibility for interpretation, reflexivity, and ethical decisions.

• Researchers analyzing patient interviews should separate descriptive coding from interpretation so that an AI-generated pattern does not become an unsupported clinical conclusion.
• Researchers working across languages should preserve original-language transcripts alongside translations because the SYNC-RF protocol identifies cultural nuance and translation loss as methodological risks.
• Researchers comparing patients, caregivers, providers, and policymakers should use cross-segment analysis to identify shared barriers, distinct experiences, and disagreements between intended and experienced referral processes.
• Researchers integrating interviews with surveys should create explicit links between qualitative themes and quantitative indicators such as referral initiation, referral completion within 30 days, follow-up contact, and travel burden.
• Research teams should treat AI outputs as reviewable analytic aids because the SYNC-RF protocol describes a reflexive thematic process that requires repeated reading, iterative coding, and team discussion.

The SYNC-RF authors write that discrepancies between patient reports and provider descriptions should not be "automatically resolved in favour of one data source." That principle is especially important when AI tools summarize multiple perspectives, because disagreement may be a substantive finding rather than a data-cleaning problem.

Healthcare research involving patient data requires privacy safeguards and approved consent procedures. The SYNC-RF protocol reports ethics approval from the Centre for Chronic Disease Control and the University of Leicester, and this article does not provide diagnosis or medical advice.

The SYNC-RF protocol shows that healthcare researchers need to connect qualitative experiences with facility assessments and survey indicators when studying continuity of care for multiple long-term conditions. AI thematic analysis can make that work more manageable by supporting transcription, translation, coding, cross-segment comparison, and evidence retrieval, while researchers retain control over interpretation and validation. The protocol's planned sample of 600 survey participants, 20 Primary Health Centres, and multiple interview groups illustrates the scale and complexity of modern health-systems research. Teams that want a secure workflow for organizing healthcare interviews and mixed-methods evidence can try Evidano for free.