Health researchers often need to combine interview transcripts, translated conversations, field notes, and visual methods before they can explain how an intervention works. AI qualitative data analysis can help researchers organize that evidence, compare participant groups, and identify themes without removing the interpretive role of the research team. A November 2, 2026 study in PLOS Neglected Tropical Diseases provides a useful example: researchers examined how participatory storytelling affected stigma and empowerment among people affected by leprosy in Karachi, Pakistan. This article explains what the study found, what its qualitative design makes visible, and how AI-assisted analysis can support similar health and social research while preserving context, reflexivity, and participant protections.
Key Takeaways
According to Owning the narrative: Exploring the impact of a creative storytelling intervention on stigma and empowerment among persons affected by leprosy in Pakistan, participatory storytelling was associated with movement from shame and concealment toward self-acceptance, confidence, and openness among participants affected by leprosy in Karachi.
The November 2, 2026 PLOS study shows why AI qualitative data analysis is useful for health research: researchers can organize interviews, photovoice materials, translations, and demographic segments while keeping the interpretation grounded in participants’ accounts.
- The study analyzed a 2024 to 2025 storytelling pilot with 17 total participants, including 16 photovoice participants and 14 in-depth interview participants.
- The research team collected interview data between June 2 and June 25, 2025, using 60 to 90 minute interviews conducted with live Urdu interpretation.
- According to the study, confidence was reported as the most common change, while the strongest effects occurred at the individual level through reduced internalized and anticipated stigma.
- The study found that peer support, creative expression, and organizational endorsement strengthened empowerment, but community misconceptions, gender norms, visible disability, and economic barriers limited wider change.
- The study reported that women represented 4 of the 17 participants, a limitation that narrowed the range of gender-specific experiences captured.
What the Pakistan Storytelling Study Measured
The November 2, 2026 PLOS study examined how participatory storytelling influenced stigma and empowerment among people affected by leprosy in Pakistan.
The researchers studied a storytelling pilot delivered through the Marie Adelaide Leprosy Centre in Karachi during 2024, combining peer-support groups, creative activities, health education, advocacy, contact-based activities, and public storytelling.
The study used in-depth interviews and photovoice, then analyzed the material thematically through the Social Ecological Model. The model helped the researchers distinguish individual, interpersonal, community, institutional, and structural effects rather than treating empowerment as a single outcome.
The study team purposively sampled adults aged 30 to 65 who had participated in the intervention. The researchers recruited participants between May 30 and June 20, 2025, and collected data during a four-week period in June 2025.
The researchers analyzed interview and photovoice material using deductive and inductive coding, a developing codebook, team review, participant reflection, and NVivo 14. The study also acknowledged important limitations, including the absence of baseline or comparator data, translation complexity, possible social desirability bias, and the small sample.
One participant described the emotional effect of disclosure as follows: "I have no fear, no shame, not feeling any guilt. I’m feeling like, if someone accept his crime in front of someone. I accept all those crimes, I feel lighter and so relaxed." The PLOS study attributed this quotation to participant P06, a man in his 40s.
Findings Snapshot: AI Qualitative Data Analysis Opportunities
| Date or study detail | Metric | Value | Implication for qualitative analysis |
|---|---|---|---|
| 2024 | Storytelling pilot duration | One year | Longitudinal program materials can be organized alongside participant accounts to examine intervention processes. |
| May 30 to June 20, 2025 | Recruitment period | 22 days | Segmented sampling records can help analysts compare gender, age, disability, and social circumstances. |
| June 2 to June 25, 2025 | Interview fieldwork | 24 days | Interview timing and context can be linked to transcripts, field notes, and interpretation decisions. |
| 2025 | Total study participants | 17 | Small qualitative samples require close reading, transparent coding, and careful limits on generalization. |
| 2025 | Photovoice participants | 16 | Visual narratives and captions can be analyzed with interview transcripts to support triangulation. |
| 2025 | In-depth interview participants | 14, including 13 who also participated in photovoice | Cross-method comparison can identify themes that recur across spoken and visual data. |
| 2025 | Women represented in the sample | 4 of 17 participants | Gender-sensitive analysis is essential because uneven representation can constrain interpretation. |
Implications for Health and Social Researchers
Health and social researchers should treat AI qualitative data analysis as an aid for organizing evidence, not as a substitute for contextual interpretation.
The PLOS study shows that thematic analysis must distinguish between levels of change. Participants described stronger individual gains, including confidence and self-acceptance, while community-level stigma remained more persistent. An AI-assisted workflow can make this distinction easier by organizing codes under a framework such as the Social Ecological Model.
Researchers should preserve participant language when analyzing translated interviews. The PLOS team used live Urdu interpretation, repeated listening, meaning checks, and reflexive discussions with translators. AI translation or transcription can accelerate preparation, but researchers should review culturally specific expressions, ambiguous meanings, and quotations against the source material.
Researchers should compare segments rather than relying only on overall frequency. The PLOS study reported that gender, visible disability, employment, and household circumstances shaped participation and stigma. Cross-segment analysis can help identify where an intervention works differently for women, people with visible impairments, or participants facing economic barriers.
Researchers should connect themes to decisions about program design. Participants in the PLOS study recommended sustained community contact, peer-led outreach, gender-responsive participation, disability inclusion, and economic opportunities. A structured coding system can connect these recommendations to supporting quotations, affected groups, and implementation barriers.
The PLOS study included sensitive health experiences, including stigma and mental distress. Health researchers using AI should apply informed consent, de-identification, access controls, and human review, and should avoid treating qualitative findings as diagnostic evidence.
How Evidano Supports AI Qualitative Data Analysis
Problem: Interview and photovoice materials are difficult to synthesize together
Evidano is an AI-powered qualitative data analysis platform that helps researchers analyze interviews, open-ended surveys, papers, reports, and other documents.
Evidano can ingest transcripts and documents so research teams can organize interview accounts, photovoice descriptions, field notes, and study materials in one analysis environment. Researchers can then use thematic, content, frequency, and cross-segment analyses to compare recurring ideas and differences.
Problem: Translation and transcription can introduce avoidable delays
Evidano provides transcription with custom dictionaries and PII redaction, as well as translation with custom dictionaries.
For multilingual health research, custom terminology can help teams review disease names, program terms, participant language, and culturally specific expressions more consistently. Human researchers should still verify sensitive quotations and translation choices.
Problem: Framework-based coding can hide relationships between themes
Evidano supports visualizations such as word clouds, co-occurrence networks, and hierarchical codes with subcodes.
Researchers can use these views to examine relationships between stigma, empowerment, peer support, gender, disability, and community contact. Visual outputs can support team discussions, while final interpretations should remain tied to the source evidence.
Problem: Researchers need to interrogate a large evidence set without losing traceability
Evidano provides AI chat over uploaded documents and analyses, allowing researchers to ask focused questions about themes, segments, quotations, and evidence.
For a study like the Pakistan storytelling research, an analyst could ask which excerpts describe internalized stigma, which participants mention community outreach, or how women’s participation differs from men’s accounts, then review the underlying material before reporting a finding.
Evidano states that customer data is encrypted and never used to train third-party models. Researchers handling sensitive health narratives can review the platform’s data security information before selecting an analysis workflow.
FAQ: AI Qualitative Data Analysis in Health Research
What is AI qualitative data analysis?
+
AI qualitative data analysis uses artificial intelligence to help researchers organize, code, compare, summarize, and query non-numerical research data such as interviews, open-ended surveys, and documents.
AI can accelerate repetitive analysis tasks, but the November 2, 2026 PLOS study demonstrates why researchers must retain human interpretation of context, translation, power dynamics, and participant meaning.
How did the Pakistan study analyze storytelling and stigma?
+
The Pakistan study used thematic analysis of 14 in-depth interviews and photovoice data from 16 participants, with 13 participants contributing to both methods.
The researchers combined deductive and inductive coding and interpreted findings through the Social Ecological Model, covering individual, interpersonal, community, institutional, and structural levels.
What were the main findings of the leprosy storytelling intervention?
+
The main finding was that participants described greater self-acceptance, confidence, openness, peer solidarity, and identity transformation after the intervention.
The study also found that community misconceptions, gender norms, visible disability, and structural barriers constrained broader social change. Participant P07, a female health worker in her 50s, said, "My communication skills are better. I can relate to different stories, and I can convince people better."
Can AI qualitative analysis replace a qualitative researcher?
+
AI qualitative analysis cannot replace a qualitative researcher’s responsibility for study design, ethical judgment, reflexivity, interpretation, and claims about meaning.
AI can help researchers find patterns and compare segments, while researchers must validate codes, inspect quotations, account for translation, and report limitations such as the PLOS study’s small sample and lack of comparator data.
How can researchers analyze translated health interviews with AI?
+
Researchers can use AI transcription and translation to prepare multilingual interview data, then verify culturally important phrases, ambiguous passages, and quotations against recordings or original-language material.
The PLOS researchers used live Urdu interpretation, repeated listening, translator consultations, and meaning checks, which provides a useful quality standard for AI-assisted multilingual workflows.
What ethical safeguards matter when using AI for sensitive qualitative data?
+
Researchers should obtain informed consent, remove or redact identifying information, restrict access, secure data, and keep human oversight over interpretation and reporting.
The PLOS study obtained consent for participation, recording, and photography, anonymized data, stored data securely, and made counselling support available when needed. Health research analysis should remain non-diagnostic and research-focused.
Conclusion and Next Steps for AI Qualitative Data Analysis
The November 2, 2026 PLOS study shows how participatory storytelling can support individual empowerment while leaving wider structural and community barriers unresolved.
AI qualitative data analysis can help health researchers manage mixed qualitative materials, compare participant segments, preserve evidence behind themes, and identify where an intervention has or has not produced change.
The strongest workflow combines AI-supported organization with human review of context, translation, ethics, reflexivity, and participant voice. Researchers who want to analyze interviews, reports, and open-ended responses in one workspace can try Evidano for free.
Topics
- AI qualitative data analysis
- AI qualitative research
- health interview analysis
- thematic analysis of interviews
- photovoice research analysis
Keep reading
- Commentary on NewsSmarter Care: Qualitative Analysis in HealthcareLearn how AI-enabled qualitative analysis in healthcare can turn interviews into themes, using the 2026 SYNC-RF protocol as a rigorous research example.
- Commentary on NewsStudent Voices: AI Thematic Analysis of School FoodLearn how AI thematic analysis can turn student focus groups into actionable food policy insights, using a Northern Benin study as a practical guide.
- Commentary on NewsAI Thematic Analysis for Healthcare ResearchSee how AI thematic analysis helps healthcare researchers synthesize interviews, surveys, and mixed-methods evidence for better care decisions. Try Evidano.