From Stories to Themes: AI Qualitative Analysis

According to the November 2, 2026 PLOS Neglected Tropical Diseases study, participatory storytelling in Karachi was associated with reduced self-stigma, greater confidence, and stronger individual empowerment, although wider community stigma persisted. The PLOS Neglected Tropical Diseases study shows why AI qualitative analysis must preserve both thematic patterns and the social context surrounding participants’ accounts.

• The study recruited 17 people affected by leprosy, with 16 participating in photovoice and 14 completing in-depth interviews during June 2025, according to the November 2, 2026 PLOS study.
• The PLOS researchers reported a movement from shame and concealment toward self-acceptance, openness, confidence, and pride after the storytelling intervention.
• The PLOS researchers found that the strongest gains occurred at the individual level, while community misconceptions, gender norms, visible disability, and economic barriers continued to limit broader change.
• The PLOS study used thematic analysis supported by NVivo 14 and interpreted findings through the Social Ecological Model across five levels: individual, interpersonal, community, institutional, and structural.
• The study’s authors concluded that future programmes need sustained community contact, peer leadership, inclusive outreach to women and people with visible impairments, and links to socioeconomic opportunities.

The November 2, 2026 PLOS study found that participatory storytelling was associated with personal empowerment but did not eliminate community-level stigma. The 2024 pilot at the Marie Adelaide Leprosy Centre in Karachi combined peer support, creative activities, health education, advocacy, public engagement, and contact-based activities.

According to the PLOS study, researchers recruited participants purposively between May 30 and June 20, 2025, and collected data during a four-week period in June 2025. Interviews lasted 60 to 90 minutes, used live interpretation from Urdu, and were complemented by field notes and photovoice activities.

The PLOS researchers combined deductive coding based on stigma, empowerment, and Social Ecological Model concepts with inductive coding from the data. A senior researcher reviewed part of the coded material, and the research team refined themes through comparison across interviews and photovoice discussions.

The results show why qualitative health research requires more than counting word frequencies. One participant described the emotional effect of disclosure by saying, “I have no fear, no shame, not feeling any guilt, ” according to the PLOS study. Another participant summarized the peer group’s value as follows: “if we are together and unite, so we can do a better work.”

The PLOS study also reported important limits: women represented 4 of the 17 participants, the sample included only people who completed the intervention, and the design did not include baseline or comparison data. The findings therefore describe perceived impact and possible mechanisms rather than proving that the intervention caused every reported change.

Qualitative health researchers should use AI to accelerate synthesis while retaining human review of context, translation, power, and participant safety. The PLOS study shows that a theme such as empowerment can operate differently at individual, interpersonal, institutional, and community levels.

• Separate outcome themes from intervention mechanisms: the PLOS researchers distinguished confidence and self-acceptance from peer solidarity, institutional endorsement, and public contact.
• Compare segments deliberately: the PLOS study identified gender, visible disability, employment, and social circumstances as factors shaping participation and stigma.
• Preserve negative and contradictory evidence: the PLOS researchers reported personal gains alongside persistent discrimination, logistical barriers, and community misconceptions.
• Use multiple data sources: the PLOS study combined interviews, photographs, captions, group discussions, and field notes, which supports triangulated analysis rather than transcript-only synthesis.
• Treat AI-generated themes as analytic aids: the PLOS study’s interpretation depended on reflexivity, translation checks, participant context, and theoretical framing that automated pattern detection cannot independently supply.

Ethics note: This article discusses research about stigma and health, not diagnosis or clinical advice; AI-assisted analysis should use informed consent, de-identification, secure storage, and human oversight.

The November 2, 2026 PLOS study shows that participatory storytelling can support self-acceptance, confidence, peer solidarity, and individual empowerment, while wider stigma requires sustained and inclusive community action. AI qualitative analysis can help researchers manage complex interview, photovoice, and document datasets, but responsible interpretation still depends on context, reflexivity, and human judgment. Evidano brings thematic analysis, cross-segment comparison, transcription, translation, and document chat into one research workflow. Try Evidano for free to turn qualitative data into structured, reviewable insight.