Health researchers often need to analyze multilingual interviews, culturally specific terms, and sensitive accounts of stigma without flattening the meaning behind participants' words. AI thematic analysis can accelerate that work, but health research requires more than automated pattern matching: researchers must preserve local language, distinguish participant beliefs from biomedical explanations, and protect confidential transcripts. The November 2026 study by Aurore Dupont, Wim Saakie, Marc-Alexandre Tareau, and Romain Blaizot on leprosy perceptions in French Guiana provides a useful example. This article explains what the study found, how its manual thematic analysis worked, and how AI-assisted qualitative research can support similar public health projects without replacing cultural interpretation or researcher judgment.
Key Takeaways
According to the November 11, 2026 PLOS Neglected Tropical Diseases study, qualitative interviews in French Guiana showed that local terms, spiritual explanations, stigma, and treatment experiences shaped how people understood leprosy.
According to the study, AI thematic analysis could help public health teams organize multilingual interview evidence and compare themes, but culturally sensitive interpretation and secure handling of sensitive transcripts remain essential.
- The PLOS study interviewed 20 patients and 37 community members between April and September 2025 across French Guiana.
- The PLOS study identified the Maroon terms “kokobe” and “gwasi” as culturally specific descriptions associated with deformities, skin patches, and leprosy-related illness.
- The PLOS study found that bacterial causation was rarely mentioned, while spiritual explanations involving taboos and curses were common among Maroon and Haitian participants.
- The PLOS study reported that local plant-based remedies were widely used, but those remedies did not appear to reduce adherence to multidrug therapy among participants.
- The PLOS study concluded on November 11, 2026 that culturally sensitive communication should address local terminology, person-to-person transmission, stigma, and care pathways.
What the French Guiana Study Shows About Qualitative Health Research
The French Guiana study used semi-structured interviews because the researchers were investigating cultural explanations, lived experiences, stigma, and care-seeking pathways in a linguistically diverse setting.
According to the PLOS study, researchers interviewed 20 patients across health centres and 37 community members in western French Guiana, continuing recruitment until thematic saturation was reached.
According to the PLOS study, patient interviews were conducted in French or with interpretation into participants' native languages, while community interviews often relied on cultural mediation and detailed field notes.
According to the PLOS study, the research team combined deductive coding based on the interview guide with inductive coding to identify emergent themes and culturally specific meanings.
The PLOS study illustrates why qualitative health analysis must preserve participants' own categories instead of treating every local term as a direct synonym for a biomedical diagnosis. The study explains that “kokobe” primarily referred to deformities and disabilities, while “gwasi” was more closely associated with skin lesions, although the meanings could overlap.
One community participant told the researchers, “It starts with spots... If you don't treat it, your fingers can bend.” The quotation shows how local descriptions can encode a progression of symptoms that may not map neatly onto clinical terminology.
The study authors wrote that “Healthcare professionals should therefore respect cultural representations without discouraging biomedical treatment.” The finding supports a research workflow that identifies cultural concepts for discussion rather than automatically treating them as misinformation.
French Guiana Leprosy Research Findings Snapshot
| Date or period | Metric | Value | Implication for qualitative analysis |
|---|---|---|---|
| April to September 2025 | Patient participants | 20, according to the PLOS study | AI-assisted coding can organize varied illness histories, explanatory models, and care pathways. |
| April to September 2025 | Community participants | 37, according to the PLOS study | Community narratives should be analyzed separately from patient experiences when the research design treats them as complementary perspectives. |
| April to September 2025 | Total interviews | 37, according to the PLOS study | Interview-level metadata can help researchers trace themes back to settings, participant groups, and language mediation. |
| 2015 to 2021 | Reported annual leprosy prevalence in French Guiana | About 1 case per 10, 000 inhabitants, according to the PLOS study's cited background evidence | Low-frequency conditions can still require intensive qualitative monitoring because stigma and delayed recognition may remain significant. |
| Study analysis reported in November 2026 | Patient interviews requiring mediation or a relative | 12 of 19 patient interviews, according to the PLOS study | Translation and mediation context should remain attached to coded excerpts during analysis. |
| Study analysis reported in November 2026 | Community interviews requiring mediation | 17 of 18, according to the PLOS study | Cultural interpretation is part of the data-generation process and should be documented as an analytic context. |
Implications for Public Health and UX Researchers
How can researchers analyze culturally specific health beliefs without erasing context?
Researchers can preserve context by coding local terms, explanatory models, language, participant group, and the surrounding quotation together.
The PLOS study shows why researchers should distinguish between a local classification such as “kokobe” or “gwasi” and the biomedical category of leprosy. An AI system can flag recurring terms and co-occurring concepts, while human researchers decide how those concepts function within the community.
How can AI thematic analysis support multilingual interview projects?
AI thematic analysis can support multilingual projects by helping researchers transcribe, translate, cluster, and compare recurring themes across interview documents.
The PLOS study involved French interviews, interpretation into native languages, Maroon-language mediation, Haitian Creole assistance, and field notes. That design indicates why translated text should be reviewed against the original wording and accompanied by language and mediation metadata.
How can researchers turn qualitative findings into better health communication?
Researchers can turn qualitative findings into better communication by identifying the exact misconceptions, fears, and trusted practices that health messaging must address.
The PLOS study found that person-to-person transmission was poorly understood, spiritual and familial explanations were common in some groups, and confusion with diabetes occurred. Those findings point toward communication that explains transmission clearly while acknowledging local terminology and treatment practices.
What ethical safeguards matter when analyzing sensitive health interviews?
Sensitive health interview analysis requires confidentiality, controlled access, careful de-identification, and restraint in sharing excerpts.
The PLOS study did not publicly share interview transcripts because rare disease narratives and small communities could make participants identifiable even after pseudonymization. The study also reported that transcripts were securely stored at the French Guiana University Hospital.
How Evidano Helps with AI Thematic Analysis
Problem: Manual coding slows synthesis across interviews
Evidano is an AI-powered qualitative data analysis platform that helps researchers analyze interview transcripts, open-ended surveys, and research documents.
Evidano can support thematic, content, frequency, and cross-segment analyses, helping teams move from a large set of interview documents to recurring themes, subthemes, and comparisons. Researchers can review the evidence behind each theme instead of relying only on an automatically generated summary.
For teams working with multilingual health research, Evidano's translation tools can help prepare material for analysis while preserving a custom dictionary for culturally specific terms such as local disease names.
Problem: Cultural terms and themes are difficult to compare
Evidano can organize hierarchical codes and subcodes, co-occurrence networks, and frequency visualizations so researchers can examine how terms such as stigma, transmission, family explanations, treatment, and local remedies appear together.
The PLOS study's distinction between “kokobe, ” “gwasi, ” visible symptoms, and biomedical explanations is an example of why qualitative analysis should compare concepts rather than simply count words.
Problem: Researchers need to interrogate findings after the first coding pass
Evidano's AI chat over documents and analyses can help researchers ask follow-up questions about participant groups, interview settings, recurring beliefs, or differences between patient and community narratives.
Researchers remain responsible for checking quotations, interpreting cultural meaning, and deciding whether a pattern is analytically meaningful. AI assistance should support reflexive research practice, not replace it.
Problem: Health research requires strong data controls
Evidano encrypts data and does not use customer data to train third-party models, according to Evidano's product information.
Researchers handling sensitive interviews should still apply informed-consent requirements, remove unnecessary identifiers, limit workspace access, and follow the governing ethics and data-protection rules for their project. Evidano's data security information provides additional product context.
FAQ: AI Thematic Analysis of Health Interviews
What is AI thematic analysis in qualitative health research?
+
AI thematic analysis is the use of artificial intelligence to help identify, organize, compare, and retrieve themes in qualitative health data such as interview transcripts and field notes.
The PLOS French Guiana study used manual thematic analysis, combining deductive and inductive coding. AI can assist with similar organizational tasks, but researchers must validate codes and interpret cultural meaning.
What did the 2026 French Guiana leprosy study find?
+
The 2026 PLOS study found that leprosy perceptions in French Guiana were shaped by local terms, spiritual and familial explanations, stigma, treatment experiences, and community support.
The study included 20 patients and 37 community members, and it concluded that culturally sensitive communication should clarify transmission while taking local terminology and practices into account.
Why are “kokobe” and “gwasi” important for qualitative analysis?
+
“Kokobe” and “gwasi” are important because the PLOS study found that the terms carry culturally specific meanings that do not exactly match the biomedical category of leprosy.
The study associated “gwasi” mainly with skin lesions and “kokobe” mainly with deformities or amputations, while also documenting overlap between the terms. Researchers should therefore preserve local terminology as an analytic category.
Can AI replace human researchers in culturally sensitive health studies?
+
AI cannot replace human researchers in culturally sensitive health studies because interpretation depends on language, history, power relationships, reflexivity, and local knowledge.
The PLOS study used a multidisciplinary team that included dermatologists, a medical anthropologist, and a Maroon cultural health mediator. That team structure demonstrates the importance of human expertise alongside systematic coding.
How should researchers protect interview transcripts about stigmatized diseases?
+
Researchers should protect stigmatized-disease transcripts through informed consent, pseudonymization, secure storage, controlled access, and careful decisions about quotation and data sharing.
The PLOS study withheld its transcripts from public access because contextual details could potentially identify participants in small communities, even after pseudonymization.
Conclusion and Next Steps for AI Qualitative Health Research
The November 2026 PLOS study shows that qualitative health research can reveal how language, cultural beliefs, stigma, treatment, and community support shape care pathways in ways that standardized measures may miss.
AI thematic analysis can make this work more manageable by organizing multilingual documents, surfacing recurring themes, comparing segments, and keeping findings connected to source excerpts.
Researchers should use AI as a transparent analysis aid, preserve cultural context, verify interpretations, and apply strict safeguards to sensitive participant data.
Teams that want to analyze interviews, open-ended survey responses, and research documents can try Evidano for free.
Topics
- AI thematic analysis
- AI qualitative analysis of health interviews
- qualitative health research
- cultural analysis of interview transcripts
- thematic analysis in public health
Keep reading
- Commentary on NewsWhat Hajj Interviews Reveal: Qualitative Sun Protection AnalysisUnderstand what qualitative sun protection analysis found among Hajj pilgrims and nurses, and learn how AI thematic analysis turns interview evidence into actionable research. Try Evidano.
- Commentary on NewsCulturally Aware AI Qualitative Analysis for LeprosyLearn how AI qualitative analysis can surface cultural beliefs, stigma, and care barriers in leprosy research, using a 2026 French Guiana study.
- Commentary on NewsSun Protection During Hajj: Qualitative ResearchLearn what qualitative research found about sun protection during Hajj, and how AI thematic analysis helps health teams turn interviews into practical action faster.