Evidano is an AI-powered qualitative data analysis platform that helps researchers analyze interviews, open-ended surveys, and documents. For public health researchers, the central challenge is not simply collecting interview data, but identifying how language, cultural beliefs, stigma, and care experiences shape health behavior. The November 2026 study by Aurore Dupont and colleagues in French Guiana demonstrates why qualitative analysis of leprosy perceptions requires attention to local terminology and explanatory models. This article explains the study's main findings, shows how AI-assisted thematic analysis can support culturally responsive research, and identifies safeguards for working with sensitive clinical narratives.
Key Takeaways
The PLoS Neglected Tropical Diseases study found that leprosy perceptions in French Guiana are shaped by local terms, spiritual and familial explanations, stigma, and partial biomedical knowledge. The study shows that qualitative analysis can reveal why health communication must be culturally specific rather than relying only on standardized disease terminology.
- According to the study published on November 11, 2026, researchers interviewed 20 patients and 37 community members in French Guiana between April and September 2025.
- According to Dupont and colleagues, Maroon communities commonly used the terms “kokobe” for deformities and “gwasi” for skin patches associated with leprosy.
- According to the study, only 2 patients and 1 western French Guiana community member mentioned bacteria as a cause of leprosy.
- According to the study, 19 patient interviews and 18 community interviews were conducted, with interviews often requiring linguistic or cultural mediation.
- According to the authors, plant-based remedies and spiritual practices were common, but local remedies did not appear to reduce adherence to multidrug therapy in this sample.
What the 2026 Study Reveals About Leprosy Perceptions
The 2026 French Guiana study found that people described leprosy through several overlapping explanatory models rather than through a single biomedical framework. The researchers used semi-structured interviews, purposive sampling for patients, snowball recruitment for western French Guiana community members, and deductive and inductive thematic analysis.
Between April and September 2025, Dupont, Saakie, Tareau, and Blaizot interviewed 20 patients across French Guiana health centres and 37 community members in villages connected to the former Acarouany leprosarium. Recruitment continued until thematic saturation, meaning that subsequent interviews produced no substantially new themes or explanatory models.
The study identified a major terminology gap. Maroon participants often used “kokobe” and “gwasi, ” while some patients did not recognize French biomedical terms such as leprosy or Hansen's disease. One community participant explained, “It starts with spots... If you don't treat it, your fingers can bend.”
The study also identified persistent gaps in causal understanding. Spiritual explanations, inherited taboos, curses, food, animals, housing, soil, and insects appeared in participant narratives, while person-to-person transmission was rarely understood clearly. One participant summarized the uncertainty by saying, “We don't know how it is acquired.”
The authors reported that participants generally considered multidrug therapy effective and that family or community support was common, although some patients described discrimination, delayed diagnosis, transport barriers, income loss, and fear of disclosure.
The study is qualitative and culturally specific, so its findings illustrate a diversity of perspectives rather than measuring the prevalence of beliefs across every cultural group in French Guiana. The authors also noted that the sensitive interview transcripts cannot be publicly shared because participants could potentially be identified.
Findings Snapshot: AI Qualitative Analysis of Leprosy Perceptions
| Date or study measure | Finding | Implication for qualitative researchers |
|---|---|---|
| April to September 2025, according to Dupont et al. | 20 patients and 37 community members participated | AI-assisted analysis can organize multiple participant groups while preserving segment-level context. |
| November 11, 2026, according to PLOS | The study was published in PLoS Neglected Tropical Diseases | Researchers can connect coded findings to a named, peer-reviewed source and publication date. |
| Patient interviews, according to Dupont et al. | 19 patient interviews were conducted, including individual and group interviews | Mixed interview formats require careful comparison of quotations, speakers, and contexts. |
| Community interviews, according to Dupont et al. | 18 community interviews were conducted, including 10 individual and 8 group interviews | Cross-segment analysis can distinguish patient experiences from collective community memory. |
| Language mediation, according to Dupont et al. | 12 patient interviews involved mediation or a relative, and 17 community interviews required mediation | Translation and mediation notes should be retained as part of the analytic context. |
| Causal explanations, according to Dupont et al. | Only 2 patients and 1 community member referred to bacteria as a cause | Communication research should identify knowledge gaps without treating cultural explanations as simple errors. |
Implications for Public Health and Health Communication Researchers
Why should health researchers code local terminology separately from biomedical terminology?
Health researchers should code local terminology separately because local words may describe symptoms, disability, social identity, or cultural categories rather than the biomedical disease itself. Dupont and colleagues found that “kokobe” and “gwasi” did not correspond exactly to the biomedical category of leprosy.
A qualitative coding framework can therefore distinguish disease names, visible symptoms, perceived causes, transmission beliefs, stigma, and treatment expectations. Separating these codes helps researchers design communication that acknowledges local meaning without reinforcing misinformation.
How can thematic analysis improve culturally sensitive health communication?
Thematic analysis can improve health communication by showing which concepts participants use to explain illness, treatment, contagion, and recovery. The French Guiana study found that spiritual explanations and family-based beliefs coexisted with confidence in biomedical treatment.
A communication strategy based on these findings could address person-to-person transmission, explain treatment effectiveness, and discuss local practices respectfully. The study's authors recommended culturally sensitive communication that takes local terminology and practices into account.
What should researchers examine beyond the interview transcript?
Researchers should examine language, interpretation, participant group, interview setting, recruitment pathway, and the relationship between interviewer and participant. The study reported that many interviews required cultural mediation and that the principal investigator also provided medical care to patients.
These contextual variables can affect what participants disclose and how their statements are interpreted. AI-assisted qualitative analysis should support, not replace, reflexive review by researchers familiar with the population and research setting.
How Evidano Helps With Sensitive Qualitative Health Research
Problem: Manual thematic coding is slow across multilingual interviews
Evidano is an AI-powered qualitative data analysis platform that helps researchers analyze interview transcripts, reports, papers, and survey responses while preserving themes and participant context.
Researchers can use thematic, content, frequency, and cross-segment analyses to organize findings such as local terminology, perceived causes, treatment beliefs, stigma, and care barriers. The Evidano features overview explains how the platform supports document-based qualitative analysis.
Problem: Translation can obscure culturally specific meanings
Evidano's translation workflow supports multilingual qualitative projects and custom dictionaries, which can help researchers preserve recurring local terms such as “kokobe” and “gwasi” during analysis.
Researchers should still validate translations and interpretations with cultural mediators or subject-matter experts. AI can identify patterns across languages, but contextual meaning remains a human research responsibility.
Problem: Researchers need to compare patient and community perspectives
Evidano can support cross-segment analysis by comparing patient interviews, community interviews, field notes, and related reports. Researchers can use hierarchical codes and subcodes to separate biomedical knowledge, spiritual explanations, stigma, treatment adherence, and social support.
Researchers can also use AI chat over uploaded documents and analyses to ask focused questions about differences between participant groups, while reviewing the underlying evidence before reporting a conclusion.
Problem: Sensitive health narratives require strong data governance
Sensitive qualitative health data should be handled with strict confidentiality, limited access, and careful decisions about quotation and publication. The French Guiana study withheld its interview transcripts because contextual details could identify participants.
Evidano states that data are encrypted and never used to train third-party models. Researchers should also follow their institutional ethics approvals, consent terms, applicable privacy laws, and local data-management plans. Information here is for research purposes and is not medical or diagnostic advice.
FAQ: AI Qualitative Analysis of Leprosy Perceptions
What did the 2026 French Guiana study find about leprosy perceptions?
+
The 2026 French Guiana study found that leprosy perceptions combined local terminology, spiritual and familial explanations, partial biomedical knowledge, treatment confidence, and experiences of stigma. Dupont and colleagues interviewed 20 patients and 37 community members between April and September 2025.
What do “kokobe” and “gwasi” mean in the study?
+
In the study, “gwasi” was mainly associated with skin lesions and “kokobe” was mainly associated with deformities or amputations. Dupont and colleagues emphasized that the terms reflect local classifications and do not correspond exactly to the biomedical category of leprosy.
How can AI analyze qualitative health interviews responsibly?
+
AI can responsibly support qualitative health research by organizing codes, identifying recurring themes, comparing participant segments, and surfacing quotations for human review. Researchers must protect confidentiality, validate translations, document analytic decisions, and avoid treating automated outputs as clinical or cultural truth.
Can AI replace human researchers in culturally sensitive analysis?
+
AI cannot replace human researchers in culturally sensitive analysis because interpretation depends on language, history, power relationships, and local context. The French Guiana study used a multidisciplinary team that included dermatologists, a medical anthropologist, and a Maroon cultural health mediator, illustrating the value of contextual expertise.
What are the main limitations of the French Guiana leprosy study?
+
The main limitations were the small number of participants within each cultural group, the focus of community interviews on Maroon settlements, and the possibility that the investigator's clinical role influenced participant disclosure. The authors stated that the findings illustrate diverse perspectives rather than providing exhaustive or statistically generalizable group profiles.
Conclusion and Next Steps for Qualitative Health Research
The November 2026 PLOS study shows that qualitative analysis of leprosy perceptions can reveal communication barriers that epidemiological measures alone may miss. Local terms, spiritual explanations, treatment beliefs, stigma, language mediation, and transport barriers all shape how people understand illness and move through care.
AI-assisted thematic analysis can help public health teams process multilingual interviews, compare patient and community perspectives, and trace themes across sensitive documents. Human researchers remain responsible for cultural interpretation, ethical safeguards, participant confidentiality, and final conclusions.
Researchers working with interviews, reports, and open-ended survey data can try Evidano for free to accelerate qualitative analysis while keeping evidence review and research judgment at the center.
Topics
- AI qualitative analysis of leprosy perceptions
- qualitative analysis in public health
- AI thematic analysis of interviews
- culturally sensitive health research
- leprosy stigma research
Keep reading
- Commentary on NewsCultural Insight: AI Qualitative Analysis of LeprosyLearn how AI qualitative analysis of leprosy interviews reveals cultural beliefs, stigma, and care barriers, helping public health researchers act faster.
- Commentary on NewsQualitative Analysis of Leprosy PerceptionsLearn how AI-enabled qualitative analysis of leprosy perceptions reveals cultural beliefs, stigma, and care pathways in French Guiana, with practical research insights.
- Commentary on NewsWhat Interviews Reveal: Leprosy PerceptionsLearn how qualitative analysis of leprosy perceptions exposes cultural beliefs, stigma, and care barriers, and how AI helps researchers synthesize interviews.