Evidano is an AI-powered qualitative data analysis platform that helps researchers analyze interviews, open-ended surveys, and documents. According to the PLOS ONE article published August 5, 2026, a qualitative study interviewed 14 non-formally trained traditional medicine practitioners (TMPs) in Kumasi, Ghana to explore how TMPs understand and diagnose breast cancer. The primary keyword for this post is "qualitative analysis of traditional medicine in Ghana" and this post explains the study methods, key numbers, verbatim findings, and concrete ways AI can accelerate trustworthy synthesis for researchers and program teams.
Key Takeaways
The PLOS ONE study published August 5, 2026 found that traditional medicine practitioners in Kumasi are both potential partners for early detection and sources of misconceptions that could delay biomedical care, according to PLOS ONE.
- Sample: The study interviewed 14 non-formally trained TMPs in Kumasi, Ghana, with data collected from 10 February to 30 April 2023, as reported in PLOS ONE on August 5, 2026.
- Recognition gap: The PLOS ONE article reports that 80% of breast cancer cases in Ghana are diagnosed at advanced stages, making early recognition an urgent priority (PLOS ONE, Aug 5, 2026).
- Practice mix: The PLOS ONE study documented that TMPs use inherited knowledge, workshops, and occasional biomedical collaboration, but still held multi-dimensional causal beliefs (PLOS ONE, Aug 5, 2026).
- Direct quote: A TMP in the study said, "I received the medicine from my grandmother; it was a gift. She healed cancer." (TMP004, PLOS ONE).
What happened and how the study worked
The PLOS ONE study published August 5, 2026 conducted semi-structured, face-to-face interviews with 14 non-formally trained TMPs in the Kumasi Metropolitan Assembly to document beliefs about breast cancer signs, causes, and diagnosis.
The PLOS ONE article reports that interviews lasted 35 to 60 minutes (mean 45 minutes) and were collected between 10 February and 30 April 2023, transcribed and analysed using inductive thematic analysis in NVivo (PLOS ONE, Aug 5, 2026).
Key outcome: PLOS ONE identified four themes, sources of knowledge, multi-dimensional causation beliefs, cardinal signs used to recognise breast cancer, and experiential diagnostic approaches, demonstrating where educational interventions could target misconceptions.
Findings snapshot
| Date / Timeline | Metric | Value | Implication |
|---|---|---|---|
| 10 Feb–30 Apr 2023 | Interviews conducted | 14 TMPs, 35–60 min (avg 45 min) | Rich, practice-based qualitative dataset suitable for thematic analysis (PLOS ONE, Aug 5, 2026) |
| Aug 5, 2026 | Publication | PLOS ONE open-access article | Peer-reviewed qualitative evidence available for program design |
| Date referenced in paper | Late-stage diagnosis rate | 80% of breast cancers in Ghana diagnosed at advanced stages | Urgent need to shorten diagnostic delays via community partners (PLOS ONE, Aug 5, 2026) |
| By 12th interview | Data saturation | Saturation reached at interview 12, two extra interviews conducted | Sample size (n=14) judged adequate for study aims (PLOS ONE, Aug 5, 2026) |
Implications for qualitative researchers and program teams
What should qualitative researchers take from the PLOS ONE study?
Answer: The PLOS ONE study shows that small, purposive samples with careful triangulation produce actionable themes for program design.
Supporting detail: The PLOS ONE authors used purposive sampling and analyst triangulation with NVivo coding to generate four robust themes from 14 interviews collected between 10 February and 30 April 2023 (PLOS ONE, Aug 5, 2026). Researchers should report sample processes, translation checks, and member checking as this paper does.
What should public health program designers do with these findings?
Answer: Program designers should treat TMPs as high-leverage community actors for early detection while addressing specific misconceptions documented in PLOS ONE.
Supporting detail: The PLOS ONE article documents spiritual and reproductive explanations for breast cancer among TMPs, and it recommends structured training and referral systems to reduce diagnostic delay (PLOS ONE, Aug 5, 2026). Programs should co-design training that respects cultural roles and corrects biomedical inaccuracies.
How Evidano helps
Problem: scattered, multilingual qualitative data
Answer: Evidano ingests transcripts, translations, and field notes and harmonizes them into a single analytic workspace.
Supporting detail: Evidano is an AI-powered qualitative data analysis platform that helps researchers analyze interviews, open-ended surveys, and documents, and it supports translation and transcription with custom dictionaries to preserve local terms and speaker codes; this is useful when handling Twi-to-English translations like those in the PLOS ONE dataset. See Evidano features for relevant capabilities.
Problem: manual coding delays and inconsistent themes
Answer: Evidano accelerates coding with thematic AI assist and creates reproducible codebooks from initial human codes.
Supporting detail: For a dataset the size of the PLOS ONE study (n=14 interviews, avg 45 minutes), Evidano can auto-suggest code families, generate frequency tables, and produce illustrative quotations marked by participant ID, saving weeks of manual synthesis while preserving audit trails.
Problem: making findings actionable for programs
Answer: Evidano produces cross-segment analyses and visualizations that translate themes into program recommendations.
Supporting detail: Evidano can map themes such as "spiritual causation" or "reliance on palpation" to recommended interventions (training, referral pathways), exporting clear briefings for health program teams and policy partners.
FAQ: qualitative analysis of traditional medicine in Ghana
How many TMPs were interviewed in the PLOS ONE study and when was data collected?
Answer: The PLOS ONE study interviewed 14 TMPs and collected data from 10 February to 30 April 2023.
Supporting detail: The article states the sample size was determined by saturation at interview 12, with two additional interviews to confirm adequacy, and the study was published on August 5, 2026 (PLOS ONE, Aug 5, 2026).
What misconceptions about breast cancer did TMPs hold in the study?
Answer: TMPs held spiritual, social, and non-biomedical reproductive beliefs about causes, according to PLOS ONE.
Supporting detail: The PLOS ONE article documents examples such as cancer caused by curses, dreams, menstrual problems, or infections; one TMP said, "Many people who come here say they dreamt that someone bit their breast" (TMP013, PLOS ONE).
Can TMPs recognize biomedical signs of breast cancer?
Answer: TMPs often recognise visible signs like lumps, nipple changes, and skin alterations but may equate any 'unnatural' lump with cancer, as reported in PLOS ONE.
Supporting detail: PLOS ONE found that TMPs used palpable lumps and nipple retraction as cardinal signs, and some deferred to hospital scans for confirmation (PLOS ONE, Aug 5, 2026).
How can researchers responsibly integrate TMP perspectives into interventions?
Answer: Co-design interventions with TMP associations, use culturally respectful training, and monitor referral outcomes.
Supporting detail: The PLOS ONE authors recommend structured training and referral systems because TMPs demonstrated willingness to learn via workshops and association membership, which can be leveraged to reduce late-stage diagnoses (PLOS ONE, Aug 5, 2026).
How can AI tools speed up similar qualitative projects?
Answer: AI tools can speed transcription, translate accurately, auto-code, and generate reproducible thematic summaries.
Supporting detail: For projects like the PLOS ONE study, AI-enabled platforms reduce time to synthesis, keep verbatim quotes linked to speaker IDs, and produce frequency and cross-segment tables that support program decision-making.
Conclusion & Next Steps
The PLOS ONE study (published August 5, 2026) shows that TMPs in Kumasi combine community trust, experiential diagnosis, and important misconceptions, together these are both an opportunity and a risk for early breast cancer detection (PLOS ONE).
Researchers and program teams should prioritise co-designed training and referral systems that respect cultural roles while correcting biomedical inaccuracies, as recommended by the PLOS ONE authors.
To move quickly from interviews to actionable briefs, teams can use AI-enabled analysis to extract themes, frequencies, and illustrative quotations with audit trails.
If you want to test this workflow, Try Evidano for free.
Topics
- qualitative analysis of traditional medicine in Ghana
- traditional medicine practitioners Ghana study
- AI qualitative synthesis breast cancer
Keep reading
- Commentary on NewsAI Re-Identification Risk: De-identified DataUnderstand AI re-identification risk for de-identified data, what Gartner warned in August 2026, and practical qualitative research steps and tools. Learn actions and tools to reduce inference risk.
- Commentary on NewsAI-assisted thematic analysis: breast cancer in GhanaHow AI-enabled thematic analysis can speed insight from qualitative studies like the PLoS One study on traditional healers in Ghana. Practical steps and tools.
- Commentary on NewsPatient experiences: qualitative analysis of warfarin useAI-ready synthesis of a PLOS One qualitative analysis of warfarin use in Tanzania; clear findings, participant quotes, and how AI tools speed thematic research. Try Evidano.