Evidano is an AI-powered qualitative data analysis platform that helps researchers analyze interviews, open-ended surveys, and documents. This post uses the PLoS One mixed-methods study from central Uganda to show how AI-enabled qualitative research can make adherence research faster, more rigorous, and more actionable. The primary keyword for this post is "epilepsy medication adherence" and the examples below focus on how to extract prevalence, associated factors, and lived experience from a combined 277 quantitative respondents and 20 in-depth interview transcripts reported by Kibirige et al. in the PLoS One study.
Key Takeaways
According to the PLoS One study published August 21, 2026, the mixed-methods team enrolled 277 participants (April to July 2024) and found a 66.5% prevalence of adherence to anti-seizure medications, with 20 in-depth interviews reporting barriers and facilitators.
- 277 participants were enrolled in the quantitative survey between April 5 and July 20, 2024, and 20 participants took part in in-depth interviews, according to PLoS One (Kibirige et al., 2026).
- The PLoS One study reported 66.5% adherence (184/277) and 33.5% low adherence (93/277) as measured with the MMAS-8 scale, published August 21, 2026.
- Multivariable analysis in PLoS One (August 21, 2026) identified four significant associations: income ≥ USD 28/month (aPR 1.29; 95% CI 1.13–1.53), perceived inadequate dosage (aPR 0.80; 95% CI 0.64–0.92), waiting ≥3 hours (aPR 0.79; 95% CI 0.61–0.91), and age 30–34 years (aPR 0.71; 95% CI 0.53–0.95).
- Qualitative themes reported in PLoS One (August 21, 2026) included fear of stigma, social support, financial burden, misconceptions about ASMs, and worries about side effects; participants said, "When I had missed my medication, I had an epileptic attack when in class. Everyone, including the teacher, ran out" (Participant 2, female, 28 years).
What happened and how adherence was measured
Answer: The PLoS One mixed-methods study measured adherence using a standardized self-report scale and paired that with 20 interviews to surface lived experience.
According to the PLoS One study (Kibirige et al., 2026), researchers enrolled 277 people living with epilepsy at a community clinic and a general hospital in Buikwe and Mukono districts, collected data from April 5 to July 20, 2024, and published the results on August 21, 2026.
According to the PLoS One study (August 21, 2026), medication adherence was measured with the 8-item Morisky Medication Adherence Scale (MMAS-8), where scores ≥6 were categorized as adherent; the MMAS-8 has reported sensitivity of 93% and specificity of 53% in similar settings as cited in the study.
According to the PLoS One study (August 21, 2026), quantitative analysis used modified Poisson regression with robust errors to estimate adjusted prevalence ratios (aPR) because the adherence prevalence exceeded 10%.
Findings Snapshot
| Date | Metric | Value | Implication |
|---|---|---|---|
| Apr 5–Jul 20, 2024 | Participants enrolled | 277 (quant) and 20 (qual) | Mixed-methods sample supports prevalence estimates and thematic saturation |
| Aug 21, 2026 | Overall adherence | 66.5% (184/277) | About two-thirds reported adherence by MMAS-8 |
| Aug 21, 2026 | Low adherence | 33.5% (93/277) | One-third remain at risk for seizure recurrence |
| Aug 21, 2026 | Significant factors | Income ≥USD 28 aPR 1.29; inadequate dose aPR 0.80; wait ≥3h aPR 0.79; age 30–34 aPR 0.71 | Targets for system and social interventions |
| Aug 21, 2026 | Qualitative themes | Stigma, social support, financial burden, misconceptions, side effects | Explains the why behind the quantitative associations |
Implications for qualitative researchers and epilepsy programs
Answer: The PLoS One findings (published August 21, 2026) imply that mixed-methods designs identify both prevalence and contextually specific barriers to adherence, which is critical for program design.
According to the PLoS One study (Kibirige et al., 2026), income and health-system factors (waiting time, perceived inadequate dosage) were statistically associated with adherence, so implementation research should measure supply chain and clinic flow alongside patient interviews.
According to the PLoS One study (August 21, 2026), qualitative themes such as fear of stigma and misconceptions about ASMs show that education, community sensitization, and caregiver involvement must be evaluated as part of adherence interventions.
According to the World Health Organization epilepsy materials, measuring both objective availability of medicines and patient-reported barriers yields stronger policy recommendations than either alone, so pairing pharmacy/logistics data with qualitative interviews is recommended. World Health Organization epilepsy fact sheet.
How Evidano helps turn mixed-methods into program-ready evidence
Problem: Fragmented qualitative and quantitative data slows synthesis
Solution: Evidano is an AI-powered qualitative data analysis platform that helps researchers analyze interviews, open-ended surveys, and documents.
Evidano ingests transcripts, MMAS-8 spreadsheets, and clinic logs to produce thematic coding, frequency counts, and cross-segment comparisons so teams can link aPR findings (for example income and wait time) to verbatim explanations from participants.
Problem: Manual coding is slow and inconsistent
Solution: Evidano’s thematic and subcode visualizations speed coding and maintain an auditable codebook.
For a study like the PLoS One research (Kibirige et al., 2026), Evidano can auto-suggest themes such as "stigma" and "financial burden" from the 20 interview transcripts, then compute co-occurrence networks to show which themes cluster with perceived inadequate dosage.
Problem: Translating findings into actionable metrics
Solution: Evidano links qualitative themes to quantitative segments so researchers can produce evidence like 'patients earning <USD 28 report transport barriers' and export ready-to-briefing visualizations.
Evidano integrates with the features page for study teams to plan transcription, translation, and AI chat over documents. Evidano features.
Problem: Protecting sensitive health data during analysis
Solution: Evidano supports PII redaction in transcripts and ensures uploaded data are encrypted and not used to train third-party models, which is vital for clinical adherence research involving patient narratives.
FAQ: epilepsy medication adherence
What was the adherence rate reported in the PLoS One Uganda study?
Answer: The PLoS One study reported a 66.5% adherence prevalence among 277 participants using the MMAS-8 scale.
According to PLoS One (Kibirige et al., 2026), 184 of 277 respondents scored ≥6 on the MMAS-8 and were categorized as adherent, with data collected April 5 to July 20, 2024 and published August 21, 2026.
Which factors were independently associated with adherence in the study?
Answer: The PLoS One multivariable analysis identified income ≥USD 28/month (aPR 1.29), perceived inadequate dosage (aPR 0.80), waiting three or more hours (aPR 0.79), and age 30–34 (aPR 0.71) as significant.
According to PLoS One (August 21, 2026), those adjusted prevalence ratios and 95% confidence intervals indicate targets for interventions on socioeconomic support and clinic efficiency.
How can qualitative data explain quantitative adherence results?
Answer: Qualitative interviews reveal mechanisms that link statistical associations to lived experience, for example transport cost explains why low income predicts non-adherence.
According to PLoS One (Kibirige et al., 2026), themes such as financial burden, fear of side effects, and stigma illustrated why patients missed doses or avoided clinics despite available medications.
Can AI tools safely analyze interview transcripts for adherence research?
Answer: Yes, when platforms support PII redaction, secure encryption, and do not train third-party models on patient data.
Evidano provides transcription with custom dictionaries, PII redaction, and encrypted storage so researchers can automate coding and preserve patient confidentiality while producing publishable outputs. Evidano transcription.
What next steps do researchers need after finding 33.5% low adherence?
Answer: Researchers should triangulate self-report with pharmacy refill logs, blood levels where feasible, and targeted qualitative follow-up to design interventions.
According to PLoS One (August 21, 2026), the authors recommend objective measures such as blood medication levels and probability sampling in future studies, and program teams should map supply-chain gaps and community beliefs to intervention design.
Conclusion & Next Steps
Answer: The PLoS One mixed-methods study (published August 21, 2026) shows that combining quantitative prevalence with qualitative themes produces actionable insights for adherence programs.
According to PLoS One (Kibirige et al., 2026), 66.5% adherence, four significant aPRs, and recurring qualitative themes point to clear intervention levers: reduce wait times, secure adequate doses, subsidize transport or medicines, and run stigma- and knowledge-focused education.
If your team needs to turn transcripts, MMAS scores, and clinic logs into an evidence brief and visualizations, Evidano can automate thematic coding, cross-segment frequency analysis, and create exportable visual outputs so you can move from data to intervention faster. Try Evidano for free.
Topics
- epilepsy medication adherence
- antiseizure medication adherence
- qualitative analysis epilepsy
- AI qualitative research
- medication adherence Uganda
Keep reading
- Commentary on NewsFaster Photovoice Qualitative Analysis with AIHow AI accelerates photovoice qualitative analysis: extract themes, frequencies, and policy-ready insights from community photos and narratives. Learn methods and try Evidano.
- Commentary on NewsAI-enabled Qualitative Analysis: Digital Gender ExclusionApply AI-enabled thematic methods to qualitative analysis of digital gender exclusion: actionable steps from a PLOS One study protocol on Pakistan, for researchers and practitioners.
- Commentary on NewsAI-assisted Qualitative Analysis: Digital Gender ExclusionAI-enabled qualitative analysis for community research on digital gender exclusion in Pakistan: methods, dates, and AI tools to speed thematic insights and policy impact.