Researchers evaluating community health interventions need to understand more than whether an intervention worked. Researchers also need to identify which experiences, relationships, behaviours, and contextual conditions may explain perceived outcomes. AI qualitative analysis can help public health teams move from interview transcripts to structured themes, comparisons, and logic models more efficiently, while keeping human interpretation at the centre. The November 5, 2026 study in PLOS ONE provides a useful example. The ELLY intervention worked with 75 residents across two disadvantaged Scottish communities during a 12-week period from September 2023 to November 2023. Researchers interviewed 20 participants and used framework analysis to examine how the intervention was experienced. This post explains what the ELLY findings show, how qualitative teams can use AI responsibly to organize similar evidence, and where human judgement remains essential. The discussion is research-focused and non-diagnostic.
Key Takeaways
The November 5, 2026 PLOS ONE study found that the ELLY community intervention was viewed as feasible and acceptable by participants, with social connection and wellbeing appearing more motivating than weight loss alone. The study shows how AI qualitative analysis could help researchers organize complex interview evidence into themes, mechanisms, and cross-community comparisons, but the authors did not use AI for their analysis.
- According to the November 5, 2026 PLOS ONE study, ELLY was delivered to 75 residents across two disadvantaged Scottish communities over 12 weeks, from September 2023 to November 2023.
- According to the study, researchers interviewed 20 participants, with 10 participants from each community, and analysed the interviews using framework analysis.
- According to the study, the main qualitative themes included positive behaviour change, social connectedness, healthier dietary habits, perceived mental and physical health improvements, and new skills and knowledge.
- According to the study, 14 of 20 interview participants, or 70%, were female, while 17 of 20 participants, or 85%, lived in areas classified as disadvantaged by Scottish Index of Multiple Deprivation quintiles 1 to 3.
- According to the study, qualitative findings alone could not establish definitive causation, and longer-term mixed-methods research is needed to assess whether reported changes were sustained.
The source for these findings is PLOS ONE, in the article by Julie Cowie, Rhonda Archibald, Scott Findlay, Danielle Hutcheon, Sinéad Currie, Louise Hay, and Elizabeth Cook.
What the ELLY Qualitative Study Found
The ELLY study found that a co-designed, incentive-based community intervention was associated with several perceived pathways to healthier behaviour and wellbeing, although the qualitative evidence did not prove causation. The research was conducted in two disadvantaged communities in Forth Valley, Scotland, and evaluated a programme that included free soup twice weekly, community activity engagement, optional goal setting, information resources, and self-monitoring.
According to the November 5, 2026 PLOS ONE report, participants received a £25 shopping card after achieving at least 9 loyalty-card stamps across 12 weeks of engagement with local activities, groups, or clubs. The study recruited 35 participants in one community and 40 in the other, and all 20 people invited to qualitative interviews agreed to participate.
The researchers conducted approximately 30-minute, face-to-face semi-structured interviews at the end of the 12-week intervention. According to the study, two researchers developed an initial coding framework after reviewing five interviews, 10% of the data was independently cross-checked, and disagreements were resolved through discussion with a third researcher.
Participant accounts connected ELLY with improved awareness of local resources, more frequent activity, stronger social relationships, healthier eating, increased confidence, and reported improvements in mental and physical wellbeing. One participant described the change as, “I’ve changed big time for getting out of the house. I was stuck in the house all the time, with my depression and that. I’m out of the house every day.”
The study also documented limits and negative experiences. Some participants found groups unwelcoming, activities irregular, goals difficult to achieve, or outcome measures burdensome. One participant said, “I found some things very specifically orientated to different groups and I didn’t fit into those groups, ” showing why qualitative analysis must preserve disconfirming and minority perspectives.
ELLY Findings Snapshot for Qualitative Researchers
| Date or period | Metric | Value | Implication for AI qualitative analysis |
|---|---|---|---|
| September to November 2023 | Intervention duration | 12 weeks | Analysis should distinguish short-term perceptions from evidence of sustained change. |
| September to November 2023 | Residents receiving the intervention | 75 | A structured coding workflow can help connect participant experiences with intervention reach. |
| At 12 weeks in 2023 | Qualitative interview participants | 20, with 10 per community | Cross-segment analysis can compare how the same intervention operated in different places. |
| 2023 interview sample | Female participants | 14 of 20, or 70% | Participant composition should be reported when interpreting whose experiences are represented. |
| 2023 interview sample | Participants living in SIMD quintiles 1 to 3 | 17 of 20, or 85% | Segment labels can help researchers examine how context shapes intervention experience. |
| 2023 intervention design | Loyalty-card threshold | 9 activities over 12 weeks | Frequency and engagement analysis can test how participation relates to reported mechanisms. |
| 2023 intervention design | Shopping-card incentive | £25 | Codebooks can separate intervention components from perceived outcomes and mechanisms. |
Implications for Public Health and Community Researchers
What should researchers code in community health interviews?
Researchers should code intervention components, participant experiences, mechanisms, outcomes, barriers, and contextual differences as separate but connected categories. The November 5, 2026 PLOS ONE study distinguishes mechanisms such as peer recruitment and community connection from outcomes such as social connectedness and positive behaviour change.
A practical AI-assisted code structure could include parent codes for motivation, opportunity, capability, social connection, diet, physical activity, wellbeing, inclusion, communication, and unintended barriers. Researchers should review AI-generated codes against the interview data and the study protocol before accepting them.
How can AI improve qualitative analysis without replacing researchers?
AI can reduce repetitive work by organizing transcripts, finding recurring concepts, comparing segments, and retrieving supporting quotations, while researchers retain responsibility for interpretation and claims. The ELLY authors used a human-led process involving independent coding, team discussion, cross-checking, field notes, and reflexive review.
AI outputs should be treated as analytic aids rather than findings. The ELLY study shows why researchers must inspect negative cases, including participants who did not achieve goals or felt excluded from community groups.
Why do cross-community comparisons matter?
Cross-community comparison matters because the same intervention can be delivered and experienced differently in different local contexts. According to the PLOS ONE study, differences in community cohesion, partner involvement, local assets, communication, and implementation support affected how ELLY was received.
An AI-assisted cross-segment analysis can surface similarities and differences between locations, demographic groups, engagement levels, or intervention components. Researchers still need contextual knowledge to decide whether a difference reflects a meaningful pattern, a sampling effect, or a local implementation detail.
How Evidano Supports AI Qualitative Analysis of Health Interviews
Problem: Interview synthesis is slow
Evidano is an AI-powered qualitative data analysis platform that helps researchers analyze interviews, open-ended surveys, and research documents. Researchers can upload interview transcripts and use thematic, content, frequency, and cross-segment analyses to organize evidence from community health studies.
For an ELLY-style project, a researcher could use thematic analysis to identify wellbeing, social connection, diet, behaviour, and inclusion themes, then use frequency analysis to review how often concepts appear across the interview set. Learn more about Evidano’s qualitative research features.
Problem: Important quotations are difficult to retrieve
Evidano’s AI chat over documents and analyses can help researchers locate passages connected to specific themes, participant groups, or intervention components. Researchers can use retrieved quotations to support reporting, then verify each quotation against the original transcript and consent requirements.
The ELLY study demonstrates the value of preserving participant voice, including both positive accounts and critical comments about exclusion, irregular activities, and difficult goals.
Problem: Differences between groups remain hidden
Evidano’s cross-segment analysis can help researchers compare communities, demographic groups, engagement levels, or other research-defined segments. Researchers could compare the two ELLY communities to examine differences in social connection, access to local assets, communication, and perceived outcomes.
Evidano also supports visualizations such as hierarchical codes and subcodes, word clouds, and co-occurrence networks. These views can help research teams inspect relationships between intervention components, mechanisms, barriers, and outcomes before final interpretation.
Problem: Sensitive health data requires careful handling
Evidano encrypts data and states that customer data is never used to train third-party models. Researchers working with health interviews should still follow informed-consent terms, institutional governance, de-identification procedures, and applicable ethics requirements.
The ELLY study restricted public access to full interview transcripts because of legal and ethical requirements, illustrating why qualitative health data should be handled as sensitive research material.
FAQ: AI Qualitative Analysis for Community Health Research
What is AI qualitative analysis?
+
AI qualitative analysis uses artificial intelligence to help researchers organize, code, compare, summarize, and retrieve meaning from qualitative data such as interview transcripts and open-ended survey responses.
AI qualitative analysis does not by itself establish causation or replace researcher judgement. The November 5, 2026 PLOS ONE study shows why interpretation, reflexivity, and review of contradictory evidence remain necessary.
How was the ELLY study’s qualitative data analysed?
+
The ELLY researchers analysed semi-structured interviews using framework analysis supported by NVivo 12. According to the study, two researchers developed and refined a coding framework, 10% of the data was independently checked, and the wider team reviewed emerging themes.
The researchers also used field notes, weekly meetings, and a COM-B-informed logic model to connect intervention components with perceived mechanisms and outcomes.
How many people were interviewed in the ELLY study?
+
The ELLY study interviewed 20 participants at 12 weeks, with 10 participants from each of two disadvantaged Scottish communities. The 20 interview participants represented 27% of the 75 residents who received the intervention.
According to the study, the interview sample included people with different ages, employment statuses, household compositions, BMI levels, and degrees of engagement.
What were the main themes in the ELLY interviews?
+
The main themes were peer recruitment, connection with community assets, positive behaviour change, social connectedness, healthier diet, mental and physical health improvements, and new skills and knowledge.
The study also identified barriers involving group inclusivity, irregular activities, communication, digital access, and the burden or intrusiveness of some outcome measures.
Can AI prove that a community intervention caused better health outcomes?
+
AI cannot prove that a community intervention caused better health outcomes from qualitative interviews alone. The ELLY authors explicitly stated that the qualitative data provided insight into perceived pathways but was insufficient to establish definitive causation.
Researchers need appropriate longitudinal, mixed-methods, effectiveness, or cost-effectiveness designs to test causal and sustained outcomes.
Conclusion and Next Steps
The ELLY study shows why community health researchers need qualitative methods that capture mechanisms, context, positive outcomes, and exclusionary experiences together. The study involved 75 residents over 12 weeks and used 20 interviews to identify how social connection, wellbeing, local activities, and healthier routines were perceived to interact.
AI qualitative analysis can make this type of evidence easier to organize and compare, but responsible research still requires transparent coding decisions, human review, attention to minority cases, and respect for sensitive participant data. Evidano helps research teams move from transcripts to thematic and cross-segment insight while keeping the original evidence available for verification. Try Evidano for free and explore a faster workflow for qualitative health research.
Topics
- AI qualitative analysis
- AI thematic analysis
- qualitative analysis of health interviews
- community health research
- qualitative interview analysis
Keep reading
- Commentary on NewsWhat ELLY Reveals About AI Qualitative AnalysisSee how AI qualitative analysis can turn community-health interviews into defensible themes, pathways, and decisions, using findings from the 2026 ELLY study.
- Commentary on NewsAI Qualitative Analysis: Lessons from ELLYLearn how AI qualitative analysis can uncover pathways in community health research, using the 2026 ELLY study to improve coding, synthesis, and decisions.
- Commentary on NewsAI Qualitative Analysis: Lessons from ELLYLearn how AI qualitative analysis can reveal intervention pathways, participant needs, and community outcomes from the 2026 ELLY health study.