Researchers evaluating community health interventions need to understand not only whether participants changed, but how and why change was experienced. AI qualitative analysis can help public health teams organize interview transcripts, compare participant groups, and trace pathways from intervention activities to reported outcomes. The 2026 ELLY study provides a useful example because researchers used semi-structured interviews and framework analysis to examine a co-designed healthy weight and wellbeing intervention in two disadvantaged Scottish communities. This article explains the study’s evidence, shows how qualitative analysis surfaced interconnected mechanisms such as social connection and improved routines, and translates those lessons into a practical workflow for researchers. The article does not treat qualitative evidence as proof of causation. Instead, it shows how carefully governed AI can accelerate synthesis while researchers retain responsibility for interpretation, reflexivity, ethics, and contextual judgment.
Key Takeaways
The PLOS One study found that qualitative interviews helped explain how the ELLY community intervention was perceived to support wellbeing, social connectedness, healthier routines, and new skills in two disadvantaged Scottish communities.
AI qualitative analysis can help researchers move from raw interview transcripts to themes, subthemes, cross-segment comparisons, and pathway models, but the ELLY study shows that human review remains necessary because qualitative findings alone cannot establish definitive causation.
- Between September and November 2023, the ELLY intervention operated for 12 weeks and reached 75 residents across two Scottish communities, according to Cowie and colleagues’ 2026 PLOS One study.
- In 2023, researchers interviewed 20 participants, 10 from each community, using approximately 30-minute face-to-face semi-structured interviews, according to the ELLY study.
- The ELLY intervention offered free soup twice weekly, community-activity engagement, optional goal setting, information resources, and self-monitoring, according to the 2026 study.
- The ELLY researchers identified reported pathways involving behaviour change, social connection, dietary habits, physical and mental wellbeing, and new knowledge, according to their framework analysis.
- The ELLY authors concluded that wellbeing and social connection appeared more closely associated with positive behaviour change than healthy weight alone, while also stating that longer-term research is required.
What the ELLY Study Shows About AI Qualitative Analysis
The ELLY study shows that qualitative health research is most valuable when analysis connects participant experience to intervention mechanisms and reported outcomes. The researchers worked with two disadvantaged communities in Forth Valley, Scotland, and designed an intervention around local assets, incentives, food access, goal setting, and wellbeing rather than a narrowly weight-focused model.
The ELLY researchers recruited 35 adults in community C1 and 40 adults in community C2 during the 2023 feasibility study. The qualitative sample included 20 interview participants, whose mean age was 51 years, while 14 of 20 participants, or 70%, were female and 17 of 20 participants, or 85%, lived in areas classified as disadvantaged by Scottish Index of Multiple Deprivation quintiles 1 to 3.
The researchers developed an initial coding framework after reviewing five interviews, refined the framework through team discussion, and independently cross-checked 10% of the data. The ELLY team used NVivo 12, weekly meetings, field notes, and researcher discussion to support reflexivity and interpretation.
The study reported that participants perceived several connected pathways. Community activities could increase social contact, social contact could support confidence and routine, and those changes could encourage healthier food choices or physical activity. The researchers explicitly cautioned that the qualitative evidence could not prove clear causality across every pathway.
One participant described the behavioural effect in concrete terms: “Now I walk every night…I continue in this situation, like walking, and soup, and healthy food, that has helped me a lot.” The ELLY study attributed this quote to participant C2, P34.
A second participant described the social mechanism as follows: “It’s actually got me going out more, meeting other people, having a laugh, which is important.” The ELLY study attributed this quote to participant C1, P36.
ELLY Findings Snapshot for Qualitative Health Researchers
| Date or study detail | Metric | Value | Implication for analysis |
|---|---|---|---|
| September to November 2023 | Intervention duration | 12 weeks | Time-bounded interviews can capture immediate and intermediate perceived pathways, but not long-term sustainability. |
| 2023 feasibility study | Residents receiving the intervention | 75 | A qualitative workflow should preserve community and participant context alongside thematic codes. |
| 2023 interviews | Interview participants | 20, with 10 per community | Cross-community comparison can reveal shared mechanisms and local differences. |
| 2023 interview sample | Female participants | 14 of 20, or 70% | Segment analysis can show whose experiences are represented in the interview evidence. |
| 2023 interview sample | Participants living in disadvantaged areas | 17 of 20, or 85% | Demographic and contextual metadata can help researchers examine inclusion and transferability. |
| 2023 intervention design | Minimum activity threshold for the shopping card | 9 activities across 12 weeks | Coding can distinguish intervention components, engagement mechanisms, and reported outcomes. |
| 2026 publication | Independent cross-check of interview data | 10% | AI-assisted analysis should be paired with audit trails, researcher review, and disagreement resolution. |
Implications for Public Health and Community Researchers
Why should researchers analyze mechanisms, not just outcomes?
Researchers should analyze mechanisms because the ELLY study found that participants described interconnected routes from intervention components to behaviour, wellbeing, and social outcomes. A simple frequency count of comments about weight would have missed the reported importance of confidence, belonging, local activities, healthier food, and routine.
AI qualitative analysis can support this mechanism-focused approach by grouping excerpts into intervention inputs, participant experiences, enabling conditions, barriers, and reported outcomes. Researchers should validate each relationship against the original transcript and avoid presenting a model as causal when the source only reports participant perceptions.
How can cross-segment analysis improve community intervention research?
Cross-segment analysis can help researchers compare how the same intervention operates across communities, demographic groups, or levels of engagement. The ELLY researchers reported broadly comparable findings across two communities, but they also described differences in community cohesion, asset participation, communication, and local implementation.
A practical analysis plan should compare themes by community, engagement level, participant characteristics, and reported barriers. The ELLY study also noted that nine participants dropped out and were not included in interviews, so researchers should examine missing or dropout perspectives where data access and ethics allow.
What limitations should researchers preserve in an AI-assisted synthesis?
Researchers should preserve limitations because the ELLY authors stated that interview data alone were insufficient to establish definitive causation and that longer-term research was needed. The ELLY study covered a 12-week intervention and did not interview participants who dropped out.
AI-generated summaries should therefore distinguish direct participant accounts, researcher interpretations, and hypotheses for future testing. Health research teams should also protect sensitive transcripts and use non-diagnostic language when analyzing mental or physical wellbeing.
How Evidano Supports AI Qualitative Analysis of Health Interviews
Problem: Interview transcripts are difficult to code consistently
Evidano is an AI-powered qualitative data analysis platform that helps researchers analyze interviews, open-ended survey responses, and documents.
Evidano can organize interview material into thematic, content, and frequency analyses, helping teams identify themes such as social connectedness, dietary change, confidence, barriers, and new skills. Researchers can review the supporting excerpts before incorporating findings into a report. Learn more about Evidano’s qualitative research features.
Problem: Researchers need to compare communities and participant groups
Evidano supports cross-segment analysis, allowing researchers to compare themes across communities, demographic groups, intervention exposure, or other spreadsheet-based metadata.
For an ELLY-style study, a researcher could compare C1 and C2, examine differences in reported inclusion or engagement, and identify whether barriers recur across segments. Cross-segment analysis does not replace sampling decisions or reflexive interpretation, but it can make comparisons faster and more systematic.
Problem: Complex pathways are hard to communicate
Evidano can help researchers turn coded findings into visual outputs such as hierarchical codes and subcodes, word clouds, and co-occurrence networks.
A hierarchical code structure could separate intervention components, mechanisms, reported outcomes, and barriers. A co-occurrence network could help a research team inspect relationships between codes such as confidence, walking, social connection, food, and wellbeing, with human review required before treating those relationships as substantive findings.
Problem: Sensitive health interviews require careful governance
Health researchers should use secure workflows because qualitative interviews can contain identifiable information and sensitive disclosures. Evidano states that data are encrypted and never used to train third-party models; teams can review the platform’s data security information before selecting a workflow.
Researchers should still obtain appropriate consent, follow institutional requirements, redact personally identifiable information where appropriate, and retain human oversight throughout analysis.
FAQ: AI Qualitative Analysis for Community Health Research
What is AI qualitative analysis?
+
AI qualitative analysis uses artificial intelligence to help organize, code, summarize, compare, and visualize unstructured research data such as interviews and open-ended survey responses.
The ELLY study demonstrates why human interpretation remains essential: researchers used framework analysis, team discussion, field notes, and independent cross-checking to interpret participant accounts.
How many interviews did the ELLY qualitative study conduct?
+
The ELLY study conducted 20 interviews in 2023, with 10 participants interviewed from each of two disadvantaged Scottish communities.
The researchers used approximately 30-minute, face-to-face, semi-structured interviews at 12 weeks after the intervention began.
What were the main qualitative themes in the ELLY study?
+
The main ELLY themes were community connection, positive behaviour change, healthier diet, mental and physical health improvements, and new skills and knowledge.
The 2026 PLOS One study also described snowball recruitment and awareness of community assets as important mechanisms or pathways.
Can qualitative interview analysis prove that an intervention caused health improvements?
+
Qualitative interview analysis cannot prove causation on its own, and the ELLY authors explicitly stated that their data were insufficient to establish definitive causality.
Qualitative evidence can explain participant perceptions, identify plausible mechanisms, reveal barriers, and generate hypotheses that future longitudinal or mixed-methods studies can test.
How can AI help analyze community health interviews?
+
AI can help community health researchers code transcripts, identify recurring themes, compare segments, retrieve supporting excerpts, and visualize relationships among codes.
The ELLY study suggests that researchers should structure analysis around intervention components, mechanisms, participant experiences, reported outcomes, and contextual barriers rather than relying only on topic frequencies.
Conclusion and Next Steps for AI Qualitative Analysis
The 2026 ELLY study shows how qualitative interviews can reveal the pathways behind a community health intervention, including social connection, confidence, healthier routines, and local engagement. AI qualitative analysis can reduce the manual burden of organizing this evidence and make thematic and cross-segment comparisons easier, while researchers remain responsible for context, ethics, reflexivity, and causal claims. A defensible workflow combines AI-supported synthesis with transcript-level evidence, transparent coding decisions, and human review. Researchers can try Evidano for free to analyze qualitative documents and uncover structured insights from interview data.
Topics
- AI qualitative analysis
- AI thematic analysis
- qualitative health research
- community intervention research
- qualitative interview analysis
Keep reading
- 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.
- Commentary on NewsWhat ELLY Shows About AI Qualitative AnalysisLearn how AI qualitative analysis can reveal intervention pathways, community needs, and participant outcomes from the ELLY health study. Try Evidano.
- Commentary on NewsWhat Java’s Study Shows: AI Qualitative AnalysisLearn how AI qualitative analysis can organize mental health research findings, compare stakeholder perspectives, and support context-sensitive decisions. Try Evidano.