What ELLY Shows About AI Qualitative Analysis

The November 5, 2026 PLOS ONE study, “Enjoy Life LocallY (ELLY)”, found that a co-designed community intervention was associated with reported improvements in behaviour, diet, social connectedness, confidence, and wellbeing, although the qualitative findings could not establish definitive causation.

• According to the November 5, 2026 PLOS ONE study, ELLY was delivered to 75 residents across two disadvantaged Scottish communities during a 12-week period from September 2023 to November 2023.
• According to the PLOS ONE study, researchers interviewed 20 participants, 10 from each community, at 12 weeks and analyzed the interviews using framework analysis.
• According to the PLOS ONE study, the strongest qualitative pathways connected community activities and social interaction with positive behaviour change, healthier diets, improved confidence, and perceived mental and physical health benefits.
• According to the PLOS ONE study, 14 of the 20 interview participants, or 70%, were female, while 17 participants, or 85%, lived in areas classified as disadvantaged.
• According to the PLOS ONE study, the interview data supported a conceptual logic model but did not demonstrate definitive causality or prove that reported outcomes would continue after the 12-week intervention.

The ELLY study showed that qualitative analysis can explain how a complex community intervention may produce outcomes through several connected pathways. According to the PLOS ONE study published on November 5, 2026, the intervention combined free soup twice weekly, a loyalty card, a £25 shopping card for attending at least 9 community assets over 12 weeks, optional goal setting, information resources, and self-monitoring of weight and wellbeing.

According to the PLOS ONE study, the intervention was co-designed with two disadvantaged communities in Forth Valley, Scotland, and used local activities, clubs, groups, and community hubs. The researchers recruited 35 participants in one community and 40 participants in the other, then invited 10 participants from each community to complete interviews.

According to the PLOS ONE study, two researchers conducted approximately 30-minute, face-to-face, semi-structured interviews at participants’ homes or in community settings. The researchers initially developed a coding framework after reviewing five interviews, refined the framework through team discussion, independently cross-checked 10% of the data, and resolved disagreements through discussion with a third researcher.

The PLOS ONE study identified themes including snowball recruitment, connection with community assets, positive behaviour change, social connectedness, healthier diet, mental and physical health improvements, and new skills and knowledge. The research team then synthesized these themes into a conceptual logic model that separated intervention mechanisms from reported outcomes.

One participant described the behavioural pathway by saying, “Now I walk every night…I continue in this situation, like walking, and soup, and healthy food, that has helped me a lot.” Another participant summarized the social pathway as “a lovely sense of community.” These quotations show why qualitative evidence adds explanatory detail that outcome measures alone cannot provide.

AI qualitative analysis can help public health researchers connect participant language to intervention mechanisms, outcomes, and differences between communities. The ELLY study demonstrates that participant accounts may reveal indirect pathways, such as social connection and confidence preceding healthier routines, rather than a simple direct effect on weight.

The PLOS ONE authors reported that “wellbeing and social connection, rather than healthy weight, were more closely associated with positive changes in behaviour.” This finding suggests that researchers should code for unexpected mechanisms and participant-defined outcomes instead of limiting analysis to the intervention’s headline metric.

AI-assisted thematic analysis can support this work by organizing recurring concepts such as confidence, isolation, access to activities, diet, and belonging across a large transcript set. Human researchers still need to review codes, examine contradictory cases, protect participant confidentiality, and distinguish reported perceptions from demonstrated effects.

Cross-segment analysis is particularly relevant to studies like ELLY because the PLOS ONE authors reported differences in community cohesion, local asset availability, group inclusivity, and communication. Comparing communities, participant characteristics, engagement levels, and goal-setting experiences can show where an intervention works consistently and where local adaptation is required.

Health researchers should treat qualitative AI outputs as an analysis aid rather than a clinical conclusion. The ELLY study itself stated that interview data alone were insufficient to establish definitive causation and recommended longer-term, mixed-methods research.

The ELLY study shows why qualitative health research must examine pathways, context, and participant-defined outcomes rather than focusing only on headline measures. The study’s 20 interviews revealed how social connection, confidence, local activities, and healthier routines were linked in participants’ accounts, while also documenting barriers such as exclusion, irregular activities, and poor communication.

AI qualitative analysis can make this type of evidence easier to organize and compare, but responsible researchers must retain human review, protect sensitive data, and avoid treating perceived outcomes as proof of causation. Evidano helps research teams analyze transcripts and documents with thematic, frequency, and cross-segment workflows designed for qualitative research. Try Evidano for free and turn complex interview data into a clearer evidence base for decisions.