Researchers and UX teams evaluating mHealth need repeatable, fast qualitative workflows. The July 2, 2026 PLoS One protocol for the MyIBDDiet pilot (n=40, 60-day crossover) documents semi-structured interviews, CFIR-framed guides, verbatim transcription and NVivo thematic analysis to assess usability (MAUQ + TFA) and acceptability. Read the original protocol at PLoS One protocol and see how Evidano can accelerate transcription, thematic coding, cross-segment comparisons, and quote selection while preserving data privacy via www.evidano.com.
Key Takeaways
The MyIBDDiet pilot pairs validated usability scales (MAUQ, TFA) with CFIR-framed semi-structured interviews in a 60-day, randomized pilot (n=40) to generate rapid, actionable usability insights.
Evidano is an AI-powered qualitative data analysis platform that automates transcription, supports AI-assisted coding, and produces stakeholder-ready visuals to compress analysis time while preserving human validation.
- The pilot was published 2 July 2026 and registered as NCT06683105 on ClinicalTrials.gov on 8 Nov 2024.
- Mixed-methods primary outcome: usability and acceptability measured by MAUQ + TFA plus interviews and NVivo thematic analysis.
- A practical 7-step checklist maps data capture to transcription, AI-assisted coding, triangulation with MAUQ/TFA scores, and rapid stakeholder reporting.
- Evidano features described include custom dictionaries for transcription, AI-assisted coding with human reconciliation, cross-segment comparisons, and one-click visuals.
Fast take + source
Fast take: The MyIBDDiet pilot (published 2 July 2026) is a single-centre, randomized 60-day trial of a diet guidance app for people with inflammatory bowel disease focused on usability as the primary outcome.
Primary outcome: usability measured with validated scales (MAUQ, TFA) and explained with semi-structured interviews; sample: 40 participants (20 per arm) with baseline, Day 30, Day 60 and Day 180 follow-ups.
Trial registration: NCT06683105 (registered 8 Nov 2024). Full protocol: PLoS One protocol.
- Why this matters: the protocol pairs validated usability scales with qualitative interviews, a model researchers and product teams can replicate to demonstrate both user experience and early clinical signal.
- Privacy note: interviews are transcribed verbatim with identifiers removed; this is research-only, non-diagnostic work and requires informed consent and ethics oversight.
Findings snapshot
| Metric | Value | Source / Note |
|---|---|---|
| Publication date | 2 July 2026 | PLoS One protocol |
| Trial registration | NCT06683105 (registered 8 Nov 2024) | ClinicalTrials.gov |
| Sample size | n = 40 (20 per arm) | Pilot RCT; informs future power calc |
| Study length | 60 days (30-day crossover) | Assess behaviour change at Day 60; persistence at 6 months |
| Primary outcome | Usability & acceptability (MAUQ + TFA + interviews) | Mixed-methods primary endpoint |
| Qual methods | Semi-structured interviews; CFIR guide; thematic analysis in NVivo | Interviews transcribed verbatim; analyzed by two coders |
| Secondary outcomes | Diet quality scores, biomarkers (urine Na/Cl), CRP, fecal calprotectin, QoL | Quant + biological correlates |
What happened, methods in plain English
This section explains the co-design, data collection, and mixed-methods analysis used in the MyIBDDiet pilot.
The MyIBDDiet team co-designed a Mediterranean-style anti-inflammatory diet app with patient partners, instrumented the app to collect backend engagement metrics, and planned a mixed-methods pilot to test usability and early signals of diet change.
Qualitative data come from virtual semi-structured interviews framed with CFIR; interviews are transcribed verbatim, de-identified, and coded independently by two researchers with reconciliation by a senior investigator using thematic analysis (NVivo).
The study pairs validated usability scales (MAUQ, TFA) with qualitative themes to explain why users do or do not engage.
- Interview process: recruit until thematic saturation, expect approximately the number per group required until no new themes emerge.
- Analytic approach: inductive thematic coding plus mapping to TFA constructs, triangulate with MAUQ scores and app analytics.
So what for UX researchers & qualitative teams
Product / UX researchers
Product and UX researchers should combine MAUQ/TFA scores with interview themes to prioritize fixes with both high user pain and low usability scores.
Use backend engagement metrics to validate self-reported barriers (for example, video loading issues linked to drop-offs).
Clinical researchers & trialists
Clinical researchers and trialists can integrate qualitative acceptability into early app trials so feasibility data inform larger RCT design and sample-size calculations.
Link qualitative findings (barriers and facilitators) to recruitment, adherence, and retention strategies for multi-centre trials.
Policy & health services teams
Policy and health services teams can use rigorous qualitative usability evidence to justify digital tools in settings with limited dietetic access by showing acceptability and real-world barriers before scale.
Do more, faster with Evidano (mapped to this use case)
Problem: manual transcription & QA → Solution: automated, research-grade transcription
Evidano is an AI-powered qualitative data analysis platform that automates transcription, supports custom dictionaries for diet and medical terms, and offers PII redaction and timestamps to reduce time spent cleaning text.
The MyIBDDiet team transcribes interviews verbatim; Evidano automates transcription with a custom dictionary for diet and medical terms, PII redaction options, and timestamps so coders spend less time cleaning text and more time interpreting themes.
Problem: slow, inconsistent coding → Solution: AI-assisted coding + codebook import
Evidano supports importing CFIR-informed codebooks or MAUQ/TFA items and provides auto-suggested codes to accelerate the initial pass while keeping human validation.
Import your CFIR-informed codebook or MAUQ/TFA items into Evidano, accept and refine auto-suggested codes, and maintain human validation and inter-rater reliability.
Problem: correlating qualitative themes with segments → Solution: cross-segment analysis
Evidano enables thematic frequency and cross-segment comparisons, and exports co-occurrence networks to show which barriers cluster with low engagement.
Run thematic frequency and cross-segment comparisons (for example, early adopters versus low-engagers) and export co-occurrence networks to see which barriers cluster with low engagement.
Problem: stakeholder-ready outputs take days → Solution: one-click visuals & quotes
Evidano generates hierarchical code-to-subcode visualizations, word clouds, and downloadable quote tables mapped to themes and MAUQ scores for rapid stakeholder reports.
Generate hierarchical code → subcode visualizations, word clouds, and downloadable quote tables mapped to themes and MAUQ scores for rapid stakeholder reports.
Security & compliance
Evidano encrypts data at rest and in transit and uses proprietary LLMs tuned for qualitative research, and user data is not used to train third-party models, which supports clinical research governance.
Data encrypted at rest and in transit; proprietary LLMs tuned for qualitative research and user data is not used to train third-party models, important for clinical research governance.
Checklist: run the MyIBDDiet-style qualitative analysis in 7 steps
This checklist lists seven steps to run a MyIBDDiet-style qualitative analysis.
1. Collect: record interviews with consent and note timestamps. Export app analytics (events, session lengths, video load errors).
2. Transcribe: batch-transcribe with custom diet/clinical dictionary and apply PII redaction.
3. Import: upload transcripts and codebook (CFIR, TFA items) into Evidano.
4. Auto-code & review: run AI-assisted coding, then have two human coders reconcile differences; calculate inter-rater agreement.
5. Triangulate: merge MAUQ/TFA scores and app analytics with themes to flag high-impact usability issues.
6. Visualize: produce co-occurrence networks, hierarchical theme trees, and quote tables for each segment.
7. Report & iterate: export a stakeholder report and update the app backlog with prioritized fixes; plan a follow-up 60–90 day usability re-check.
FAQ: qualitative analysis of app usability
How many interviews until saturation?
Recruit until thematic saturation is reached: the MyIBDDiet pilot plans to recruit until saturation and pilots commonly reach saturation between 12–30 interviews per group depending on heterogeneity.
Use Evidano’s clustering to detect diminishing returns in new-theme emergence.
How do I link qualitative themes to quantitative usability scores?
Map interview codes to TFA and MAUQ domains and run cross-tab or regression analyses with MAUQ scores and engagement metrics to quantify relationships.
Evidano supports cross-segment frequency and co-occurrence analysis to make that mapping explicit.
Is automated coding reliable enough for trials?
AI-assisted coding speeds initial passes but should be paired with human validation and inter-rater reliability reporting for trials.
Use AI to surface patterns, not to replace expert judgement.
Wrapping up & next steps
Wrapping up: The MyIBDDiet pilot demonstrates pairing validated usability scales with rigorous qualitative interviews to inform app refinement and future trials.
The MyIBDDiet pilot (published 2 July 2026) is a practical example of pairing validated usability scales with rigorous qualitative interviews to inform app refinement and future trials.
If you run usability pilots or RCTs, adopt an AI-enabled pipeline to compress analysis time from weeks to days while keeping human oversight.
Ready to convert transcripts and MAUQ/TFA interviews into prioritized product changes? Try Evidano for free for secure transcription, AI-assisted thematic coding, cross-segment analysis, and stakeholder-ready visuals.
Original protocol: PLoS One protocol