Qualitative analysis of hyperphagia in Bardet‑Biedl syndrome shows semi-structured interviews detect far more severe cases than a 14-item questionnaire in a July 2026 mixed-methods study of 51 adults with BBS. The study (accepted 9 June 2026; deposited 2 July 2026) compared questionnaire results to in-depth interviews and found interviews classified severe hyperphagia in 66.7% (10/15) of interviewees versus 5.9% (3/51) by questionnaire. Read the full deposit at LSE Research Online. Note: findings are research-focused and non-diagnostic.
Key Takeaways
Evidano is an AI-powered qualitative data analysis platform that transcribes, securely stores, redacts, codes, and visualizes clinical interview data.
Semi-structured interviews in a July 2026 mixed-methods study of 51 adults with Bardet‑Biedl syndrome identified substantially more severe hyperphagia than a 14-item questionnaire, reclassifying 86.7% (13/15) of interviewees to a higher severity.
Research teams should add targeted interviews and auditable qualitative coding when questionnaires inform prevalence, eligibility, or triage decisions.
- Study snapshot: Mossman et al. compared a 14-item questionnaire (n=51) with semi-structured interviews (n=15); interviews classified severe hyperphagia in 66.7% vs 5.9% by questionnaire.
- Reclassification: 86.7% (13/15) of interviewees were rated at a higher severity by interview than by questionnaire.
- Practical implication: relying solely on self-report questionnaires risks underestimating clinical and lived burden, so add interviews and auditable coding.
Fast take + source
Fast take: Mossman et al. (accepted 9 June 2026; deposited 2 July 2026) compared a 14-item hyperphagia questionnaire (n=51) with semi-structured interviews (n=15) and found interviews classified severe hyperphagia in 66.7% versus 5.9% by questionnaire.
- Population: 51 adults with BBS (mean BMI 36.5 [SD 8.8]; 56.9% aged 18–34).
- Core finding: 86.7% (13/15) of interviewees were rated at a higher severity via interview than by questionnaire.
- Why it matters: relying on self-report questionnaires alone risks underestimating clinical and lived burden.
Source: see the deposited manuscript at LSE Research Online.
Findings snapshot
| Metric | Value | Note / Source |
|---|---|---|
| Total sample | n = 51 | Mixed-method cross-sectional study (Mossman et al., Frontiers in Endocrinology) |
| Interview subset | n = 15 | Semi-structured interviews; transcripts coded and reviewed with a BBS clinical expert |
| Questionnaire: severe | 5.9% (3/51) | 14-item hyperphagia questionnaire |
| Interview: severe | 66.7% (10/15) | Systematic qualitative coding raised severity ratings |
| Reclassified higher via interview | 86.7% (13/15) | Interview led to upward reclassification compared with questionnaire |
| BMI ≥30 cohort | n = 39 (overall); interview within cohort n = 13 | Interview: severe 69.2% (9/13); none classified as mild |
What happened (methods, briefly)
The study was a non-interventional, cross-sectional UK mixed-methods design combining a structured 14-item hyperphagia questionnaire with in-depth semi-structured interviews.
Participants currently or previously treated with setmelanotide were excluded, and interview transcripts (n=15) were coded by a qualitative researcher then reviewed with a clinical BBS expert to assign severity ratings.
- Sample demographics: 82.4% White; age at diagnosis ranged from birth to 50 years; mean BMI 36.5 (SD 8.8).
- Key analytic choice: compare categorical severity across instruments and check interviewer-assisted reclassification.
- Result: qualitative interviews surfaced behaviours and context missed by the questionnaire, producing far higher severe ratings.
So what for researchers, UX teams and clinicians
For qualitative and clinical researchers
Qualitative and clinical researchers should not assume that face-valid questionnaires capture compulsive, context-dependent behaviours.
Add targeted semi-structured interviews to detect intensity, frequency, and functional impact.
Plan for coder triangulation: the study paired a qualitative coder with a clinical expert to improve severity assignments, an approach you can operationalize.
For survey and instrument designers
Survey and instrument designers should adjust question formats to account for cognitive differences and stigma that compress self-ratings.
Use mixed items (vignettes, behavioural anchors) and validation via interviews to calibrate cut-points.
Run small interview pilots (n≈10–20) to identify missing response frames before scaling a questionnaire.
For policy analysts & clinicians
Policy analysts and clinicians should expect that prevalence estimates based only on questionnaires may understate need for services or interventions.
Qualitative augmentation reveals unmet burden and can change triage or eligibility decisions.
When evaluating treatments or services, include qualitative severity metrics to capture real-world change beyond numeric scores.
Do more, faster with Evidano (map to this study)
Transcription & secure handling
Use Evidano for automated transcription, PII redaction, and encrypted storage to reduce manual transcription and protect clinical transcripts.
Problem: interviews require accurate transcripts and PII safeguards; Evidano offers automated transcription with custom dictionaries (medical terms, names) plus PII redaction and end-to-end encryption so clinical transcripts stay private.
Benefit: get clean, searchable text ready for coding without manual re-typing.
Qualitative coding, theme extraction & cross-segment checks
Use Evidano to accelerate coding, extract themes, and run cross-segment analyses such as by BMI or age to replicate the study’s coder plus clinical review pattern.
Problem: manual coding is slow and inconsistently applied; Evidano ingests transcripts and generates thematic, content frequency, and cross-segment analyses.
Benefit: import your codebook, use AI-suggested codes, then review and finalize with human oversight.
Reclassification and evidence for stakeholders
Use Evidano to produce auditable evidence linking coded quotes to severity tags to show why interview-based severity changed.
Problem: proving that interviews change severity assignments requires auditable evidence; Evidano links coded quotes to severity tags and produces shareable visualizations.
Benefit: produce an audit trail showing why a participant’s severity label moved from ‘mild’ to ‘severe’.
Scale interviews ethically
Use Evidano to scale semi-structured interviews ethically with consent flows, PII controls, and autonomous collection options.
Problem: interview scale is resource-heavy; Evidano supports AI avatar interviewers for autonomous qualitative collection with custom prompts and consent flows.
Benefit: run reproducible semi-structured interviews at scale while preserving data governance; Evidano data is never used to train third-party models.
Two-week reproducible workflow (practical checklist)
This two-week reproducible workflow shows how to design, pilot, transcribe, code, compare, and report a mixed-methods study using the study’s approach.
Week 1: Design & pilot
- Day 1–2: Import existing questionnaire items and study metadata into Evidano; set BMI and age segments.
- Day 3–5: Run 5–10 pilot interviews (AI avatar or live), transcribe with custom dictionary, and flag PII redaction rules.
Week 2: Code, compare, and report
- Day 6–9: Auto-suggest codes, finalize codebook, and apply to all transcripts. Use cross-segment frequency and co-occurrence visualizations to spot severe-symptom clusters.
- Day 10–12: Compare questionnaire vs interview severity labels across segments; export a stakeholder brief with quotes tied to reclassified cases.
- Day 13–14: Prepare decision memo (eligibility, triage, or follow-up study) and share interactive visualizations with clinicians/policy teams.
FAQ: qualitative analysis of hyperphagia
Why did interviews find more severe cases?
Interviews found more severe cases because interviews capture context, compensatory behaviours, and obsessive food-seeking that short questionnaires miss.
Interviews are better able to surface intensity, functional impact, and behaviours where cognitive differences or stigma suppress self-report.
How many interviews do I need to validate a questionnaire?
A small purposive sample, typically about 10–20 interviews, is often sufficient to reveal missing frames and validate a questionnaire.
Select participants to span key segments such as BMI and age at diagnosis and use Evidano to scale to larger samples if needed.
Is this approach ethical for sensitive clinical topics?
Yes, this approach is ethical when combined with informed consent, PII redaction, and secure storage.
The study’s findings are research-focused and non-diagnostic; clinical decisions should involve qualified practitioners.
Wrapping up & next steps
Mossman et al.’s July 2026 study shows qualitative interviews uncover hyperphagia severity and lived impact that questionnaires can miss.
- Next move: replicate the paper’s mixed-methods comparison on a pilot cohort (n≈50) and use Evidano to transcribe, code, and visualize reclassifications.
- Start: Try Evidano for free to see how Evidano maps transcripts to severity labels and produces shareable evidence for clinicians and policymakers.