Evidano is an AI-powered qualitative data analysis platform that auto-ingests interview audio, transcribes with a custom medical dictionary and PII redaction, and generates thematic analyses and prioritized requirements. Fast, reliable qualitative analysis of telemedicine is a practical requirement for product teams, policymakers, and clinical UX researchers. A June 8, 2026 PLOS One study of 17 general practitioners in Indonesia (interviews 19 June–11 Sept 2024) maps clinical tasks to specific application requirements and identifies persistent gaps in integration, exam support, and decision-making. Read the full study at PLOS One. In this post you will learn a reproducible AI-enabled workflow to convert transcripts like these into validated design requirements and prioritised recommendations, and how Evidano automates the heavy lifting.
Key Takeaways
This post explains how to extract prioritized product requirements from telemedicine interviews using an AI-enabled qualitative workflow, illustrated by a PLOS One study of 17 GPs (interviews 19 June–11 Sept 2024). The study identifies top unmet needs (data/services integration, clinical communication, clinical task support) and reports measurable performance gains in efficiency, service quality, and physician–patient communication.
- Seventeen general practitioners were interviewed (interview window: 19 June–11 Sept 2024) to map clinical tasks to app capabilities using the Task‑Technology Fit model.
- Top unmet needs reported by clinicians were data/services integration, clinical communication, and clinical task support, which should be product design priorities.
- Platform exposure in the study was Good Doctor 100%, Halodoc 29.41%, and Alodokter 17.64%, and clinician experience was concentrated around ~47.06% with 5–7 years and ~47.06% with 2–4 years of telemedicine use.
- An AI-enabled workflow (transcription, codeframe application, cross-segment synthesis, clinician validation) reproduces these insights faster and yields prioritized requirements and stakeholder-ready artifacts.
Findings snapshot (quick view)
| Item | Value | Why it matters | Source |
|---|---|---|---|
| Published | June 8, 2026 | Current peer-reviewed evidence (PLOS One) | PLOS One |
| Sample | 17 GPs | Saturation reached; in-depth perspectives | Study methods |
| Interview window | 19 June – 11 Sept 2024 | Recent hands-on telemedicine practice | Study methods |
| Platform exposure | Good Doctor 100%; Halodoc 29.41%; Alodokter 17.64% | Cross-platform experience increases generalisability | Results |
| Top unmet needs | Data/services integration; clinical communication; clinical task support | Design priorities for product teams | Theme: Telemedicine requirements |
| Performance gains reported | Efficiency, service quality, physician–patient communication | Measurable outcomes to track post-release | Theme: Performance impact |
What the study actually found (concise)
The study found that applying the Task‑Technology Fit (TTF) model to 17 GP interviews mapped clinical tasks (anamnesis, remote physical exams, diagnosis, prescribing, referrals, follow-up) to required app capabilities and revealed persistent gaps in referrals, physical exam support, diagnostic accuracy, and inter-physician communication. While most GPs reported routine telemedicine use, key gaps remain in integrated referrals, thorough remote physical exams, diagnostic accuracy, and inter-clinician communication.
- All GPs perform anamnesis remotely; physical exams are often constrained to photos or patient-reported device readings.
- Critical requirement groups identified were clinical communication, data & services integration, clinical task support, reliability, and privacy/security.
- Most GPs used available features regularly; ~47.06% had 5–7 years' telemedicine experience and another ~47.06% had 2–4 years.
Why this matters for researchers, UX & policy teams
For UX/Product teams
Product teams should prioritise integration and structured clinical task templates because clinicians repeatedly flagged these as high-value, cross-task features. Prioritise EHR/labs/pharmacy integration and structured templates for prescriptions and referral letters to reduce clinician burden and improve safety.
Track the impact metrics the study highlights: service efficiency, decision quality, diagnostic accuracy, and physician–patient communication.
For qualitative researchers
Qualitative researchers should use an integrated deductive+inductive content analysis approach, because the study shows value in combining a TTF coding frame with emergent themes to ensure comparability and discovery. Use similar codebook scaffolding to replicate findings and maintain analytic rigor.
Document saturation method clearly: the authors used code-meaning saturation and stopped at 17 interviews, which is a replicateable and defensible approach for focused clinical topics.
For policy & clinical leaders
Policy and clinical leaders should define regulatory boundaries and minimum functional standards because clinicians in the study asked for explicit guidance on telemedicine scope, including emergencies and complex diagnostics. Establish secure data integration, referral workflows, and uptime/reliability SLAs as baseline requirements.
Minimum functional standards should include secure data integration, referral workflows, and uptime/reliability service-level agreements to protect patient safety and clinical workflows.
Do more, faster with Evidano (mapping features to gaps)
Problem: Manual transcript coding is slow
Evidano maps study gaps to features by automating ingest, transcription, and codebook application, because manual coding is time-consuming. Evidano auto-ingests interview audio, transcribes with a custom medical dictionary and PII redaction, then applies the TTF codebook to produce thematic and frequency analyses in minutes.
Problem: Hard to compare segments (experience, platform used)
Evidano lets teams compare segments quickly by automating cross-segment analysis, because manual comparisons are error-prone and slow. Cross-segment analysis in Evidano lets you compare themes by clinician seniority, platform (Good Doctor vs Halodoc), or use frequency, revealing which groups most demand referrals or integration.
Problem: Stakeholders want actionable requirements, not raw quotes
Evidano produces prioritized requirements with supporting evidence, because product managers need actionable outputs rather than unstructured quotes. Evidano generates prioritized requirement lists (clinical communication, integration, templates) with supporting quotes and co-occurrence networks so product managers can build a roadmap with evidence.
Problem: Multilingual, messy inputs from devices/patients
Evidano handles mixed-language corpora with translation and custom dictionaries, because many telemedicine settings include multiple languages and device inputs. Built-in translation with a custom dictionary plus transcription accuracy controls make it practical to analyze mixed-language corpora common in settings like Indonesia.
Security note
Evidano enforces data protection controls appropriate for clinical transcripts, because handling sensitive clinical data requires encryption and clear model-use policies. Evidano uses end-to-end encryption and does not use customer data to train third-party models.
Actionable 7-step workflow to reproduce the study’s product insights (AI-enabled)
This checklist converts interview transcripts into prioritized product requirements within two weeks by automating transcription, coding, synthesis, and validation.
- 1) Import & prep: Upload audio/video and survey files into Evidano; apply the medical custom dictionary and enable PII redaction.
- 2) Transcribe & translate: Auto-transcribe interviews (the study used Microsoft Teams recordings); review flagged low-confidence segments.
- 3) Apply codeframe: Seed the TTF codebook (task, tech, utilization, performance) and run AI-assisted coding across transcripts.
- 4) Synthesize themes: Use thematic and co-occurrence analyses to surface critical requirement clusters, for example integration and referral workflows.
- 5) Cross-segment checks: Run frequency and cross-segment analyses (experience, platform) to prioritize which features serve the largest or highest-risk groups.
- 6) Validate with clinicians: Export clickable quotes and a one-page evidence brief; conduct a one-hour validation session to confirm priorities.
- 7) Deliver artifacts: Generate stakeholder-ready visualizations (word clouds, hierarchy of codes) and a prioritized requirements spreadsheet for engineering.
FAQ: qualitative analysis of telemedicine
What is 'qualitative analysis of telemedicine' and when should I use it?
Qualitative analysis of telemedicine is the systematic coding and synthesis of interviews, notes, and transcripts to surface user needs, workflows, and failure modes. Use qualitative analysis when product decisions affect clinical safety, workflows, or regulations, and when you need rich, contextual evidence to prioritize design choices.
How do I compare segments (e.g., platform, clinician experience) reliably?
You should compare segments with a standardized codebook and frequency analysis, because consistent coding enables reliable cross-group comparisons. Use standardized codebooks (the study used TTF constructs) and run cross-segment frequency and significance checks; Evidano automates these comparisons and highlights divergent themes.
How secure is AI-enabled qualitative research with clinical data?
AI-enabled qualitative research can be secure if you enforce PII redaction, encrypted storage, and vendor commitments not to use your data for third-party model training. Ensure PII redaction, encrypted storage, and vendor commitments not to use your data for third-party model training; Evidano enforces these controls for health-sensitive projects.
Wrapping up & next moves
The PLOS One study (June 8, 2026) provides a clear, clinician-driven roadmap: integrate data/services, improve clinical communication, and add task templates to improve telemedicine fit and performance. If your team runs interviews or has existing transcripts, you can reproduce the study’s insights faster by applying an AI-enabled qualitative workflow and the seven-step checklist above.
- Try a pilot: upload a small set (5–10) of interviews to www.evidano.com and run the TTF codebook to produce prioritized requirements in hours, not weeks.
- If you need a template: export the study-aligned TTF codebook and a one-page evidence brief from Evidano to share with product and policy stakeholders.
Read the original paper: PLOS One. Ready to accelerate your qualitative telemedicine research? Try Evidano for free.