Evidano home
Commentary on News

Faster Insights: mHealth palliative care qualitative analysis

Evidano7 min read

Fast, defensible qualitative analysis of mHealth palliative care data is possible and necessary. The JMIR pilot of mPallCare in Bidibidi Refugee Settlement (Uganda) shows how routine patient-reported symptom and outcome data (n=32; Aug–Oct 2022) produced a 6-week longitudinal dataset with clear themes, operational problems, and action signals (combined completion 67%). This post translates those findings into a repeatable, AI-enabled qualitative analysis workflow researchers, UX teams, and policy analysts can use to transform transcripts, form data, and dashboards into themes, segment comparisons, and prioritized recommendations. Follow the playbook below and see how Evidano can ingest your transcripts/surveys, auto-code themes, run cross-segment analyses, and produce shareable visuals, all with data encryption and no third-party model training.

Key Takeaways

Evidano is an AI-powered qualitative data analysis platform that ingests transcripts, CSVs, and app exports and automates thematic, frequency, and cross-segment reporting. This post presents a pragmatic 7-step workflow to analyze mHealth palliative care data from the mPallCare pilot and turn routine reports into prioritized operational recommendations.

  • mPallCare pilot (Bidibidi, Uganda) was feasible and acceptable over 6 weeks with 32 participants and a combined report completion rate of 67%.
  • Top-reported symptoms were headache, muscle pain, and dizziness, and main barriers were navigation, translation accuracy, intermittent sync/connectivity, and device availability.
  • Use an AI-assisted codebook, cross-segment comparisons, and POS time-series to surface equity gaps and to prioritize UX fixes, language support, and triage rules.

Fast take: what the study found (source)

The mPallCare pilot was feasible and acceptable over a 6-week, uncontrolled mixed-methods study in Bidibidi Refugee Settlement, Uganda. Full report: JMIR.

  • Sample: 32 people with advanced cancer; VHTs collected reports (training May 2022; enrollment Aug–Oct 2022).
  • Completion: 163/192 symptom reports (84.9%), 266/448 outcome reports (59.4%), combined 429/640 (67%).
  • Top symptoms: headache (84.4% of participants), muscle pain (84.4%), dizziness (81.3%).
  • Main barriers: navigation, translation accuracy, intermittent sync/connectivity, device availability; users valued improved communication and faster clinical responses.

Findings snapshot

MetricValueSource / Note
Study design6-week uncontrolled mixed methods feasibilitymPallCare pilot (JMIR, published 26 June 2026)
Participants32 patients with advanced cancer; 20 trained VHTsRecruited from Bidibidi; Aug–Oct 2022
Report completionSymptom 84.9% | Outcome 59.4% | Combined 67%Feasibility threshold ≥65%
Common symptomsHeadache, muscle pain, dizzinessTable 2 in source
LanguagesEnglish, French, Swahili, Juba Arabic (later Bari, Ma’di, Lingala planned)Translation/back-translation used
Key UX issuesNavigation, translation, sync errors, device logisticsFrom qualitative interviews

How mPallCare worked (plain English)

mPallCare used village health teams with an Android app to collect weekly symptom checklists and APCA POS outcome ratings, and submitted data appeared on a clinician dashboard with color flags. VHTs used an Android app to collect weekly symptom checklists (17 yes/no items) and APCA Palliative Outcome Scale (POS) outcome ratings every 3 days. Submitted data appear on a clinician dashboard with flags (green/yellow/red). Self-management content (text/graphics) is delivered after reports. The platform captured longitudinal patient-level signals that informed clinician outreach and supply planning.

  • Data pipeline: VHT entry → secure server → clinician dashboard → clinician action log.
  • Measures: 17-item symptom checklist + APCA POS (7 domains; 0–5 scales).
  • Operational constraints: intermittent connectivity, device sharing, translation gaps, VHT workload.

What this means for researchers & UX teams: qualitative analysis of mHealth palliative care

Researchers (design & evaluation)

Researchers should treat the dataset as mixed structured and unstructured material suitable for mixed-methods synthesis. The dataset blends structured (POS, checklists) and unstructured (interview transcripts, clinician notes) material, ideal for mixed-methods synthesis. Priorities: construct temporal theme maps (change in Pain, Peace, Feelings), estimate minimal clinically important differences for APCA POS in context, and link reported flags to clinician response lag times.

Use-case: rapid thematic coding + cross-segment comparisons (e.g., by ECOG score, language group, VHT zone) to surface equity gaps and triage rules.

UX & product teams

Product teams should focus on simplifying input flows, improving local-language support, and surfacing clinician feedback to users. The qualitative findings point to three product levers: simplify input flows (fewer fields, pictorial/audio prompts), expand true local-language support, and surface clinician feedback to users (so patients/VHTs see actions taken).

Metric to track: weekly active reporting, time-to-clinician-action after flagged report, and local-language comprehension errors.

Policy & implementers

Policy teams should integrate routine patient-reported data into national systems and use longitudinal trends to justify resource allocation. Integration with national systems (DHIS2) and establishing escalation pathways are key for sustainability. Policy teams should use longitudinal patient-reported data to justify resource allocations for palliative care in displacement settings.

Do more, faster with Evidano (mapped to this use case)

Ingest & unify the messy inputs

Evidano ingests interview transcripts, clinician notes, app exports (CSV), and survey sheets so teams can analyze APCA POS time series alongside interview themes without manual reformatting. Evidano ingests interview transcripts, clinician notes, app exports (CSV), and survey sheets so you can analyze APCA POS time series alongside interview themes without manual reformatting.

Automated thematic + frequency analysis

Evidano can run AI-assisted code suggestions, lock a codebook, and auto-apply it across transcripts and forms to return theme frequencies, co-occurrence networks, and prioritized quotes. Run AI-assisted code suggestions, then lock a codebook and auto-apply it across transcripts and forms. Evidano returns theme frequencies, co-occurrence networks, and prioritized quotes (clickable for audit).

Benefit: turn 6 weeks of mixed data into an evidence-backed list of top barriers (navigation, sync, translation) in minutes, not weeks.

Cross-segment and longitudinal views

Evidano provides segment comparisons and POS domain trajectories so teams can spot where outcome reporting declines or diverges by access, language, or disease status. Compare segments (e.g., mobile access vs no mobile access, by ECOG score, or by language) and chart POS domain trajectories. Evidano creates segment breakdowns and highlights statistically meaningful divergences for follow-up.

This maps directly to needs identified in the mPallCare pilot, e.g., where outcome reporting declines, identify whether device access or disease progression explains attrition.

Secure, auditable, and research-first

Evidano keeps project data encrypted and supports PII redaction, transcript/translation verification, and export-ready visuals for funders and Ministries. Data are encrypted and never used to train third-party models. Evidano supports PII redaction, transcript/translation verification, and export-ready visuals for funders and Ministries.

Checklist: 7-step AI-enabled qualitative workflow to reproduce these insights

Use this 7-step checklist to reproduce the analysis quickly on a mixed mHealth + qualitative pilot.

  • 1) Collect & centralize: export app CSVs, interview transcripts, and clinician notes into one folder (deidentify first).
  • 2) Ingest to Evidano: upload transcripts, CSVs, and codebook (if you have one).
  • 3) Auto-suggest codes: run AI code suggestion; review and finalize the codebook with a small team.
  • 4) Auto-code + validate: apply codes across all documents; validate on 10% of items and reconcile discrepancies.
  • 5) Run cross-segment analysis: compare by language, access-to-phone, ECOG, and timeline windows.
  • 6) Create decision-ready outputs: ranked barriers, quote bank, co-occurrence networks, and POS trend charts.
  • 7) Export & integrate: create a one-page stakeholder brief and export structured counts to DHIS2-compatible CSVs or policy dashboards.

FAQ: mHealth palliative care

Can AI handle translations and local language variants?

Yes, AI can assist with translations but always requires human validation. Use custom dictionaries and human-in-the-loop review; the mPallCare pilot found initial translation gaps, so validation is essential.

How do I measure sustained engagement?

Measure sustained engagement with weekly completion rates, time-to-response after flagged reports, and segment attrition, plus qualitative follow-up to diagnose causes. Track weekly completion rates, time-to-response after flagged reports, and segment attrition. Use qualitative follow-up to diagnose causes such as device access, disease progression, or VHT workload.

What completion rate defines feasibility in the mPallCare pilot?

The pilot used a feasibility threshold of 65% and achieved a combined completion rate of 67%. The combined completion rate across symptoms and outcomes was 67%, exceeding the stated feasibility threshold of ≥65%.

Conclusion & next step (strong CTA)

The mPallCare feasibility study demonstrates that routine symptom and outcome monitoring provides operational signals to prioritize UX fixes, language expansion, triage rules, and DHIS2 integration. The mPallCare feasibility study (Bidibidi, Uganda) demonstrates the value of routine symptom and outcome monitoring in humanitarian settings, and highlights exactly where qualitative analysis can turn signals into operational changes (UX fixes, language expansion, triage rules, DHIS2 integration).

  • Ready to run this workflow on your transcripts, survey sheets, and clinician logs? Start a secure pilot: upload your files to Evidano and generate thematic, frequency, and cross-segment reports in hours, not weeks.
  • Try the platform now: Try Evidano for free and reference the mPallCare findings (JMIR) to accelerate your palliative care or humanitarian research pipeline.

Topics

  • mHealth palliative care
  • qualitative analysis workflow
  • mPallCare pilot
  • APCA POS
  • Evidano

Keep reading

Browse all articles

Company

HomeContactBlogsGuides

About

FeaturesThematic Analysis SoftwareAI Thematic AnalysisAI Qualitative CodingQuote TraceabilityMultilingual AnalysisCompare QDA ToolsPlansCase StudiesData SecurityHow We Trained Our AI ModelsAffiliate ProgramHow to Cite & Journal AI PoliciesCompare Evidano with Other SolutionsCertified TrainersHuman vs AIEnvironmental Impact

Preview Evidano

View sample AI docx/pdf analysisView sample AI xlsx/csv analysisView sample AI audio transcriptionTry a demo AI survey / interviewPreview AI ChatbotAI Detection

Account

LoginRegister for Free Plan (free forever, no expiry)Book demo

Newsletter

© Evidano. All rights reserved.

Privacy PolicyTerms of Service