Evidano is an AI-powered qualitative data analysis platform that provides secure ingestion, AI-assisted coding, and reproducible exports for perception studies. This post shows researchers and UX teams how to run a rigorous qualitative analysis of Electronic Immunization Registry (EIR) perceptions using the Rajshahi City pilot as a worked example. The primary keyword is qualitative analysis of EIR perceptions. Published July 2, 2026, the PLOS study (n=321: 305 caregivers, 16 HCPs) found high satisfaction but flagged connectivity and device limits. Read the original report at PLOS One. Ethics note: this is research-focused synthesis, not clinical guidance.
Key Takeaways
This post gives a reproducible, two-week workflow to run qualitative analysis of EIR perceptions using the Rajshahi pilot as an example. The Rajshahi pilot (published Jul 2, 2026) reported high acceptability but highlighted intermittent internet and device constraints that limited operations.
- Sample and timing: 321 participants (305 caregivers, 16 healthcare providers), data collected July–Sept 2024, paper published Jul 2, 2026.
- Top findings: 87% electronic registration at EPI centers; caregiver satisfaction mostly 59–78% satisfied and 16–22% highly satisfied.
- Reproducible workflow: secure ingestion → AI-assisted code suggestions → human validation → thematic frequency and cross-segment analysis, exportable as reproducible reports.
Fast take: Rajshahi EIR pilot (Jul 2, 2026)
The Rajshahi EIR pilot reports mostly satisfied or highly satisfied perceptions but operational friction from intermittent internet and device limitations. Caregivers and healthcare providers reported primarily positive acceptability, while implementation issues centered on connectivity and device constraints.
- Study published: July 2, 2026; data collection: July–Sept 2024.
- Sample: 321 respondents (305 caregivers, 16 HCPs).
- Key figures: 87% electronic registration at EPI centers; caregiver satisfaction mostly 59–78% satisfied, 16–22% highly satisfied.
- Original source: PLOS One
Findings snapshot
| Date | Location | Sample (caregivers/HCPs) | Registration setting | Main sentiment | Top operational issues | Source |
|---|---|---|---|---|---|---|
| Jul 2, 2026 (paper) | Rajshahi City Corporation, Bangladesh | 321 (305 / 16) | 87% at EPI centers; outreach house-to-house | Mostly satisfied / highly satisfied | Intermittent internet; device limitations | PLOS One |
What happened, data & methods in plain English
The study is a descriptive cross-sectional survey (July–Sept 2024) designed to capture perceptions of EIR-supported services. Data included Likert-scale items and open-ended responses, quantitative items were summarized with frequencies, and open-text answers were summarized descriptively. The authors note the study is not an effectiveness trial and open-ended data are not publicly shared due to identifiability concerns.
- EIR features in pilot: unique child IDs, tablet data entry, SMS reminders; system introduced in 2019, paper published July 2, 2026.
- Survey instruments: pretested semi-structured questionnaire; 5-point Likert scale for acceptability.
- Analytic approach: SPSS for descriptive stats; open responses summarized manually.
So what for researchers, UX teams, and policy analysts
Researchers
Researchers should treat the dataset as suitable for mixed-method triangulation, pairing published frequencies with granular thematic analysis of open-text responses (subject to ethical approvals). Pay attention to selection bias because participants were purposively sampled from EPI sessions and are likely more engaged users.
Action: request restricted access to open-ended responses through the IRB contact listed in the paper if deeper qualitative coding is needed.
UX & implementation teams
UX and implementation teams should prioritize offline-first UX, light-weight forms, and robust error handling to address the reported connectivity and device friction. The perceived benefits (reminders, record access) versus friction points (connectivity, devices) indicate those UX priorities.
Action: use thematic frequency analysis to quantify how often connectivity versus training appear in complaints and map issues by site/ward.
Policy & program managers
Policy and program managers should not assume urban acceptability guarantees rural scalability because the Rajshahi pilot is urban and infrastructure gaps vary by setting. Use cross-segment analysis (urban vs. peri-urban; facility vs. outreach) to target infrastructure investments.
Action: combine EIR quantitative metrics (timeliness, registration completeness) with coded qualitative themes to prioritize interventions.
How to run qualitative analysis of EIR perceptions (Evidano workflow)
Step 1; Secure ingestion
Collect questionnaires, transcripts, and device logs, then ingest documents into Evidano for secure processing and encryption. Evidano encrypts data at rest and in transit and does not use customer data to train third-party models.
If audio interviews exist, use Evidano transcription with a custom dictionary and PII redaction to standardize names like ward labels or local vaccine terms.
Step 2; Automated code suggestions
Run an initial unsupervised thematic extraction to surface common topics, such as connectivity, SMS reminders, and the registration process. Review and accept suggested code labels to seed a codebook.
Use hierarchical codes and subcodes to capture details, for example connectivity → intermittent signal and connectivity → slow upload.
Step 3; AI-assisted coding + validation
Apply AI-assisted coding across the full corpus, then run a human validation pass on a 10–20% sample to estimate coding precision and adjust rules. Track inter-rater metrics inside Evidano to document reliability for publication.
Step 4; Thematic frequency & cross-segment analysis
Compute theme frequencies, co-occurrence networks, and compare segments such as caregivers versus HCPs and facility versus outreach to reveal which themes drive dissatisfaction or satisfaction. Visualize co-occurrence patterns to identify candidate root causes.
Export tables and visualizations for stakeholder briefs and policy discussions.
Step 5; Deliverables
Produce a one-page evidence brief containing the top five themes, representative quotes, segment comparisons, and recommended fixes such as offline data capture. Generate a reproducible codebook and an appendix of method metrics suitable for peer review.
Checklist: 7 quick steps to reproduce the Rajshahi synthesis
This checklist lists seven reproducible steps to reproduce the Rajshahi synthesis in Evidano in two weeks.
- 1) Gather instruments, consent forms, and de-identified transcripts; confirm IRB access for open-text data.
- 2) Ingest documents into Evidano; run secure transcription if needed (custom dictionary for local terms).
- 3) Auto-extract themes; review top 15 candidate codes with subject-matter experts.
- 4) Run AI-assisted coding across the corpus; validate on a stratified 15% sample.
- 5) Run frequency tables and cross-segment comparisons (caregiver vs. HCP; facility vs. outreach).
- 6) Create visuals: co-occurrence network, hierarchical code tree, and quote packs for each theme.
- 7) Export reproducible reports and a policy brief; link recommendations to measurable KPIs (for example, percent of sessions with offline-capable devices).
FAQ: qualitative analysis of EIR perceptions
Can I analyze Likert items and open-text together?
Yes, you can analyze Likert items and open-text together as a mixed-methods synthesis. Treat Likert distributions as quantitative context and map open-text themes to high and low scorers to explain drivers of satisfaction.
How do I handle small provider samples (for example, n=16)?
Handle small provider samples by reporting provider findings descriptively and triangulating with richer qualitative quotes. Use caution about generalizability and clearly state limitations in any brief.
Is Evidano secure for handling potentially identifiable open-text responses?
Yes, Evidano encrypts data at rest and in transit and does not use customer data to train third-party models. Use built-in PII redaction at ingestion and follow IRB agreements when requesting restricted access to open-ended responses.
Wrapping up: next moves
The Rajshahi pilot (published Jul 2, 2026) provides a concise example showing positive acceptability tempered by infrastructure limitations. To move from descriptive impressions to prioritized action, combine thematic coding with cross-segment frequency analysis and visualize co-occurrence patterns to reveal root causes.
- Ready to reproduce this workflow on your dataset? Start a secure pilot and Try Evidano for free to ingest transcripts, run AI-assisted coding, and produce reproducible reports.
- Ethics reminder: handle open-text responses under IRB agreements and treat this synthesis as research-only, not diagnostic.