Evidano is an AI-powered qualitative data analysis platform that ingests documents and interviews, auto-transcribes, redacts PII, and runs thematic and frequency analysis. Fast, actionable insights matter when lives and scarce resources are on the line. A 2026 PLOS NTDs study of the September 2021 mass rabies vaccination campaign in southern Cercado (Cochabamba, Bolivia) found that none of 17 local health centers met the 80% protocol adherence threshold and identified concrete logistical, coordination, and communication failures (see original study: PLOS Neglected Tropical Diseases). In this post you will learn: 1) what the implementation fidelity gaps were (numbers, dates); 2) how mixed-methods qualitative analysis singled out root causes; and 3) an AI-enabled, reproducible workflow using Evidano (www.evidano.com) to convert interview transcripts, checklists, and campaign logs into prioritized, evidence-backed fixes. Ethics note: this analysis is research-focused and non-diagnostic.
Key Takeaways
The September 2021 mass rabies vaccination campaign in southern Cercado (Cochabamba, Bolivia) failed to meet the 80% protocol adherence threshold at any of 17 evaluated centers, with widespread cold chain, staffing, and communication failures.
A mixed-methods fidelity diagnostic using a 27-item Carroll checklist and CFIR-guided interviews identified operational and governance bottlenecks that are directly addressable with targeted logistics, coordination, and communication fixes.
- None of 17 centers reached ≥80% overall adherence, content scores ranged 32%–79%, and 12/17 centers met coverage targets.
- Cold chain gaps were common: 13/17 centers (76.5%) lacked refrigerators, and 8/17 (47.1%) reported damaged coolers.
- Binary checklist verification (YES/NO) plus 154 interviews (46 ETD staff, 108 pet owners) highlighted inner-setting and governance barriers rather than vaccine efficacy issues.
Fast take + source
Fast take: Gómez-Vidal et al. (published 10 July 2026) used a 27-item Carroll checklist and CFIR-guided interviews (46 ETD staff, 108 pet owners) to assess fidelity of the September 2021 campaign in southern Cercado and found no center exceeded 80% adherence, widespread cold chain and team composition failures, and weak communication and interinstitutional coordination.
- Study context: September 2021 campaign; retrospective data collection in late 2022; published 10 July 2026.
- Units: 17 first-level health centers in southern Cercado, Cochabamba.
- Methods: 27-point quantitative checklist (Carroll) plus CFIR-guided qualitative interviews (COREQ reported).
- Full paper: PLOS Neglected Tropical Diseases.
Findings snapshot (by the numbers)
| Date/Metric | Value | Source detail | Implication |
|---|---|---|---|
| Campaign date | September 2021 | Study design | Baseline event analyzed |
| Health centers evaluated | 17 | Carroll checklist | Complete municipal coverage |
| Overall adherence | 0 centers ≥80% | Checklist results (published Jul 10, 2026) | Protocol not consistently delivered |
| Content adherence range | 32%–79% | Per-center content scores | Large variability in core operations |
| Centers with adequate coverage | 12/17 (70.6%) | Coverage construct | Vaccine availability often sufficient for campaign day but postcampaign continuity varied |
| Centers maintaining frequency | 14/17 (82.4%) | Frequency construct | Daily sessions largely met but advance promotion low |
| Centers with sufficient duration | 11/17 (64.7%) | Duration construct | Some centers ran <8-hour sessions |
| Cold chain gaps | 13/17 lacked refrigerators (76.5%) | Content subitems | High risk of vaccine spoilage |
| Incomplete brigades | 15/17 (88.2%) | Content subitems | Shortage of channelers & registry staff |
| Damaged coolers | 8/17 (47.1%) | Content subitems | Field logistics compromised |
| Local burden (2021) | 69 confirmed canine cases in Cochabamba; 34 in Cercado | Background epidemiology | High local transmission risk |
| Human deaths (2021) | 5 reported in Bolivia | Background epidemiology | Ongoing public health impact |
What happened: implementation fidelity gaps explained
Implementation fidelity gaps were widespread in the campaign, driven by failures in cold chain, brigades, and promotion.
The study operationalized fidelity through four Carroll constructs: content (19 items), coverage (2), frequency (4), and duration (1), plus a cross-cutting documentation check.
Using a strict YES/NO documentary verification approach, researchers found systemic failures in basic logistics (cold chain, carriers), human resources (incomplete brigades), and communication (advance promotion under one month in most centers).
- Binary scoring (YES/NO) prioritized verifiable documentation; partial compliance was counted as NO, which is conservative but reproducible.
- Qualitative CFIR coding mapped barriers and facilitators to organizational context, values, expected efficacy, and communication/coordination.
- Triangulation (documents plus 154 interviews) showed the bottlenecks were operational (inner setting) and governance-related, not a lack of vaccine efficacy.
Implications for researchers and program teams: implementation fidelity of rabies vaccination
For implementation researchers
Implementation researchers should use parallel mixed-methods designs to separate 'what' (fidelity scores) from 'why' (CFIR contextual drivers).
The Bolivia study demonstrates reproducible mapping from checklist gaps to organizational causes.
Report binary verification but consider adding graded scales in follow-ups to capture partial compliance dynamics.
For program/UX teams running campaigns
Program and UX teams should prioritize inner-setting fixes such as refrigeration workarounds, formalized brigade staffing, and mandatory pre-campaign promotion timelines.
Replace or work around refrigeration gaps (thermotolerant vaccines or insulated carriers), formalize brigade staffing plans, and set mandatory pre-campaign promotion timelines.
Formalize OTB–ETD coordination via written agreements and standardized WhatsApp protocols to convert ad hoc facilitation into institutional practice.
For policy/One Health stakeholders
Policy and One Health stakeholders should integrate veterinary associations and institute routine pet censuses and facility-level vaccination targets.
Integrate veterinary associations in planning and execution to increase technical capacity and public trust.
Institute routine pet censuses and facility-level vaccination targets for data-driven resource allocation.
Do more, faster with Evidano: operationalizing the study’s methods
Turn documents and interviews into diagnostic themes
Evidano is an AI-powered qualitative data analysis platform that ingests PDF/CSV/MP3, auto-transcribes with custom dictionaries and PII redaction, and runs thematic and frequency analysis to surface prioritized, evidence-backed barriers.
Problem: You have checklists, WhatsApp logs, meeting minutes, and interview transcripts but need prioritized causes.
Evidano ingests PDF/CSV/MP3, transcribes with a custom dictionary and PII redaction, then runs thematic plus frequency analysis to surface the most frequent, evidence-backed barriers (for example, 'no refrigerator' and 'missing channeler').
Map qualitative codes to operational fixes
Evidano builds code hierarchies and co-occurrence networks so teams can see which barriers cluster and which actors appear in the same quotes.
Evidano automatically builds code hierarchies and co-occurrence networks so you can see which barriers cluster (for example, cold chain plus transport) and which actors appear in the same quotes (for example, 'OTB' plus 'lack of budget').
Exportable visualizations and quote-linking help program managers craft targeted SOPs and one-page decision memos for funders.
Run cross-segment analysis and track impact
Evidano computes frequency and sentiment by segment, enabling split analyses by center, zone, or stakeholder to identify differential barriers.
Split analyses by center, zone, or stakeholder (ETD versus pet owners) to identify differential barriers; Evidano computes frequency and sentiment by segment.
Set up repeated uploads (post-intervention evaluations) to measure fidelity changes over time, with data encrypted and not used to train third-party models: see Evidano for platform details.
Quick workflow: from raw data to prioritized fixes (2-week pilot)
This run-book reproduces the Bolivia study’s diagnostic in two weeks and produces an operational plan.
- Day 1: Ingest materials, upload 27-item checklists (CSV), vaccination logs, WhatsApp exports, and interview recordings. Enable PII redaction and add local terms to the custom dictionary (for example, OTB, SEDES).
- Days 2–3: Auto-transcribe and translate transcripts, verify a 10% sample, and run an initial topic extraction and frequency table.
- Days 4–6: Auto-code using a CFIR-aligned codebook template (apply Carroll checklist mapping). Review and refine codes with two coders to reach consensus.
- Day 7: Generate a co-occurrence network and a top-10 barrier list (frequency plus impact tags).
- Days 8–10: Produce stakeholder-specific briefs: (a) health centers (logistics fixes), (b) municipal managers (budget and coordination), (c) community leaders (communication plan).
- Days 11–14: Run a cross-segment comparison (for example, southern versus northern zones), finalize a prioritized 90-day action plan, and export visuals and a one-page executive memo for funders.
FAQ: AI-enabled qualitative analysis
Can AI accurately code local terms (Quechua and Spanish)?
Yes. Evidano supports custom dictionaries and translation with glossaries so local terms and place names retain meaning during coding.
Teams should add local terms and acronyms to the custom dictionary during ingestion to preserve semantic accuracy.
How do you compare segments reliably?
Use a locked codebook and reproducible analytics to compare segments reliably.
Evidano’s cross-segment frequency and co-occurrence analyses, when combined with locked code definitions, let you run the same codebook across groups for reproducibility.
Is sensitive data secure?
Yes. Evidano encrypts data end-to-end and does not use customer data to train third-party models.
For platform terms, see Evidano platform terms.
Wrapping up: what to do next
The Bolivia study (published 10 July 2026) shows that vaccines are effective but delivery systems often fail, and mixed-methods fidelity diagnostics produce actionable fixes.
- If you run vaccination campaigns or evaluate public health interventions, pilot the two-week Evidano workflow on one past campaign to prioritize the top three operational fixes and produce decision-ready briefs.
- Ready to convert transcripts, checklists, and messages into prioritized actions? Try Evidano for free and bring documentary evidence to your next planning table (original study: PLOS Neglected Tropical Diseases).