Fast take: A June 24, 2026 PLOS One focus-group study of 36 residents in Kashongi, Southwestern Uganda, found universal concern about water quality, heavy time burdens fetching and preparing water, and mixed reactions to a prototype plant-xylem point-of-use (POU) filter (cost ≈ $11, slow flow, durability worries). Read the original study at PLOS One. If you run qualitative research on prototypes, this study shows how transcript-level patterns (who says what, when, and why) determine adoption risks. In this post we explain what researchers and product teams can extract from the study, and show an AI-enabled workflow (using Evidano) to move from raw audio/transcripts to prioritized design and implementation actions.
Key Takeaways
The core finding: the June 24, 2026 PLOS One focus-group study shows that technical performance alone will not guarantee uptake of point-of-use filters, time, cost, repairability, and trust determine adoption in this community of 36 participants.
Evidano is an AI-powered qualitative data analysis platform that auto-transcribes, translates, and runs thematic and cross-segment analyses to convert FGDs into prioritized product and program actions.
- POU adoption drivers: cost (prototype ≈ $11), slow flow, and durability concerns were the dominant barriers reported across six focus groups.
- Operational realities: fetching and preparing water consumed 1–7 hours per household, making time-to-water a critical usability metric.
- Research design note: segmented FGDs revealed richer transcripts from older women and sparser data from young men, informing recruitment and facilitation plans.
Findings snapshot: summary sentence
This table summarizes key study metadata and the most frequent user concerns reported by participants.
Findings snapshot
| Item | Value | Source / Note |
|---|---|---|
| Published | June 24, 2026 | PLOS One (original study) |
| Study period | April–May 2024 | Fieldwork dates reported by authors |
| Participants | 36 (6 focus groups of 6; segregated by age & gender) | Transcripts translated from Runyankore to English |
| Location | Kashongi, Kiruhura District, Southwestern Uganda | Rural, limited treated water access |
| Reported average household water use | 142 liters/day | Study report (participants' estimate) |
| Prototype cost | ≈ $11 USD (~40, 000 UGX) | Prototype uses Pinus halepensis xylem + bicycle pump |
| Top user concerns | Cost, durability, slow filtration, replacement parts | Common themes across groups |
What happened: study design & core results
This section summarizes study design and core results: the study used six local-language focus groups (April–May 2024) with recordings transcribed and translated to English, yielding 36 participants and published June 24, 2026.
- Community context: limited or no access to treated tap water; primary sources were wells/dams shared with livestock; many households reported water-related illness (children most affected).
- Time burden: fetching and preparing water took from 1 to 7 hours depending on distance/season; boiling requires fuel (cost/environmental burden).
- Prototype reaction: mixed, excitement about avoiding boiling and improving child health, but persistent concerns about the device’s slow flow, fragility, and replacement costs.
- Implementation history: community reported prior external projects that ended without follow-up, sustainability and local supply chains are decisive factors.
Qualitative analysis of point-of-use water filtration: what researchers and teams should note
For field researchers
Field researchers should note that design matters beyond lab efficacy, participants equated time-to-water and maintenance costs with utility.
Design matters beyond lab efficacy: participants equated time-to-water and maintenance costs with utility. Ask: will a technically effective filter be used daily?
Segmentation reveals risk: older women produced longer, richer transcripts (useful for lived-experience evidence); young men had sparse data, plan recruitment and facilitation accordingly.
History and trust shape adoption: prior, unsupported interventions reduced willingness to sustain behavior change, code for 'project fatigue' and include it in thematic reports.
For product & implementation teams
Product and implementation teams should focus on price sensitivity, operational friction, and supply-chain education before distribution.
Price sensitivity is critical: $11 prototype may be inexpensive for developers but unaffordable locally, model total cost of ownership, replacement cadence, and local sourcing.
Operational friction kills usage: slow throughput and need for manual pumping were repeated disincentives, prioritize flow-rate testing and alternatives (gravity-fed vs pressurized trade-offs).
Education & supply chains: participants misunderstood acceptable wood types; plan targeted, local-language educational assets and a parts supply plan before distribution.
Do more, faster with Evidano
Turn focus-group audio into prioritized insights
Use Evidano to import audio and translated transcripts, then run automated thematic, content frequency, and cross-segment analyses to surface dominant barriers and who emphasizes each concern.
Import audio and translated transcripts to Evidano, then run automated thematic, content frequency, and cross-segment analyses to surface the dominant barriers (cost, durability, slow flow) and which groups emphasize each concern.
Features to use: automated transcription with custom dictionary (Runyankore terms, local names), translation support, and AI chat over your documents to answer stakeholder questions in plain language.
Validate and quantify qualitative signals
Use Evidano’s frequency analysis and co-occurrence networks to confirm which topics cluster and prioritize fixes and messaging.
Use Evidano’s frequency analysis and co-occurrence networks to confirm which topics cluster (e.g., 'cost' co-occurs with 'replacement' and 'politician-led projects'), helping prioritize design fixes and messaging.
Cross-segment comparisons: run age/gender comparisons to detect whether, for example, older women vs. young men differ in acceptance, map results to pilot-user targeting.
Operationalize for scale & sustainability
Export codebooks, quotes, and visualizations from Evidano to support partners’ supply-chain planning and funder reporting.
Export hierarchical codebooks, annotated quotes, and visualizations (word clouds, co-occurrence networks) for funders and local partners to support supply-chain planning.
Preserve privacy: Evidano offers PII redaction and encrypted storage and does not use your data to train third-party models, helpful when handling sensitive community transcripts.
Collect more data without a bigger field team
Deploy AI avatar interviewers via Evidano for follow-up structured interviews to reduce travel costs and speed iterative design.
Use AI avatar interviewers to run follow-up, structured interviews about specific barriers (willingness-to-pay, acceptable flow rates, repair preferences), reducing travel costs and speeding iterative design.
Two-week reproducible workflow (run-book)
This section provides a compact two-week workflow you can run on a new prototype trial.
- Day 0–2: Gather audio, photos, field notes. Import into Evidano and attach metadata (group, age, gender, session date).
- Day 2–4: Auto-transcribe (use custom dictionary for local terms) and translate. Run initial theme extraction and quote pull (top 20 quotes by theme).
- Day 4–7: Do cross-segment frequency analysis (who mentions cost/durability/flow) and build co-occurrence network to spot linked concerns.
- Day 7–10: Create a one-page stakeholder brief with top 3 design changes, illustrative quotes, and a recommended pilot cohort.
- Day 10–14: Deploy AI-avatar follow-ups for a small subsample to test messaging and willingness-to-pay; update themes and finalize the pilot plan.
Ethics & safeguards (brief)
This section summarizes ethics and safeguards used in the study and recommended practices for handling community data.
This PLOS study kept transcripts restricted for confidentiality; similarly, research teams should secure consent for audio use and storage.
Evidano supports PII redaction and encrypted storage, use these where community expectations or IRB requirements apply.
- Non-diagnostic: findings about health outcomes are research-focused and not clinical advice.
- Local-language fidelity: retain original-language quotes where possible and document translation choices to preserve meaning.
FAQ: point-of-use water filtration
What were the main user concerns about the prototype?
+
The main user concerns were cost, slow filtration flow, durability, and replacement parts.
The study reported that participants were excited about avoiding boiling and improving child health but repeatedly cited slow flow, fragility, and replacement costs as barriers to daily use.
How was the study conducted and who participated?
+
The study was a qualitative focus-group design conducted April–May 2024 with six group discussions totaling 36 participants, transcribed and translated from Runyankore to English.
Local facilitators ran the FGDs, recordings were transcribed and translated, and four team members used an Excel coding template for analysis, as reported in the June 24, 2026 publication.
Can AI accelerate analysis of these focus groups?
+
AI can accelerate analysis by auto-transcribing, translating, extracting themes, and enabling cross-segment comparisons to prioritize design and program actions.
Using automated tools speeds quote extraction, frequency counts, co-occurrence mapping, and iterative follow-ups without enlarging the field team.
How should teams handle privacy and consent for audio and transcripts?
+
Teams should secure consent for audio use and storage, restrict access to transcripts, and apply PII redaction when required.
The PLOS study restricted transcripts for confidentiality; similarly, use encrypted storage and redaction tools and document translation choices to preserve meaning.
Wrapping up: next moves
This section summarizes the study's implication and provides next-step links for trialing the workflow.
The PLOS June 24, 2026 study (PLOS One) shows that technical performance alone won't guarantee uptake of POU filters, time, cost, repairability, and trust matter as much.
Ready to convert FGDs into prioritized product and program actions? Start a trial on Evidano to auto-transcribe, translate, and run thematic and cross-segment analyses; export stakeholder-ready briefs and visualizations in days, not weeks.
Strong CTA: Try Evidano for free
Topics
- point-of-use water filtration
- qualitative analysis
- focus groups Uganda
- AI qualitative research
- POU filter adoption
Keep reading
- Commentary on NewsAI Qualitative Research: Lessons From RANDRAND's 2026 dissertation explains AI qualitative research, worker outcomes, and medical coding, with evidence synthesis guidance from Evidano for research teams.
- Commentary on NewsAI Qualitative Research: Lessons from RANDLearn what RAND’s 2005-2023 evidence reveals about AI qualitative research, coder training, and worker outcomes, then turn interviews into usable insight with Evidano.
- Commentary on NewsKey Findings: Antibiotic Stewardship in PoultryLearn how qualitative research reveals antibiotic stewardship barriers in Madagascar’s poultry sector and how AI analysis helps researchers act faster.