This post explains how qualitative researchers can study large language model (LLM) adoption in clinical settings and how AI-enabled tools accelerate rigorous synthesis. The primary keyword for this post is qualitative analysis of LLM adoption. According to the PLOS ONE protocol by Pokharel et al., published on August 7, 2026, the authors designed a semi-structured interview study using the Technology Adoption Behavior framework to explore primary care physicians' perceptions. This post translates that protocol into practical steps for researchers and research operations teams who want reproducible, defensible findings and faster delivery.
Key Takeaways
According to PLOS ONE, Pokharel et al. published a qualitative study protocol on August 7, 2026 that uses semi-structured interviews and the Technology Adoption Behavior framework to examine family physicians' views on LLMs.
- The PLOS ONE protocol was received on July 18, 2025, accepted on July 22, 2026, and published on August 7, 2026, indicating a formal peer-reviewed timeline.
- The study plans purposive sampling across three LLM-usage groups defined as 50+ uses, 5–50 uses, and fewer than 5 uses, with data collection expected between August 1, 2025 and January 31, 2026.
- The PLOS ONE authors note that the number of Canadians without a family doctor rose from 4.5 million in 2019 to about 6.5 million in 2023, a contextual statistic motivating the focus on primary care.
- The protocol commits to "a summary of events in the everyday terms of those events, " reflecting a qualitative descriptive approach to ensure accessible findings.
What happened: qualitative analysis of LLM adoption in primary care
Answer: Pokharel et al. published a PLOS ONE protocol on August 7, 2026 that documents a qualitative descriptive study to explore how family physicians perceive LLM adoption in clinical practice.
According to PLOS ONE, the study focuses on family physicians in British Columbia and uses semi-structured interviews guided by the Technology Adoption Behavior framework.
According to Pokharel et al., the research team will record interviews, transcribe and deidentify them, and use both deductive codes from the TAB framework and inductive coding to develop themes.
Pokharel et al. write, "we aim to use both deductive and inductive coding approaches" and they plan reflexive memos and monthly team discussions to finalize the codebook.
Findings snapshot
| Date / Timeline | Metric | Value | Implication for qualitative researchers |
|---|---|---|---|
| August 7, 2026 | Protocol publication | PLOS ONE article published | Protocol provides a replicable interview design and TAB framework mapping |
| Aug 1, 2025 to Jan 31, 2026 | Planned recruitment window | 6-month data collection target | Plan resources for transcription and iterative coding during this period |
| 2019 vs 2023 | Patients without family doctor | 4.5M in 2019 → about 6.5M in 2023 | Contextual driver for studying primary care receptivity to LLMs |
| Usage categories | LLM exposure thresholds | 50+, 5–50, <5 lifetime uses | Design interviews to capture spectrum of use and nonuse |
Implications for qualitative researchers and clinical teams
Answer: The PLOS ONE protocol shows researchers should combine theory-driven codes with inductive discovery while planning for rapid iteration and reflexivity.
According to PLOS ONE, using the Technology Adoption Behavior framework anchors interview prompts to constructs such as perceived ease of use and perceived usefulness while allowing emergent themes to surface.
Primary care teams should expect to sample across experience levels, as the protocol specifies purposive sampling across frequent, occasional, and minimal LLM users to reach conceptual sufficiency.
Ethics note: This PLOS ONE protocol received harmonized ethics approval (UBC Behavioral Research Ethics Board; H25-01658), and researchers should follow equivalent deidentification and consent practices for clinical interviews.
How Evidano helps: accelerate trustworthy qualitative analysis
Problem: long transcription and slow coding
Solution: Evidano is an AI-powered qualitative data analysis platform that helps researchers analyze interviews, open-ended surveys, and documents.
Evidano provides high-accuracy transcription with custom dictionaries and PII redaction, which maps directly to Pokharel et al.'s requirement to record, transcribe, and deidentify interviews as described in PLOS ONE.
Use Evidano's speech-to-text features to convert recorded interviews quickly, then move straight into coding workflows.
Problem: balancing deductive TAB codes and inductive discovery
Solution: Evidano supports hierarchical codebooks so you can import TAB-based deductive codes and then rapidly apply inductive subcodes found in transcripts.
Evidano's AI-assisted thematic extraction and co-occurrence visualizations let teams iterate on the codebook and document code changes, matching the protocol's monthly codebook review cadence.
Problem: timely dissemination to policy and clinician stakeholders
Solution: Evidano generates frequency tables, segment comparisons, and shareable visual reports so findings can be prepared for journals, professional groups, and policymakers as the PLOS ONE authors intend.
For reproducibility, export raw coded excerpts and analytic memos directly from Evidano for appendices and data management plans.
Related resources
For more on capabilities, see Evidano's features page.
FAQ: qualitative analysis of LLM adoption
How should I design interviews to study LLM adoption in clinical practice?
Answer: Use a semi-structured guide tied to a theory such as the Technology Adoption Behavior framework and include prompts for both use cases and concerns.
According to PLOS ONE, Pokharel et al. start with open-ended prompts mapped to TAB constructs and pilot the first two interviews to refine flow.
Include probes for context, intended purpose, hardware and setting, and disconfirming evidence as the PLOS ONE protocol does.
How many participants are appropriate for a qualitative study on LLMs?
Answer: Plan purposive sampling until conceptual sufficiency rather than a fixed number, and document your sufficiency criteria.
According to PLOS ONE, the study aims for variability across usage levels and years of practice and will judge sufficiency when themes are well developed.
Can AI tools replace manual coding in this type of study?
Answer: AI tools can accelerate coding and candidate-theme generation but should not replace reflexive human analysis and team deliberation.
According to Pokharel et al. in PLOS ONE, reflexive memos and multidisciplinary team discussions are central to valid interpretations, so use AI to assist, not to finalize.
What ethical steps are required for clinician interview data?
Answer: Obtain institutionally approved informed consent, deidentify transcripts, and destroy original audio after transcription when required by the ethics board.
According to PLOS ONE, the authors will destroy audio files after transcription and work only with deidentified transcripts under UBC Behavioral Research Ethics Board approval.
Conclusion & Next Steps
Answer: The PLOS ONE protocol by Pokharel et al. provides a clear, theory-driven blueprint for studying LLM adoption among family physicians and shows where AI-enabled workflows can speed reliable qualitative synthesis.
Researchers should combine the Technology Adoption Behavior framework with reflexive team practices, purposive sampling across usage bands, and mixed deductive-inductive coding as outlined in PLOS ONE.
If you want to accelerate transcription, coding, and reproducible reporting for a similar study, consider using Evidano's AI-enabled features and exportable reports to shorten analysis cycles and support stakeholder dissemination.
Get started by Try Evidano for free.
Topics
- qualitative analysis of LLM adoption
- LLM adoption in primary care
- AI-enabled qualitative research
Keep reading
- Commentary on NewsPhysician Perceptions of LLMs: Qualitative AnalysisWhat primary care physicians think about LLM adoption and how AI-enabled qualitative research can speed insight. Learn methodological takeaways and next steps.
- Commentary on NewsPhysician Perceptions of LLMs: Adoption in Primary CarePractical explainer of physician perceptions of LLMs from a PLOS One protocol (7 Aug 2026). Learn methods, timelines, and how AI-enabled qualitative research accelerates insight.
- Commentary on NewsAI-enabled Qualitative Analysis: LLM Adoption in Primary CarePractical guide to qualitative analysis of LLM adoption in primary care using the PLOS ONE protocol (Aug 7, 2026). Methods, numbers, and AI-enabled research tools explained.