Primary care teams need timely, grounded evidence on clinician attitudes to large language models, and the primary keyword here is physician perceptions of llms qualitative analysis. The PLoS One protocol by Pokharel et al., 2026 sets out a structured plan to interview family physicians in British Columbia and to analyse their views using the Technology Adoption Behavior (TAB) framework, providing a reproducible roadmap for qualitative researchers and health system decision makers. This post translates that protocol into practical steps for AI-enabled qualitative research teams and points to how modern tools shorten the path from audio recording to validated themes and cross-segment insights.
Key Takeaways
The PLoS One protocol by Pokharel et al. (2026) PLoS One describes a qualitative study to document how primary care physicians perceive adopting LLMs, using semi-structured interviews and the Technology Adoption Behavior (TAB) framework.
- The study protocol was published on August 7, 2026 and lists key timeline dates: received July 18, 2025; accepted July 22, 2026, according to Pokharel et al., 2026.
- The authors plan purposive sampling with LLM-use strata defined as 50+ uses, 5–50 uses, and fewer than 5 uses, and recruitment is expected from August 1, 2025 to January 31, 2026, according to Pokharel et al., 2026.
- The protocol highlights a population context: the number of Canadians without a family physician rose from 4.5 million in 2019 to about 6.5 million in 2023, as cited in Pokharel et al., 2026.
- The study will combine deductive TAB-based coding with inductive theme generation, and the authors note that "summary of events in the everyday terms of those events" is the goal of their qualitative descriptive approach (Pokharel et al., 2026).
What Happened and How the Study Works
Pokharel et al., 2026 published a study protocol to explore primary care physician perceptions of LLM adoption using semi-structured interviews and the TAB theoretical framework.
According to Pokharel et al., 2026, the research team will purposively sample practicing family physicians in British Columbia with varying self-reported LLM experience (defined as 50+, 5–50, and <5 lifetime uses in clinical contexts).
According to Pokharel et al., 2026, interviews will be recorded, transcribed verbatim, deidentified, and analysed using both deductive codes drawn from the TAB framework and inductive codes developed from the data, with NVivo used to conduct the analysis.
According to Pokharel et al., 2026, the TAB framework integrates perceptions (perceived ease of use and perceived usefulness) with task-technology fit traditions to explain uptake, and the study team will write reflexive memos and hold multidisciplinary discussions to check interpretations.
Findings Snapshot
| Date | Metric | Value | Implication |
|---|---|---|---|
| July 18, 2025 | Protocol received | Submission date | Marks start of peer review timeline in Pokharel et al., 2026 |
| July 22, 2026 | Protocol accepted | Acceptance date | Protocol refined and approved for publication as reported in Pokharel et al., 2026 |
| August 7, 2026 | Protocol published | Public availability | Enables other teams to adopt the TAB-guided interview and coding approach |
| Aug 1, 2025 to Jan 31, 2026 | Planned recruitment window | 6 months | Expected window to reach sufficiency in interview-based sampling per Pokharel et al., 2026 |
| 2019 vs 2023 | Patients without a family doctor | 4.5 million → ~6.5 million | Contextual pressure on primary care capacity cited in Pokharel et al., 2026 |
Implications for primary care researchers
The protocol signals that qualitative protocols can and should be theory-guided: Pokharel et al., 2026 position the TAB framework to produce transferable themes rather than purely descriptive lists.
According to Pokharel et al., 2026, using purposive stratification by prior LLM exposure (50+, 5–50, <5 uses) helps researchers capture both adopters and sceptics, which improves analytic generalizability.
According to Pokharel et al., 2026, combining deductive TAB codes with inductive coding and monthly team codebook reviews supports reflexivity and reduces bias in theme development.
Researchers planning LLM perception studies should predefine use strata, record contextual system pressures (for example, the 2019 to 2023 primary care access shift noted in Pokharel et al., 2026), and budget for iterative memoing and multidisciplinary adjudication.
How Evidano Helps: AI-enabled qualitative research mapped to the protocol
Problem: Slow transcription and messy raw data → Solution: Fast, compliant transcription
Evidano is an AI-powered qualitative data analysis platform that helps researchers analyze interviews, open-ended surveys, and documents.
Pokharel et al., 2026 specify verbatim transcription and deidentification as core steps; Evidano offers automated transcription with a custom dictionary and PII redaction to match those needs, reducing transcription time from days to hours.
Use Evidano's transcription features to keep consistent labels for LLM tool names and clinical terms that the protocol suggests clarifying at interview start.
Problem: Combining theory-driven and emergent coding → Solution: Thematic + deductive-inductive workflows
Pokharel et al., 2026 plan a hybrid deductive and inductive coding process using NVivo; Evidano supports the same workflow with hierarchical codebooks, code frequency tables, co-occurrence networks, and exportable codebooks for monthly team reviews.
Evidano's AI-assisted code suggestions accelerate initial code generation from the TAB framework while preserving manual adjudication for reflexive memos.
Problem: Cross-segment analysis (LLM-use strata) → Solution: Cross-segment and visualization tools
Pokharel et al., 2026 stratify participants by LLM exposure; Evidano produces segmented frequency analyses and side-by-side theme comparisons, enabling researchers to quantify which themes dominate the 50+, 5–50, and <5 use groups.
Evidano also provides exportable visuals for policy briefings and stakeholder dissemination, matching the protocol's stated dissemination goals.
Relevant platform pages
See features and technical details at Evidano Features.
For secure transcription and PII handling see Evidano Speech-to-Text and Evidano Data Security.
FAQ: physician perceptions of llms qualitative analysis
How will the study sample physicians and why does that matter?
Answer: The study will purposively sample family physicians in British Columbia across three self-reported LLM-use strata to capture diverse perspectives.
Supporting detail: Pokharel et al., 2026 define the strata as 50+ uses, 5–50 uses, and fewer than 5 uses in clinical contexts, and they justify purposive sampling as a route to analytic sufficiency and variability.
What is the Technology Adoption Behavior (TAB) framework in this protocol?
Answer: The TAB framework integrates perceived ease of use and perceived usefulness with task-technology fit traditions to explain uptake.
Supporting detail: Pokharel et al., 2026 describe TAB as unifying perception-focused models like TAM and environment-fit models like Task-Technology Fit to guide both interview questions and deductive coding.
How will analysis combine deductive and inductive coding?
Answer: The team will start with TAB-derived deductive codes and iteratively add inductive codes that emerge from interviews, with monthly codebook reviews and reflexive memos.
Supporting detail: Pokharel et al., 2026 state that primary coding will be performed by a trained team member and that all code adjustments will be discussed with the senior author and the full team.
When will the study results be available?
Answer: The protocol expects recruitment and data collection to occur from August 1, 2025 to January 31, 2026 and anticipates producing results within one year after data collection begins, according to Pokharel et al., 2026.
Supporting detail: Pokharel et al., 2026 list the projected timelines explicitly in their 'Study status' section.
Conclusion & Next Steps
The PLoS One protocol by Pokharel et al., 2026 provides a clear, theory-guided blueprint for studying primary care physicians' perceptions of LLM adoption and for producing policy-relevant themes.
Researchers can mirror the protocol's purposive strata, hybrid coding, and reflexivity practices while using AI-enabled workflows to shorten transcription-to-theme timelines and preserve analytic rigor.
If you plan to run semi-structured interviews on clinician attitudes, consider tools that support secure transcription, hierarchical codebooks, and cross-segment visualizations to match the protocol's aims.
To accelerate your own TAB-aligned qualitative project, Try Evidano for free.
Topics
- physician perceptions of llms qualitative analysis
- qualitative analysis of llm adoption
- llm adoption in primary care
- ai-enabled qualitative research
Keep reading
- Commentary on NewsAI-Enabled Qualitative Analysis of LLM AdoptionHow primary care physicians view LLMs and what qualitative researchers need to know about LLM adoption. Practical methods and AI-assisted analysis advice.
- 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.