AI-enabled Qualitative Research: LLM Adoption in Primary Care

According to the PLOS One protocol (Pokharel et al., 2026), published on August 7, 2026, the study will use semi-structured interviews with primary care physicians in British Columbia to examine perceptions of LLM adoption (PLOS One).

According to Pokharel et al. (2026), the study uses the technology adoption behavior (TAB) framework and combined deductive and inductive thematic analysis to generate policy-relevant insights.

• Pokharel et al. (2026) expect recruitment and data collection from August 1, 2025 to January 31, 2026 and reported the study design status on August 7, 2026.
• Pokharel et al. (2026) cite that the number of Canadians without a family doctor rose from 4.5 million in 2019 to about 6.5 million in 2023, framing primary care as a priority setting for LLM evaluation.
• Pokharel et al. (2026) will purposively sample physicians by self-reported LLM use: 50+ uses, 5–50 uses, and fewer than 5 uses, to ensure contrasting perspectives.
• Pokharel et al. (2026) note strengths and limits: “The semi-structured interview may reveal the interviewees’ internal views but limits the variety of insights gleaned, ” a caution the authors list under limitations.

Answer: PLOS One published a qualitative study protocol on August 7, 2026 that specifies how primary care physicians’ perceptions of LLM adoption will be sampled, interviewed, and analyzed.

According to Pokharel et al. (2026) in PLOS One, the study uses a qualitative descriptive design with purposive sampling across usage strata to capture contrasting clinician experiences.

According to Pokharel et al. (2026), interviews will be semi-structured, recorded, transcribed verbatim, deidentified, and analyzed using both deductive codes derived from the TAB framework and inductive codes emerging from the data.

According to Pokharel et al. (2026), the interdisciplinary team plans reflexive memos and monthly team discussions to refine the codebook and move from codes to broader themes.

Answer: The PLOS One protocol shows how to combine a theory-driven interview guide with purposive sampling and mixed deductive/inductive coding to produce policy-relevant themes.

According to Pokharel et al. (2026), using the TAB framework ties interview prompts directly to constructs such as perceived ease of use, perceived usefulness, and task-technology fit, which helps teams develop a focused codebook before fieldwork begins.

According to Pokharel et al. (2026), purposive sampling across experience strata (50+, 5–50, <5 uses) supports analytic comparisons that reveal how familiarity with LLMs shifts perceived benefits and risks.

• Design tip: Use a pre-specified theoretical code set to speed initial coding and enable cross-case matrices, as modeled in the PLOS One protocol.
• Data quality tip: Record pilot interviews and refine prompts early, as Pokharel et al. (2026) pilots the first two interviews to ensure smooth flow.
• Policy translation tip: Map emergent themes to practice and regulatory levers, following Pokharel et al. (2026) plan to disseminate to policymakers and professional organizations.

According to Pokharel et al. (2026) in PLOS One, a transparent, theory-driven qualitative protocol can produce actionable themes to guide safe and effective LLM integration into primary care.

Researchers who adopt AI-enabled qualitative research workflows can speed transcription, maintain deidentification, and run cross-segment analyses that mirror the TAB-guided contrasts in the PLOS One protocol.

If you want to operationalize a protocol like Pokharel et al. (2026) and scale from interview to policy-ready themes, try tools that combine secure transcription, iterative coding, and cross-segment visualizations.

Get started: Try Evidano for free