AI Qualitative Analysis in Healthcare: UTI Care

According to the PLoS ONE study, “General practitioners’ perspectives on treatment decisions and interprofessional care for women with uncomplicated urinary tract infections”, antibiotic-sparing UTI care depends on more than the GP’s prescribing decision: patient expectations, trust, workload, practice staff, and pharmacy communication also shape care.

The study shows why AI qualitative analysis in healthcare is useful for primary care research: researchers can use thematic analysis to organize complex interview evidence while retaining quotations, divergent views, and the wider care pathway.

• Between June 10 and July 17, 2025, the researchers interviewed 15 GPs in Switzerland, and the interviews averaged 44 minutes.
• The November 11, 2026 study generated six themes about antibiotic prescribing, relationships, practical constraints, practice staff, professional autonomy, and pharmacy-based care.
• The participating GPs had a mean age of 46.6 years, 53.3% were female, and their general-practice experience ranged from 5 to 25 years.
• The study found that antibiotic-sparing intentions were conditional because symptom severity, patient expectations, trust, time pressure, and access to reassessment affected treatment decisions.
• The authors concluded that clearer professional roles, appropriate training, and feasible communication structures may support coordinated antibiotic-sparing care.

The 2026 PLoS ONE study found that GPs treated antibiotic-sparing management as a negotiated and context-dependent process rather than a simple prescribing choice.

The researchers used virtual semi-structured interviews, verbatim transcription, and thematic analysis informed by Braun and Clarke’s framework. The analysis combined predefined interview domains with inductive codes developed from the interview material.

The study’s six themes connected clinical reasoning with relationships and workflow. GPs described different thresholds for prescribing antibiotics, the influence of patient trust and expectations, the effect of workload and weekends, the role of practice staff, disagreement about staff autonomy, and limited visibility into pharmacy-based care.

One GP quoted in the study summarized the implementation gap as follows: “By now, the antibiotic-sparing approach is actually the approach of choice in our practice... However, it is still often the case that aspiration and reality are very far apart.”

The study also reported that interviews stopped after 15 participants because later interviews produced no substantively new content relevant to the research questions. The authors cautioned that the sample consisted of teaching physicians affiliated with the University of Zurich, so the findings are not necessarily representative of all GPs.

For healthcare researchers, the study indicates that UTI treatment decisions should be analyzed across the full primary care pathway rather than only at the point of prescribing.

The November 11, 2026 PLoS ONE findings show that a transcript may contain several interacting layers of evidence: clinical thresholds, patient communication, organizational pressure, delegated work, professional boundaries, and information exchange with pharmacies.

• Researchers should code both explicit treatment rationales and contextual factors, such as workload, weekend access, and previous patient experiences, because the PLoS ONE participants connected these factors with prescribing decisions.
• Researchers should preserve contrasting accounts because the PLoS ONE study found that GPs differed on how much autonomy practice staff should have.
• Researchers should distinguish visibility from frequency because the PLoS ONE authors noted that GPs may hear more about unsuccessful pharmacy-based care than successful cases.
• Researchers should connect quotations to themes and participant characteristics because the PLoS ONE study used verbatim interview evidence to illustrate variation in clinical reasoning.
• Research teams should treat AI-generated themes as analytical support rather than clinical conclusions because the PLoS ONE study’s interpretations depended on contextual judgment, reflexive discussion, and review of divergent accounts.

The study includes an important data-governance lesson: the full interview transcripts were not publicly released because detailed qualitative data could permit deductive identification despite pseudonymization. Healthcare research teams should therefore assess privacy, consent, and access controls before using AI tools with sensitive transcripts.

The November 11, 2026 PLoS ONE study demonstrates that antibiotic-sparing UTI care is shaped by clinical judgment, patient expectations, workload, practice roles, and communication across primary care.

The study also shows why healthcare qualitative researchers need structured thematic analysis that preserves participant context and competing perspectives. AI can accelerate transcript organization and comparison, but human researchers must retain responsibility for interpretation, ethics, and validation.

Evidano brings thematic analysis, cross-segment comparisons, secure document workflows, and AI chat together for teams working with interviews, surveys, and healthcare documents. Try Evidano for free and turn complex qualitative data into evidence your research team can review and act on.