What AI Thematic Analysis Reveals About UTI Care

According to the General practitioners’ perspectives on treatment decisions and interprofessional care for women with uncomplicated urinary tract infections: An exploratory qualitative study, antibiotic-sparing UTI care depends on clinical judgments, patient expectations, trust, workload, practice staff, and pharmacy relationships rather than on a single prescribing decision.

AI thematic analysis can help healthcare researchers organize interview transcripts into recurring themes, compare professional perspectives, and trace how qualitative evidence supports implementation decisions.

• The PLOS One study published on November 11, 2026, analyzed interviews with 15 general practitioners in Switzerland.
• The 15 interviews conducted between June 10 and July 17, 2025, lasted an average of 44 minutes, with interview lengths ranging from 32 to 54 minutes.
• The researchers identified six themes covering prescribing thresholds, trust and expectations, practical constraints, practice staff, professional autonomy, and pharmacy-based care.
• The study reported that 53.3% of participating GPs were female, the mean participant age was 46.6 years, and GP experience ranged from 5 to 25 years.
• The study’s central implication is that clearer roles, appropriate training, and feasible communication structures may support coordinated antibiotic-sparing care.

The PLOS One study found that general practitioners viewed antibiotic-sparing treatment as conditional rather than automatic. Participants considered symptom severity, symptom duration, diagnostic findings, patient expectations, trust, access to follow-up, time pressure, and the availability of other healthcare professionals.

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

The study identified six themes: differing thresholds for antibiotic prescribing; trust and established expectations; practical constraints; the formal and informal influence of practice staff; disagreement about professional autonomy; and the partial visibility of pharmacy-based care.

One GP quoted in the study summarized the implementation gap by saying, “aspiration and reality are very far apart.” Another participant explained that “shared decision-making becomes somewhat difficult” when patients believe that only antibiotics will help.

The study also reported an important limitation for qualitative interpretation: the full interview transcripts were not publicly released because detailed data could permit deductive identification of participants despite pseudonymisation.

Healthcare qualitative researchers should analyze treatment decisions as processes distributed across people, settings, and time. The PLOS One findings show that a GP’s decision may be influenced by prior self-treatment, patient expectations, practice staff triage, weekend access, and pharmacy interactions.

Researchers should preserve contradictory accounts instead of forcing consensus. The study reported that some GPs supported expanded roles for practice staff while retaining physician oversight, whereas others were more open to independent management after training and legal clarification.

Researchers should also distinguish visibility from frequency. The PLOS One authors reported that GPs often heard about pharmacy cases that persisted or required referral, while successfully resolved pharmacy encounters could remain invisible to them. An interview analysis that counts only reported problems could therefore overrepresent negative experiences.

• Code clinical factors separately from relational, practical, organizational, and interprofessional factors.
• Compare themes by participant characteristics, practice model, professional role, or care setting when the dataset supports those comparisons.
• Link each synthesized theme to supporting quotations so that readers can audit how interpretations were developed.
• Record divergent cases and uncertainty, especially when a study concerns professional autonomy or patient safety.
• Treat AI-generated themes as analytical support, not as a replacement for researcher judgment, reflexivity, or clinical governance.

The 2026 PLOS One study demonstrates how qualitative interviews can reveal the clinical, relational, organizational, and interprofessional forces behind healthcare decisions. AI thematic analysis can help healthcare researchers process transcript collections, compare participant groups, visualize code relationships, and preserve a clear link between themes and evidence.

A defensible AI-assisted workflow still requires human review, careful handling of sensitive data, and transparent reporting of how themes were developed. Researchers analyzing interviews about antimicrobial stewardship, primary care, or patient pathways can use Evidano to make synthesis more structured and auditable.

Start your next qualitative healthcare analysis with Try Evidano for free.