The goal of this clinical trial is to learn if an artificial intelligence (AI) tool that listens to appointments and drafts clinical notes can help clinicians in psychiatric care. Researchers also want to know if the tool changes how patients experience their care. The study includes mental health care providers (physicians, psychologists/psychotherapists, nurses, and social workers) at Psychiatrie St. Gallen and the adult patients they see in their appointments. The main questions it aims to answer are: * Does the AI tool lower the time clinicians spend writing notes? * Are notes written with the AI tool as good as or better than notes written without it? * Do patients notice a difference in their relationship with their clinician when the tool is used? Researchers will compare clinicians who use the AI tool to clinicians who keep writing notes their usual way, before and during the intervention period, to see if writing time, note quality, and the patient-rated therapeutic relationship differ between the two groups. Clinicians will: Join one of two groups. One group starts using the AI tool on August 14, 2026. The other group keeps writing notes the usual way. Both groups will be observed until the end of November 2026, with note quality evaluation completed in December 2026. Clinicians who cannot contribute to all measurements may, with their consent, remain in the study for automated measurement of documentation time only. Patients will: * Be invited to fill out a short, anonymous survey about their experience with their clinician. * Be asked to agree, in writing, that researchers may review the report from their initial assessment appointment.
Study Type
INTERVENTIONAL
Allocation
NON_RANDOMIZED
Purpose
HEALTH_SERVICES_RESEARCH
Masking
SINGLE
Enrollment
47
Listener Tool of the SwissGPT application (AlpineAI AG). After the patient's verbal consent, the provider records the consultation audio via mobile phone or laptop. All audio and text data are processed exclusively on AlpineAI's own GPU infrastructure located in Switzerland and are encrypted with individual keys per customer and per data type. Transcription uses the speech-to-text models Whisper 3 Turbo; from the transcript, a draft clinical note is generated by open-weight large language model Mistral Large 3 based on a standardized prompt developed for the institution. The provider transfers the generated draft note unedited into the electronic health record (ORBIS) and performs all review and editing directly in the EHR, so that the effort of correcting AI-generated drafts is captured by the automated documentation-time measurement. Audio recordings and transcripts are automatically deleted after 30 days.
Psychiatrie St. Gallen
Heerbrugg, Canton of St. Gallen, Switzerland
Psychiatrie St. Gallen
Pfäfers Dorf, Canton of St. Gallen, Switzerland
Psychiatrie St. Gallen
Rorschach, Canton of St. Gallen, Switzerland
Psychiatrie St. Gallen
Sankt Gallen, Canton of St. Gallen, Switzerland
Psychiatrie St. Gallen
Sargans, Canton of St. Gallen, Switzerland
Psychiatrie St. Gallen
Uznach, Canton of St. Gallen, Switzerland
Psychiatrie St. Gallen
Wattwil, Canton of St. Gallen, Switzerland
Psychiatrie St. Gallen
Wil, Canton of St. Gallen, Switzerland
Time spent on psychiatric documentation
Mean minutes spent writing clinical notes (progress notes, initial psychiatric assessment reports, and discharge letters), extracted automatically from EHR (ORBIS) audit logs as form-level editing durations, compared pre-intervention vs. during intervention between groups.
Time frame: 24 Weeks
Documentation quality of initial psychiatric asessment reports
Quality of initial psychiatric assessment reports rated with the Physician Documentation Quality Instrument, 9-item version (PDQI-9). Each of the 9 items is scored on a 5-point Likert scale (1 = strongly disagree to 5 = strongly agree) and summed to a total score ranging from 9 (minimum) to 45 (maximum); higher scores indicate better documentation quality. Each report is independently rated by 2 of 8 experienced mental health specialists in a balanced rotating-pair design, blinded to provider identity, group, time point, and site.
Time frame: 24 Weeks
Working Alliance
Patient-rated working alliance measured with the Working Alliance Inventory-Short Revised, German patient version (WAI-SR). The 12 items are each scored on a 5-point scale (1 to 5) and summed to a total score ranging from 12 (minimum) to 60 (maximum); higher scores indicate a stronger (better) working alliance. Reported anonymously by patients after outpatient consultations. In addition to the two-sided superiority test, non-inferiority is prespecified: a lower bound of the two-sided 95% CI of the difference-in-differences estimate above -3.5 WAI-SR sum-scale points will be interpreted as evidence against a clinically relevant deterioration of the alliance.
Time frame: 5 weeks pre intervention and 5 weeks during the intervention.
Time passed from initial psychiatric assessment note creation to its completion
Calendar time from the administrative creation of the initial psychiatric assessment note in the EHR to its completion, capturing documentation backlog. This outcome is separate from total active time spent writing the note.
Time frame: 24 Weeks
Guideline-relevant information coverage and density in initial psychiatric assessment reports
Coverage and density of guideline-relevant psychiatric assessment content, scored by a large language model (SwissGPT) against a 71-item checklist derived from the American Psychiatric Association practice guidelines for the psychiatric evaluation of adults. Each item is scored as covered (1), partially covered (0.5), or not covered (0); report length in characters is used to compute information density.
Time frame: 24 Weeks
Patient-perceived empathy of mental health care providers during consultations
Patient-perceived empathy measured with the Consultation and Relational Empathy (CARE) measure. The 10 items are each scored on a 5-point scale (1 = poor to 5 = excellent) and summed to a total score ranging from 10 (minimum) to 50 (maximum); higher scores indicate greater perceived empathy (better). Completed anonymously by patients alongside the WAI-SR. A study-specific German translation is used; its psychometric properties will be examined and reported.
Time frame: 5 weeks pre intervention and 5 weeks during the intervention.
Documentation quality of ambient scribe-generated initial psychiatric assessment reports (raw AI output)
AI-generated initial psychiatric assessment reports, exactly as produced by the ambient scribe and before any human correction, rated by the generating providers (intervention group only) with the Provider Documentation Summarization Quality Instrument (PDSQI-9), an LLM-adapted 9-attribute instrument screening for hallucinations, omissions, redundancy, and biased language. Each attribute is scored on a 5-point scale (1 to 5) and summed to a total score ranging from 9 (minimum) to 45 (maximum); higher scores indicate better quality of the AI-generated summary.
Time frame: 15 Weeks
Clinician Burnout
Work-related burnout of participating providers in both groups, measured with the German version of the Burnout Assessment Tool (BAT). The 23 core-symptom items (subscales: exhaustion, mental distance, cognitive impairment, emotional impairment) are each scored on a 5-point frequency scale (1 = never to 5 = always) and averaged to a core score ranging from 1.00 (minimum) to 5.00 (maximum); higher scores indicate more burnout symptoms (worse outcome). The 10 secondary-symptom items are scored and averaged on the same 1.00-5.00 metric, with higher scores likewise indicating a worse outcome.
Time frame: 5 weeks in the pre-intervention period and the last Week (Week 15) of the post-intervention period.
This platform is for informational purposes only and does not constitute medical advice. Always consult a qualified healthcare professional.