- ICH GCP
- US Clinical Trials Registry
- Klinisk utprøving NCT07706920
Testing an AI Tool to Help Primary Care Clinicians With Specialty Consultation Questions (SAGE)
Physician Usability and Output Quality Evaluation of a Retrieval-Augmented Language Model System for Specialty Medical Consultation: A Human-Computer Interaction Study
The goal of this study is to test an artificial intelligence (AI) tool called SAGE. SAGE helps primary care doctors with questions that often need a specialist. Primary care doctors are the doctors people usually see first. SAGE reviews a case and suggests what a specialist might advise.
The main questions it aims to answer are:
- Do doctors make sound, timely care decisions when they use SAGE?
- Do those decisions match what a specialist would advise?
Researchers will compare the decisions doctors make with and without SAGE.
Doctors in the study will:
- Review made-up patient cases (these are not real patients)
- Make a decision for each case, their usual way and with SAGE
Studieoversikt
Status
Intervensjon / Behandling
Studietype
Registrering (Antatt)
Fase
- Ikke aktuelt
Kontakter og plasseringer
Studiesteder
-
-
California
-
Stanford, California, Forente stater, 94305
- Stanford University
-
Ta kontakt med:
- Jonathan Chen, MD, PhD
- Telefonnummer: 650-725-3655
- E-post: jonc101@stanford.edu
-
-
Deltakelseskriterier
Kvalifikasjonskriterier
Alder som er kvalifisert for studier
- Voksen
- Eldre voksen
Tar imot friske frivillige
Beskrivelse
Inclusion Criteria:
- Physician practicing in primary care internal medicine and/or family medicine
- Attending physician, or resident physician at postgraduate year 2 (PGY-2) or higher
- Stanford-affiliated (faculty, staff, or trainee)
- Able to complete a single, approximately 60-minute remote (Zoom) session conducted in English
Exclusion Criteria:
- Not currently practicing in primary care internal medicine or family medicine
- Member of the study team or otherwise involved in the development of SAGE
- Unable to complete the remote study session
Studieplan
Hvordan er studiet utformet?
Designdetaljer
- Primært formål: Helsetjenesteforskning
- Tildeling: N/A
- Intervensjonsmodell: Enkeltgruppeoppdrag
- Masking: Ingen (Open Label)
Våpen og intervensjoner
Deltakergruppe / Arm |
Intervensjon / Behandling |
|---|---|
|
Eksperimentell: Primary care physicians
Primary care physicians make management decisions on synthetic patient cases.
Each physician reviews four cases in random order: two with SAGE (AI clinical decision support) and two using a standard eConsult approach.
|
SAGE is an artificial intelligence (AI) based clinical decision support tool.
It generates specialty-informed recommendations for a primary care consultation question.
Physicians review the SAGE output while making a management decision for the synthetic case.
A standard electronic consultation (eConsult) format, without SAGE.
Physicians review this information while making a management decision for the synthetic case.
|
Hva måler studien?
Primære resultatmål
Resultatmål |
Tiltaksbeskrivelse |
Tidsramme |
|---|---|---|
|
Clinical actionability of the clinician's decision
Tidsramme: Post-session independent review, through study completion (up to approximately 2 months)
|
For each case, independent reviewers judge whether the physician's decision is clinically actionable, meaning the physician can act on it now rather than wait for a specialist.
The outcome is the proportion of cases judged actionable, measured separately for decisions made with SAGE and for decisions made the standard way.
|
Post-session independent review, through study completion (up to approximately 2 months)
|
|
Concordance of the clinician's decision with the specialist reference answer
Tidsramme: Post-session independent review, through study completion (up to approximately 2 months)
|
Agreement between the physician's management decision and a specialist reference answer, rated by independent reviewers, with inter-rater agreement summarized.
Assessed for decisions made with SAGE and for decisions made the standard way.
|
Post-session independent review, through study completion (up to approximately 2 months)
|
Sekundære resultatmål
Resultatmål |
Tiltaksbeskrivelse |
Tidsramme |
|---|---|---|
|
System usability (System Usability Scale, SUS)
Tidsramme: End of study session (Day 1, approximately 60 minutes)
|
Perceived usability of SAGE measured with the 10-item System Usability Scale (SUS), scored from 0 to 100.
|
End of study session (Day 1, approximately 60 minutes)
|
|
Order acceptance in the SAGE condition
Tidsramme: During study session (Day 1)
|
Proportion of SAGE-suggested orders that the clinician accepts, assessed in the SAGE condition.
|
During study session (Day 1)
|
|
Time to complete case review
Tidsramme: During study session (Day 1)
|
Time (in minutes) taken to review each case and reach a management decision, compared between the SAGE and standard eConsult conditions.
|
During study session (Day 1)
|
Samarbeidspartnere og etterforskere
Sponsor
Etterforskere
- Hovedetterforsker: Jonathan Chen, MD, PhD, Stanford University
Studierekorddatoer
Studer hoveddatoer
Studiestart (Antatt)
Primær fullføring (Antatt)
Studiet fullført (Antatt)
Datoer for studieregistrering
Først innsendt
Først innsendt som oppfylte QC-kriteriene
Først lagt ut (Faktiske)
Oppdateringer av studieposter
Sist oppdatering lagt ut (Faktiske)
Siste oppdatering sendt inn som oppfylte QC-kriteriene
Sist bekreftet
Mer informasjon
Begreper knyttet til denne studien
Andre studie-ID-numre
- 85430
Plan for individuelle deltakerdata (IPD)
Planlegger du å dele individuelle deltakerdata (IPD)?
Legemiddel- og utstyrsinformasjon, studiedokumenter
Studerer et amerikansk FDA-regulert medikamentprodukt
Studerer et amerikansk FDA-regulert enhetsprodukt
Denne informasjonen ble hentet direkte fra nettstedet clinicaltrials.gov uten noen endringer. Hvis du har noen forespørsler om å endre, fjerne eller oppdatere studiedetaljene dine, vennligst kontakt register@clinicaltrials.gov. Så snart en endring er implementert på clinicaltrials.gov, vil denne også bli oppdatert automatisk på nettstedet vårt. .