UnknownNANCT07457840

Transforming Clinical Decision Support Systems: Using Continuous Bayesian Updates to Integrate AI Predictions With Clinician Expertise

Sponsor: University of California, San Francisco

Trial Details

Condition	Diagnostic Decision Making
Intervention	Bayesian-Updated Post-Test Probability; Standard AI Predicted Probability; Uncertainty Quantification (95% Confidence Interval)
Phase	NA
Status	Unknown
Participants	100
Est. Completion	1 Dec 2026
Last Updated	14 July 2026

Primary Outcome

Clinician Diagnostic Accuracy

Eligibility Criteria

Inclusion Criteria

Inclusion Criteria: * Must hold one of the following clinical roles: Nurse Practitioner (NP), Physician Assistant/Physician Associate (PA), Resident Physician, Physician Fellow, or Attending Physician * Able to complete the survey in English * Access to a computer or tablet (mobile phones are not recommended due to the visual nature of the survey) Exclusion Criteria: * Does not hold an eligible clinical role as defined above * Completes fewer than 2 of 8 clinical vignettes (less than 25% of the survey) * Has previously participated in this study * Unable to complete the survey in English

Trial Sites

Countries

United States

View Registry Entry

Related Analysis

Phase N/A Vignette Trial Opens: Bayesian-Updated AI Outputs vs Standard Predicted Probabilities in Clinical Decision Support — What Clinicians Need to Know

A new UCSF-sponsored randomised vignette trial is investigating whether presenting AI diagnostic outputs as Bayesian-updated post-test probabilities — rather than raw model predictions — improves clinician diagnostic accuracy in chest pain and dyspnea cases. The study also tests whether uncertainty quantification can protect clinicians against the anchoring effects of deliberately misleading AI predictions, with direct implications for clinical decision support system design globally.

Read full article →

This content is for educational purposes for healthcare professionals only and does not constitute clinical advice. Clinical decisions should be based on individual patient assessment, current guidelines, and appropriate specialist consultation. Editorial Standards · Privacy Policy · Terms of Service