Clinician-grounded platform helps check AI psychiatric intake quality
Clinician-Grounded Quality Assurance for AI-Assisted Psychiatric Intake
Artificial IntelligenceComputation and Language
Summary
It can be hard to know if AI systems interviewing patients for psychiatric care are doing a good job. The authors built a special tool that lets clinicians test AI interviewers using simulated patients, so they can see how well the AI finds important information. In a small test, the AI found more key details but also made extra guesses that might not be accurate. This approach helps health systems keep an eye on AI tools to make sure they meet clinical standards.
What this means in practice
- •For health system quality teams: Test AI psychiatric intake systems against clinical standards using simulated patient interviews.
- •For mental health software developers: Use the clinician-grounded platform to evaluate and improve AI interviewer accuracy and clinical relevance.
Authors
King Shi, Amanda Li, Jonathan Ivey, Synthia Qia Wang, Guan Gui, Hyunseo Kim, Peter Zandi, Jason Straub, Jacob Taylor, Ananya Joshi
Abstract
Before patients can use AI-assisted psychiatric intake systems, health systems need practical ways to routinely evaluate these tools against their clinical standards for quality assurance. Because clinicians may use different intake styles, evaluation for this task must (1) support comparison across interviewing approaches, (2) minimize clinician burden, and (3) measure clinically relevant performance for health systems deploying these technologies. We present a clinician-grounded evaluation platform built around a memory-augmented patient simulator for open-ended AI interviewing, InterviewPlayground. We created interactive patients using InterviewPlayground with our expert-authored vignettes, constructed a simulated intake platform for the interviews, and designed evaluation modalities relevant to intake. In a pilot of 6 clinicians in a 25-minute assessment compared to a GPT-based LLM intake interviewer, the LLM recovered more of the clinically relevant items embedded in the patient vignettes (88.0% vs. 38.9%), but made more clinical inferences not based on the interview (56.8% vs. 27.8%), and characterized identified safety concerns less often (33.3% vs. 66.7%), setting the stage for deployed quality assurance for this task.