TriageSim: A Conversational Emergency Triage Simulation Framework from Structured Electronic Health Records

TL;DR

TriageSim is a novel simulation framework that generates synthetic nurse-patient triage conversations from structured EHR data, enabling research and evaluation in emergency triage communication.

Contribution

It introduces a new framework for creating controlled, multi-turn triage conversations with synthetic transcripts and audio from structured records, supporting research in emergency triage.

Findings

Generated corpus includes ~800 synthetic transcripts and audio.

Modest agreement observed for acuity levels across text, transcripts, and audio.

Automated and manual evaluations confirm linguistic, behavioral, acoustic, and medical fidelity.

Abstract

Research in emergency triage is restricted to structured electronic health records (EHR) due to regulatory constraints on nurse-patient interactions. We introduce TriageSim, a simulation framework for generating persona-conditioned triage conversations from structured records. TriageSim enables multi-turn nurse-patient interactions with explicit control over disfluency and decision behaviour, producing a corpus of ~800 synthetic transcripts and corresponding audio. We use a combination of automated analysis for linguistic, behavioural and acoustic fidelity alongside manual evaluation for medical fidelity using a random subset of 50 conversations. The utility of the generated corpus is examined via conversational triage classification. We observe modest agreement for acuity levels across three modalities: generated synthetic text, ASR transcripts, and direct audio inputs. The code, persona schemata and triage policy prompts for TriageSim will be available upon acceptance.