CareTransition-Audit: A Benchmark to Audit Discharge Summaries for Efficient Care Transitions
Researchers have developed a new benchmark called CareTransition-Audit to evaluate how well large language models can audit clinical discharge summaries. The benchmark, which uses the MIMIC-IV database and clinician-provided labels, assesses documentation completeness and agreement with human experts. While current LLMs show moderate agreement with clinicians, they struggle to identify ambiguous information, indicating a need for further development in automated clinical documentation quality improvement. AI
IMPACT This benchmark could accelerate the development of LLMs for clinical documentation auditing, improving patient safety and care transitions.