We propose an experiment to evaluate the use of an expert system (Iliad) to detect diagnostic errors that may lead to significant quality problems. We propose that these diagnostic errors are not detected by the current Peer Review Organization (PRO) nurse review procedures. We shall compare the review results of 500 Utah PRO cases (inpatient Medicare and Medicaid care) using Iliad and using the current nurse review procedures to determine whether: 1. Iliad correctly detects more quality problems than nurse review. 2. Iliad detects fewer """"""""false positive"""""""" quality problems than nurse review. 3. The quality problems detected by Iliad are more severe by PRO standards than the quality problems detected by nurse review.
