Variation between hospitals and reviewers in detection of adverse events identified through medical record review in Korea.

Objectives: This study utilized the method of medical record review to determine characteristics of adverse events that occurred in the inpatient units of hospitals in Korea as well as the variations in adverse events between institutions.

Design: A two-stage retrospective medical record review was conducted. The first stage was a nurse review, where two nurses reviewed medical records of discharged patients to determine if screening criteria had been met. In the second stage, two physicians independently reviewed medical records of patients identified in the first stage, to determine whether an adverse event had occurred.

Setting: Inpatient units of six hospitals.

Participants: Medical records of 2 596 patients randomly selected were reviewed in the first stage review.

Intervention(s): N/A.

Main Outcome Measure(s): Adverse events.

Results: A total of 277 patients (10.7%) were confirmed to have had one or more adverse event(s), and a total of 336 adverse events were identified. Physician reviewers agreed about whether an adverse event had occurred for 141 patients (5.4%). The incidence rate of adverse events was at least 1.3% and a maximum of 19.4% for each hospital. Most preventability scores were less than four points (non-preventable), and there were large variations between reviewers and institutions.

Conclusions: Given the level of variation in the identified adverse events, further studies that include more medical institutions in their investigations are needed, and a third-party committee should be involved to address the reliability issues regarding the occurrence and characteristics of the adverse events.