Abstract
BACKGROUND: Although there have been studies that compared outcomes of patients with acute myocardial infarction (AMI) across countries, little focus has been placed on institutional variance of outcomes. The aim of the present study was to compare institutional variance in mortality following percutaneous coronary intervention (PCI) for AMI and factors explaining this variance across different health systems. METHODS: Data on inpatients who underwent PCI for AMI in 2016 were obtained from the National Health Insurance Data Sharing Service in Korea, the Diagnosis Procedure Combination (DPC) Study Group Database in Japan, and the National Health Insurance Research Database (NHIRD) in Taiwan. Multilevel analyses with inpatient mortality as the outcome and the hierarchical structure of patients nested within hospitals were conducted, adjusting for common patient-level and hospital-level variables. We compared the intraclass correlation coefficient (ICC) and the proportion of variance explained by hospital-level characteristics across the three health systems. RESULTS: There were 17 351 patients from 160 Korean hospitals, 29 804 patients from 660 Japanese hospitals, and 10 863 patients from 104 Taiwanese hospitals included in the analysis. Inpatient mortality rates were 6.3%, 7.3%, and 6.0% in Korea, Japan, and Taiwan, respectively. After adjusting for patient and hospital characteristics, Taiwan had the lowest variation in mortality (ICC, 1.8%), followed by Korea (2.2%) and then Japan (4.5%). The measured hospital characteristics explained 38%, 19%, and 9% of the institutional variance in Korea, Taiwan, and Japan, respectively. CONCLUSION: Korea, Japan, and Taiwan had similarly uniform outcomes across hospitals for patients undergoing PCI for AMI. However, Japan had a relatively large institutional variance in mortality and a lower proportion of variation explainable by hospital characteristics, compared with Korea and Taiwan.