Abstract
BACKGROUND: The quality of healthcare at the regional level is of great interest. Regional differences in the quality of healthcare for cardiovascular diseases have been pointed out in previous research, but the regional factors that influence the quality remain unclear. METHODS: Using the national Diagnosis Procedure Combination (DPC) database, risk-adjusted mortality (RAM) in each secondary medical area (SMA) was derived as an indicator for the quality of the healthcare for patients with acute myocardial infarction (AMI). Factors associated with RAM were analysed using the partial least squares (PLS) regression model. RESULTS: There is a wide variation in RAM in the SMAs; the maximum value was 0.593 and the minimum value was 1.445. The PLS regression identified two components positively correlated with RAM. The first component (Component 1) positively correlates with the proportion of the elderly within the population and negatively correlates with the number of medical facilities per area, the population density, and the intra-SMA access to centres with a high volume of emergency percutaneous coronary interventions (ePCI). The second component (Component 2) positively correlates with the number of medical facilities per area and negatively correlates with the number of doctors per person and the intra-SMA access to centres with a high volume of ePCI. CONCLUSIONS: There was wide regional variation in the quality of in-hospital AMI treatments. Results suggested the degree of rurality, the sufficiency of medical resources, the access to high-volume ePCI centres, and coordination of healthcare delivery were associated with healthcare quality.