Geographic Variation in Process and Outcomes of Care for Patients With Acute Myocardial Infarction in China From 2001 to 2015

This cross-sectional study evaluates geographic variation and its change with time in treatment process and outcomes for patients with acute myocardial infarction in China. Question What are the variations in process of care and outcomes for patients with acute myocardial infarction in China by geographic area? Findings In this cross-sectional study of 27046 hospitalizations from 2001 to 2015, patients in the region with worst performance were 17% less likely to receive optimal treatments comparing with the region with the best performance. Patients experienced a higher risk of in-hospital death by 46% in the region of highest mortality rate than in that of lowest mortality rate. Meaning In this study, geographic variation in care delivery and in-hospital outcomes persisted, suggesting the need for more targeted research and investment in different geographic regions. Importance Variations across regions for managing acute myocardial infarction (AMI) in China are little understood. Objective To evaluate geographic variation and its change with time in treatment process and outcomes for patients with AMI. Design, Setting, and Participants This cross-sectional study used data from the Patient-Centered Evaluative Assessment of Cardiac Events-Retrospective AMI project in 2001, 2006, 2011, and 2015 in 153 randomly selected hospitals across China. Patients were hospitalized for AMI. Data were analyzed from October 1 to October 31, 2019. Exposures Hospitalization in 3 geographic regions (Eastern, Central, and Western) stratified according to China's official definition. Main Outcomes and Measures Process of care measures included reperfusion therapies, aspirin, clopidogrel, beta-blockers, angiotensin-converting enzyme inhibitors or angiotensin receptor blockers, and statins. Therapy use was analyzed among patients who were clinically eligible without contraindications (considered as ideal candidates for treatments). Outcome measures included in-hospital mortality and 5-day mortality. Mixed models were used to assess the regional disparities and time-region interactions in those measures, adjusting for patient characteristics. Results In 153 hospitals across China, 27046 patient hospitalizations for AMI were sampled. There was a significant difference across regions in process of care and the odds ratio (OR) of delivering any 1 of the 6 treatments to an ideal patient was 0.83 (95% CI, 0.76-0.91; P < .001) for the lowest region compared with the highest region. The variation between the 2 higher regions narrowed (time-by-Eastern region interaction: OR, 0.83; 95% CI, 0.76-0.91; P < .001). The region with the highest in-hospital mortality had 1.46 times greater in-hospital mortality (95% CI, 1.07-2.00; P < .001) than the lowest region and the region with the highest 5-day mortality had 1.52 times greater 5-day mortality (95% CI, 1.09-2.11; P = .04) than the lowest region. The geographic variation in mortality did not change over time. Conclusions and Relevance In this study, significant geographic variations in process of care and outcomes were found to persist in China; further targeted and region-based approaches to AMI management are warranted.