Abstract 230: The Impact of Hospital Closures on Outcomes for Myocardial Infarction and Stroke

• Published in: Circulation Cardiovascular quality and outcomes Vol. 6; no. suppl_1
• Main Authors: Joynt, Karen E, Orav, E. John, Jha, Ashish K
• Format: Journal Article
• Language: English
• Published: 01.05.2013
• ISSN: 1941-7713; 1941-7705
• DOI: 10.1161/circoutcomes.6.suppl_1.A230

Abstract only Objective: U.S. hospitals face substantial financial challenges as policymakers try to reduce healthcare spending. As a result, it is likely that hospital closures will accelerate in the coming years. Hospital closures threaten access to care, and may negatively impact patient outcomes by lengthening travel times, particularly for illnesses for which treatment depends heavily on timely receipt of treatment. Therefore, our objective was to assess the impact of hospital closures on patient outcomes for acute myocardial infarction (AMI) and stroke over the past decade. Methods: We used data from Medicare files as well as the American Hospital Association from 2002 through 2010 to identify hospital closures. For each hospital that closed, we identified Medicare fee-for-service patients that sought care at that hospital for AMI or stroke in the year prior to closure. We then determined the nearest alternative hospital using geocoding for each patient in the sample, and determined the additional travel time associated with this alternative site of care. We compared the characteristics of the closing versus alternative hospitals. We then created multivariate hierarchical logistic regression models in which we reassigned each of the patients at a hospital that closed to the nearest alternative hospital, and compared mortality rates between the closed and alternative facilities. Results: Between 2002 and 2010, we identified 121 hospital closures. There were 1,109 patients that received AMI care and 1,104 patients that received stroke care at a hospital that closed in the subsequent year. The median age was 79 for both conditions; of the AMI patients, 45% were male, and of the stroke patients, 39% were male. Travel time increased by 2.8 minutes for AMI patients and by 1.5 minutes for stroke patients. For AMI patients, the alternative hospital to which they would have traveled was more often large (0% versus 28%, p<0.001), non-profit (63% versus 79%, p<0.001), and teaching (1.1% versus 8.2%, p<0.001) compared with hospitals that closed. Patterns were similar for patients with stroke. For AMI, there was no difference in 30-day risk-adjusted mortality rates for the closed versus alternative hospitals (19.03% versus 19.17%, difference 0.14%, p=0.90). Similarly, for stroke, there was no difference in 30-day risk-adjusted mortality rates (19.57% versus 19.96%, difference 0.39%, p=0.75). Conclusions: We found no evidence that hospital closure was associated with worse clinical outcomes for AMI or stroke for patients. Though there were increased travel times for both conditions, this was likely offset by an increase in hospital quality in the alternative hospitals. These findings should provide some reassurance to those concerned that hospital closures will lead to significantly worse patient outcomes, even for conditions for which timely receipt of treatment is critical.