Abstract TP318: The Hospital Level Variation in Interhospital Transfer in Ischemic Stroke

Abstract

Background: Interhospital transfer of patients with stroke to higher level of care is a resource intensive practice. This study aimed to understand the patterns of interhospital transfer in the context of hospital characteristics. Methods: This study included Medicare fee-for-service beneficiaries aged ≥65 years who were hospitalized in 2012 for ischemic stroke and underwent interhospital transfers. The data obtained from the American Hospital Association Annual Survey were linked to the 2012 Medicare inpatient and outpatient files. This study included patients admitted to the hospitals which were categorized as “general hospitals” with emergency departments. Hospitals were classified into receiving (high transfer in rate), sending (high transfer out rate), low flow (low transfer in or out rates), and high flow (both high transfer in and out) hospitals. Pearson’s chi-square tests were used for categorical variables and Wilcoxon Rank-Sum tests for continuous variables. Results: Interhospital transfers for ischemic stroke occurred in 2876 out of 4198 hospitals (68.5%), and 5.7% of ischemic stroke admissions (19,283 of 338,306 admissions). Using national average of 5.7% as cut off, the four hospital groups : 411 receiving hospitals (14.3%), 559 sending hospitals (19.4%), 1863 low-flow hospitals (64.8%). Receiving hospitals were larger than low-flow and sending hospitals by the number of beds (Median 371, 189, and 88, respectively, p<0.001) and by annual stroke volume (median 205, 86, and 26, respectively, p<0.001). The majority of receiving (75%) and low-flow hospitals (54%) were in the Metropolitan area, while sending hospital were more evenly distributed in both urban and rural area. Higher proportion of teaching hospitals were in receiving hospitals(28%) compared to low-flow(6%) and sending hospitals (1%) with p<0.001. Higher proportion of receiving (75%) and low-flow (47%) hospitals were certified stroke centers, compared to sending hospitals (16%) with p<0.001. Conclusions: The national patterns of interhospital transfer for ischemic stroke varies depending on the hospital size, geographical location, academic status, and stroke certification. Further study of the associated outcomes will aid in health care resource utilization.