Geographic Modeling to Quantify the Impact of Primary and Comprehensive Stroke Center Destination Policies

Background and Purpose— We evaluated the impact of a primary stroke center (PSC) destination policy in a major metropolitan city and used geographic modeling to evaluate expected changes for a comprehensive stroke center policy. Methods— We identified suspected stroke emergency medical services encounters from 1/1/2004 to 12/31/2013 in Philadelphia, PA. Transport times were compared before and after initiation of a PSC destination policy on 10/3/2011. Geographic modeling estimated the impact of bypassing the closest hospital for the closest PSC and for the closest comprehensive stroke center. Results— There were 2 326 943 emergency medical services runs during the study period, of which 15 099 had a provider diagnosis of stroke. Bypassing the closest hospital for a PSC was common before the official policy and increased steadily over time. Geographic modeling suggested that bypassing the closest hospital in favor of the closest PSC adds a median of 3.1 minutes to transport time. Bypassing to the closest comprehensive stroke center would add a median of 8.3 minutes. Conclusions— Within a large metropolitan area, the time cost of routing patients preferentially to PSCs and comprehensive stroke centers is low.