Abstract
Emergency medical services (EMS) crews often wait for emergency department (ED) beds to become available to offload their patients. Presently there is no national benchmark for EMS turnaround or offload times, or method for objectively and reliably measuring this. This study introduces a novel method for monitoring offload times and identifying variance. We performed a descriptive, observational study in a large urban community teaching hospital. We affixed radio frequency identification (RFID) tags (Confidex Survivor™, Confidex, Inc., Glen Ellyn, IL) to 65 cots from 19 different EMS agencies and placed a reader (CaptureTech Weatherproof RFID Interpreter, Barcoding Inc., Baltimore, Maryland) in the ED ambulance entrance, allowing for passive recording of traffic. We recorded data for 16 weeks starting December 2009. Offload times were calculated for each visit and analyzed using STATA to show variations in individual and cumulative offload times based on the time of day and day of the week. Results are presented as median times, confidence intervals (CIs), and interquartile ranges (IQRs). We collected data on 2,512 visits. Five hundred and ninety-two were excluded because of incomplete data, leaving 1,920 (76%) complete visits. Average offload time was 13.2 minutes. Median time was 10.7 minutes (IQR 8.1 minutes to 15.4 minutes). A total of 43% of the patients (833/1,920, 95% CI 0.41–0.46) were offloaded in less than 10 minutes, while 27% (513/1,920, 95% CI 0.25–0.29) took greater than 15 minutes. Median times were longest on Mondays (11.5 minutes) and shortest on Wednesdays (10.3 minutes). Longest daily median offload time occurred between 1600 and 1700 (13.5 minutes), whereas the shortest median time was between 0800 and 0900 (9.3 minutes). Cumulative time spent waiting beyond 15 minutes totaled 72.5 hours over the study period. RFID monitoring is a simple and effective means of monitoring EMS traffic and wait times. At our institution, most squads are able to offload their patients within 15 minutes, with many in less than 10 minutes. Variations in wait times are seen and are a topic for future study.