MP36: Can one emergency physician improve department flow? A proof-of-concept trial of a physician float role

Introduction: Emergency departments (EDs) are overcrowded and patient acuity and volumes are ever-increasing. While changes to the flow of ED patient input and output are outside the control of frontline ED teams, the efficiency of ED throughput can be optimized. One widely studied intervention is the implementation of a physician liaison role to assist in managing overall ED flow. The Physician Float (PF) acts as a triage liaison, second physician for resuscitations, ED procedural sedation physician, and fields ED referral calls. This is a first-iteration proof-of-concept trial to plan, implement and evaluate if the PF role could decrease ED length of stay (LOS) by a goal of 30 minutes, over a four-week period, without adverse changes to left without being seen (LWBS) and bounce-back rates. Methods: The PF role was implemented as a scheduled emergency physician shift in the fall of 2017. Ongoing iterations of this role implementation are being reviewed for re-implementation. The primary outcome measure was ED LOS; secondary outcomes included time-to-physician initial assessment (PIA), EMS offload rates, and LWBS and 72-hour bounce-back rates. Qualitative data including patient concerns and physician feedback were also collected. Data were collected after the trial from a centralized, de-identified ED information system database with time-stamp quantifiers and compared to the following four-week time period where the shift is a regular ED physician shift at the same time. The ED physician and nursing team planned and implemented the PF role, then results were evaluated and shared with the wider ED staff in departmental grand rounds and quality council presentation formats, and recommendations were gathered from to adjust and strengthen future iterations of PF role implementation. Results: Descriptive statistics and Mann-Whitney and Median tests were calculated. On average there were 185 daily ED visits in the trial and comparison periods. Median ED LOS decreased by 12 minutes in the PF trial period (p<0.05). Furthermore, there was a 12 minute decreased ED LOS for all discharged patients (p<0.05). PIA time decreased by 13 minutes for patients that were admitted. The average percentage of EMS offloads within 60 min improved from 75% to 80.7% for admitted patients. LWBS and 72-hour bounce-back rates were unchanged. No additional patient concerns arose related to or during the trial. Physician feedback on the PF role was mainly positive. Conclusion: The defined role of a PF in an ED can decrease ED LOS, albeit not achieving the desired 30-minute reduction on the first iteration, this trial supported proof-of-concept for implementation of a PF role in a tertiary care centre ED. Further iterations are needed to evaluate the scalability and sustainability of this role.