• Fareed Saleh, Kristen Breslin, Paul C Mullan, Zachary Tillett, and James M Chamberlain.
• From the Emergency Medicine, Children's National Health System.
• Pediatr Emerg Care. 2020 Feb 1; 36 (2): e50-e55.

ObjectiveThe aim of this study was to create a multivariable model of standardized relative value units per hour by adjusting for nonprovider factors that influence efficiency.MethodsWe obtained productivity data based on billing records measured in emergency relative value units for (1) both evaluation and management of visits and (2) procedures for 16 pediatric emergency medicine providers with more than 750 hours worked per year. Eligible shifts were in an urban, academic pediatric emergency department (ED) with 2 sites: a tertiary care main campus and a satellite community site. We used multivariable linear regression to adjust for the impact of shift and pediatric ED characteristics on individual-provider efficiency and then removed variables from the model with minimal effect on productivity.ResultsThere were 2998 eligible shifts for the 16 providers during a 3-year period. The resulting model included 4 variables when looking at both ED sites combined. These variables include the following: (1) number of procedures billed by provider, (2) season of the year, (3) shift start time, and (4) day of week. Results were improved when we separately modeled each ED location. A 3-variable model using procedures billed by provider, shift start time, and season explained 23% of the variation in provider efficiency at the academic ED site. A 3-variable model using procedures billed by provider, patient arrivals per hour, and shift start time explained 45% of the variation in provider efficiency at the satellite ED site.ConclusionsSeveral nonprovider factors affect provider efficiency. These factors should be considered when designing productivity-based incentives.

20 keywords...

hide…