PC1016: Derivation and Internal Validation of a Novel Clinical Workload Metric for the Pediatric Emergency Department

Background: The need to measure clinical workload in the emergency department (ED) has become especially acute during the pandemic-related healthcare crisis, with decreased resources and unpredictable surges in patient volumes. Patients have been impacted with increased wait times prior to physician evaluation, prolonged boarding times for hospitalized patients, and unacceptably high left without treatment (LWOT) rates.

Methods: We performed a retrospective observational study using administrative data from a large, urban, academic pediatric ED. For each patient between January 2018 and December 2021, we collected arrival time, disposition, and current procedural terminology (CPT) evaluation and management codes with corresponding relative value units (RVUs). We derived a novel ED workload metric by aggregating the number of RVUs for all patients in the Emergency Department at a given time. We grouped ED workload into 4-hour blocks to generate workload with units RVU*h. Finally, we performed a Poisson regression to measure the association of workload with the outcome of total patients who LWOT during each 4-hour block.

Results: During the entire study period, mean workload per 4-hour block, was 402.1 (SD 201.6) RVU*h. Of 302,825 total visits, 9035 (3.0%) LWOT. More than half of those LWOT cases were during the period of July-December 2021, when the LWOT rate was 10.8%. In Poisson regression, our novel workload metric of RVU*h showed a relatively strong association with LWOT, including a pseudo R2 of 0.38 and a coefficient of 0.0053 (95%CI: 0.0051, 0.0054). For each increase of 188.7 RVU*h, approximately equivalent to 13 hospitalized boarders remaining in the ED for an additional 4 hours, at least one patient would be expected to LWOT. Expected increases in LWOT are greater for critically ill patients boarding in the ED, given the higher RVUs assigned to these patients.

Conclusion: We developed a parsimonious metric for ED clinical workload based on RVUs, a measure that already has face validity for workload since it is the scaffold on which healthcare payments are structured in the United States. Since patients continue to contribute to aggregated RVUs for their entire ED length of stay, this simplistic measure also captures the increased workload associated with protracted inpatient boarding in the ED. The metric was internally validated as a potential predictor for LWOT, the ultimate measure of patient satisfaction, representing a judgment by families that their time is more valuable than any medical service that might be provided by extending their wait time. Additional study of the relationship between RVU*h and LWOT is necessary in other pediatric centers as well as adult ED settings. Future work to predict workload, using this metric, may offer opportunities to help with staff scheduling and other elements of ED operational planning.