Abstracts

Rationale: Washington University Pediatric Epilepsy Center at St. Louis Children’s Hospital (SLCH) is a Level 4 epilepsy center with an 8-bed EEG Monitoring Unit (EMU). Currently, the EMU runs at 51% room utilization with patients waiting 1-2 months for their EMU admissions.

We hypothesized we could improve the EMU’s occupancy to 80% by designing interventions to address the primary drivers of our EMU census. Increasing occupancy will allow us to care for more patients and make better use of this valuable resource.

Methods: We began by participating in the SLCH Quality Improvement Boot Camp taught by a multidisciplinary team of faculty trained in Improvement Science. We developed a Key Driver Diagram (Figure 1) with the aim of increasing the EMU occupancy from 51% to 80%.

We identified three main drivers affecting our census levels: (1) early discharges, (2) cancellations and no-shows, and (3) scheduling too few patients. We then designed two Plan-Do-Study-Act (PDSA) cycles to address preventable early discharges and cancellations.

PDSA 1 was designed to reduce preventable early discharges. We modified our EEG-Video (vEEG) order form to provide clinical decision support to providers to assist in determining the optimal number of nights to order. PDSA 2 was designed to reduce last-minute cancellations and no-shows. We instituted reminder calls to caregivers two weeks prior to the vEEG to determine if the admission was still necessary and feasible for the family.

To assess if our interventions improved the EMU occupancy, we tracked our outcome measure, Actual EMU Occupancy (Figure 2) and two process measures, nights lost to early discharge, and no-shows and cancellations. Our balancing measures included vEEG add-ons accepted and declined.

Results: Our interventions have not yet increased overall EMU occupancy (Figure 2). However, we have realized slight improvements in a few process measures. For example, our scheduled 2-night admissions have improved from 36% staying the full 2 nights in 2020 to 54% staying 2 nights in 2021 to date. Looking at the indications for EMU admissions, we have also identified opportunities to optimize our clinical decision support to help providers determine the optimal number of nights to order. We have continued to monitor our balancing measures, but as expected, there have been no changes to date.

Conclusions: Employing Improvement Science methods has allowed us to better understand the challenges of EMU scheduling. Although we have seen slight improvements in certain drivers, the variable impact of all drivers is so great these improvements have not resulted in increased census. However, the ability to see how each driver affects the overall census and to assess whether there has truly been an improvement provides knowledge that allows us to more appropriately focus our efforts.

Through this process, we have realized more significant interventions may be required to alter our primary outcome of increasing EMU census, such as overbooking patients and implementing tighter scheduling necessitating turning over rooms more quickly between patients.

Funding: Please list any funding that was received in support of this abstract.: No funding was received in support of this abstract.