Background: Hospital length of stay (LOS) serves as a key quality metric for evaluating efficient hospital management within health systems. improving LOS enables hospitals to effectively handle both elective and emergent admissions. Excessive demand without corresponding capacity results in emergency department and unit crowding, leading to unfavorable mortality outcomes, impacts patient and provider experiences, contributes to avoidable complications like hospital-acquired infections and falls. While various strategies have emerged to improve LOS, interventions targeting hospitalists’ morning census distribution strategies lack sufficient data. Most hospitalist groups currently employ a model that evenly distributes patients at the beginning of the day, theoretically fair but unevenly increase workload for hospitalists with higher discharges leading to dissatisfaction and burnout.This study aims to outline and implement a model where the hospitalists’ morning census is based on equal distribution of overnight admits rather than level setting each team, and assess its impact on average hospital LOS.
Methods: Saint Michael Medical Center in Silverdale, Washington, is a 300-bed hospital within the Virginia Mason Franciscan Health system. Sound Physicians hospital leadership initiated new overnight patients distribution intervention to improve LOS. The current staffing model involves 10 daytime rounding teams (7 am-7 pm), one admitter team (9 am-7 pm), one swing team (4 pm-2 am), and one night team (9 pm-7 am). The average daily starting census, ranging from 150-170 patients, each rounding team has similar morning census. A pilot project aimed to allocate new overnight patients equally among teams ensuring a more equitable workload. The focus was establish a fair workload for all providers, considering new admits and discharges require more time in management and planning. The process involved documenting starting census for each team prior to adding overnight admits, calculating overnight admits and total census, and distributing patients from the bottom up. Final census adjustments were made to minimize disparities between teams (Table 1). A quantitative analysis, both pre- and post-intervention, was conducted to assess the correlation between hospitalists’ daily morning census workload distribution and hospital LOS.
Results: The pilot project spanned three months, January 1st – March 31st. LOS data was gathered from the Common Spirit dashboard. Post-intervention, there was a notable improvement in LOS, decreasing from 7.10 in December 2022 to 6.56 in January 2023, and maintaining a lower range at 5.44 and 5.81 in February and March 2023, respectively, as illustrated in Figure 1. The project was eventually halted due to Common Spirit’s nationwide policy mandating geographical rounding. The hospital’s structural layout, geographical multiple disciplinary rounds (MDRs), posed challenges in implementing new distribution policy.
Conclusions: Fair patients workload with even distribution of overnight admits to hospitalist rounding team rather then equal morning census for each team represents an organizational level change that not only improves LOS but also has the potential of improving hospitalist burnout and career satisfaction. It is unclear if the work can be applied to larger hospitals with strict ge-rounding. More data needs to be collected in different centers to determine a cause effect relationship to determine widespread implications in clinical practice.
