Development of a hospitalist‐run short‐stay unit to improve throughput and reduce length of stay

Emergency departments (EDs) nationwide are overcrowded, a problem that is exacerbated when hospital systems close and more patients lack access to healthcare.1, 2 As hospital volumes increase, admitted patients remain in EDs until a disposition has been made.3 This is associated with worse patient safety outcomes: admitted patients boarding in the ED may not receive appropriate levels of care, and boarding due to overcrowding may lead to hospital diversion.4, 5 Clinical decision units (CDUs), ED-run observational units with an intended <24-h turnaround time for patients with low acuity not meeting admission criteria, arose in the 1990s. CDUs are strictly protocolized with significant exclusion criteria; they are run by ED and/or hospitalist physicians and have a time limit of 24 h before further disposition is made.6, 7 Traditional ED-run CDU/observation units often use protocol-driven models to manage patient care. Increasingly, these units encounter patient intake challenges due to complex patient comorbidities and socioeconomic factors.8 Simultaneously, as an understanding of the utility of CDU/observation units has grown and facilities are incentivized by policy to facilitate timely disposition, more patients are being sent to such units. This complicates patient management and potentially bottlenecks throughput, necessitating more flexible and inclusive approaches. Involving hospitalists in the care of complex short-stay patients can enhance both continuity and quality of care, often extending beyond the scope of ED practice. For instance, hospitalists are well versed in performing complex admission and discharge medication reconciliations, a critical component of not only inpatient care but also the successful transition to outpatient care. This expertise allows hospitalists to manage more complex patients in the short-stay unit (SSU). Finally, effective collaboration between ED consultants and hospitalists is key to accurately identifying patients suitable for admission9 and facilitates the management of patients who are more likely to require admission after observation. We aimed to decompress the ED and improve patient throughput and care progression through the development of a hospitalist-run SSU. As each patient admitted to the SSU opened up a corresponding bed on the floor, we expected an ancillary benefit of an improved length of stay (LOS) for SSU patients compared to admitted patients. With the support of hospital leadership and in partnership with ED providers, we instituted the SSU to accommodate patients with a wide range of acuity and baseline chronic complex medical conditions who did not meet the criteria for the CDU. The SSU opened in October 2022 and provides 24/7 coverage of admissions, progressing care, and discharges, operating in conjunction with the CDU rather than replacing it. To determine its effectiveness, we compared data collected for the time period before (January to September 2022) and after (November 2022 to October 2023) implementation. We evaluated the number of observation status patients in the hospital per day and set the number of beds in the SSU at 10. We collaborated closely with ED colleagues to design inclusion/exclusion criteria (Figure 1, , tip sheet) and coordinate patient flow, aiming to maximize efficiency and potentially redirect admissions. Patients requiring airborne or enteric precautions, or those with severe illnesses necessitating intensive care unit (ICU)-level care, were excluded from the SSU. Meanwhile, those with low or intermediate care needs, such as those on bilevel positive airway pressure (BiPAP) or insulin infusion, were eligible for admission to the SSU if their anticipated stay was <48 h. The distinguishing factor for the SSU is that we can take any patient not meeting exclusion criteria as long as the triaging hospitalist physician thinks their stay will be <48 h, including those who need intermediate care (which distinguishes our model from others with flexible criteria, such as Abenheim et al.9). Our SSU is not bound to diagnosis, observation status, or even level of care (except ICU), setting the unit apart from primarily observation-status units with strictly protocolized patients. The SSU is team-staffed (Figure 1b, logic model): one hospitalist physician triages admissions and collaborates with an Advanced Practice Provider (APP) for admissions. The physician oversees all admissions and discharges. From 7 p.m. to 7 a.m., 7 days a week, an APP cross-covers and admits patients to the unit. We worked closely with nursing administration to staff this unit with med/surg and intermediate care nurses, rather than ED nurses. This intentional component is novel, given the SSU is considered an inpatient unit accommodating observation and inpatient status patients. The SSU is run by hospitalists 24 h/day, utilizing the physician-APP model to streamline admission, care progression, and planned patient discharge. Through direct ED/hospitalist communication, short-stay patients are identified and then rapidly streamlined from the main ED to the near-adjacent SSU. Once admitted to the SSU, these patients receive care parallel to inpatient counterparts but with expedited necessary testing (e.g., echocardiograms, imaging studies) and SSU-dedicated ancillary services including physical and occupational therapy, case management, and social work. To facilitate care progression, we worked with support staff to have resources prioritized including imaging, stress test, phlebotomy, nursing staff, therapy services, dedicated care coordination, and social workers. This allocation allows us to quickly care for and discharge patients who would not benefit from extended time in the hospital. We measured ED throughput improvement by comparing SSU versus non-SSU medicine patients' median time from admission order placement to the patient leaving ED/checking into a unit (Table 1), noting a 71% efficiency gain for the SSU. Patient progression improvement was measured by comparing the LOS of common primary diagnoses among SSU versus non-SSU medicine patients. ED colleagues perceived the implementation of the SSU positively, as our hospitalist perspectives informed alternative dispositions and outpatient care coordination. This collaboration ultimately helped decompress the ED and the hospital, an important change needed for our overcrowded safety-net facility. Our hospitalist-run closed SSU has several innovative features: liberal criteria tailored to meet the needs of patients suspected to only require a short stay to effectively decompress the ED and reduce LOS for common diagnoses with a relatively small investment of staff time and resources. By utilizing diagnosis-based criteria rather than a strictly protocolized diagnosis algorithm, we leverage our clinical acumen and experience as hospitalists to select patients who are most likely to benefit from the SSU, including those with intermediate care needs (e.g., diabetic ketoacidosis on insulin infusion, hypertensive emergency on infusions, chronic obstructive pulmonary disease exacerbation on BiPAP). While SSU patients receive priority testing, which undoubtedly contributes to the efficiency of their stay, the decreased LOS observed is multifactorial and reflects the dedication of hospitalist providers to monitor and progress SSU patients with the goal of discharge within 48 h. The closed unit design enables quick, frequent reassessment of patients by hospitalist providers, supported by dedicated SSU-unit nurses and case managers who also prioritize rapidly advancing patient care and facilitating high daily admissions and discharges. Due to the SSU's high volume and turnover, and its success as outlined above, we are enhancing the model to a one hospitalist, two APP model where the hospitalist supervises all admissions and discharges and primarily focuses on triaging admissions. This will also allow additional direct collaboration with our ED colleagues, a key component to this unit's success. We developed the SSU in close partnership with ED colleagues. Building credibility and trust has allowed the SSU to operate as a complement to the ER, CDU, and medicine floors, facilitating their function and improving patient care. Broad scope: Our SSU accepts patients regardless of diagnosis, level of care, or admission status who are anticipated to require <48 h of care. Flexible: This SSU is much more flexible in approach to patient selection and admission compared to the CDU, allowing for inclusivity of degrees of chronic conditions and socioeconomic factors—however, we run in parallel with our CDU rather than competing with it to best serve all our patients. Resource-rich: Our SSU was supported by a dedicated 24/7 team with the prioritization of hospital resources allowing flow progression and enhanced throughput, decreased LOS, alleviation of ED congestion, and improvement in overall hospital efficiency. Collaborative: This SSU operates in close collaboration with ED colleagues to maximize efficiency and help in alternative dispositions by providing outpatient resources and coordinating care when appropriate. Interdisciplinary: We use an innovative physician/APP model, and this interdisciplinary collaboration—along with emergency physicians and other healthcare professionals—is also key in ensuring this model is successful, given patients' diverse acuity levels and complex comorbidities. The authors would like to thank Dr. Daniel P. Hunt, MD for his thoughtful review of our paper. The authors declare no conflict of interest.