The association of closed-collaborative SICU modeling on emergency general surgery patient outcomes.

Objective: Surgical intensive care unit (SICU) optimization is a critical factor impacting patient outcomes and resource utilization. SICUs operate using an open or closed model, where the surgeon or intensivist, respectively, manages critically-ill patients. In 2017, we adopted a closed-collaborative model. This study aimed to compare patient outcomes in the closed-collaborative model vs. the previous open model in a cohort of emergency general surgery (EGS) patients. Methods: A retrospective review of EGS SICU patients from August 2015 to July 2019 was performed. Patients were divided into "Open" and "Closed" cohorts before or after closed-collaborative model implementation on August 1, 2017. Demographic variables and clinical outcomes were analyzed. Results: We identified 434 patients (O:191; C:243). While no significant demographic differences were observed, there was a higher proportion of patients with qSOFA scores greater than 2 in the closed cohort. There were no differences regarding sepsis, cerebrovascular accident, myocardial infarction, venous thromboembolism, anemia, SICU length of stay (LOS), SICU costs, ventilation requirements, or ventilator duration; mortality rate was higher, but hospital LOS was shorter in the closed cohort. Conclusion: Overall, outcomes were not statistically different between the two models, despite sicker patients in the closed group, which we suspect accounts for the higher mortality in this group. We expect the decreased hospital LOS observed in the closed cohort improved bed management, patient flow, and ultimately led to institutional cost savings. Further investigation is warranted to examine SICU modeling effects in other surgical specialties and to evaluate potential hospital-level administrative benefits.