57 Improving the Safety and Quality of Inter-Hospital Transfer for Nontraumatic Intracerebral and Subarachnoid Hemorrhage

Abstract

Patients with nontraumatic intracranial hemorrhage (ICH, including intracerebral or subarachnoid hemorrhage) often require inter-hospital transfer (IHT) for subspecialty care. IHT involves multiple care transitions that harbor safety threats and quality gaps, yet few studies have implemented and evaluated interventions to improve this process. We re-engineered the IHT process at our institution for patients with nontraumatic ICH and implemented a multi-modal intervention that standardized clinical care goals and enhanced inter-disciplinary communication. We examined the impact of this intervention on process metrics, communication during transition periods, and the timeliness/effectiveness of care. Robust pre-intervention problem analysis identified several safety threats in this patient population. Guided by these results, we instituted a multi-modal intervention in May 2017 that included: 1) statewide clinical guideline dissemination, 2) standardization of the IHT acceptance process, 3) implementation of a patient arrival notification system, and 4) EMR enhancements. The intervention was evaluated through an EMR-based dashboard, case-by-case audit and feedback, and content analysis of IHT calls. We performed pre/post-intervention descriptive statistics analysis. The primary outcomes were a set of quality and safety measures: 1) process outcomes (ED length of stay [LOS], time to admit order, and ED boarding), 2) inter-departmental coordination/communication (% of times consulting services were notified upon patient arrival, % of times an ED attending to NICU attending conversation occurred), and 3) timeliness/effectiveness outcomes (adequate blood pressure control, and time to reversal of anticoagulation). The interventions resulted in significant improvements in the process of care: ED LOS (300 to 149 minutes, p<.01), ED time to admission order (66 to 20 minutes, p<.05) and ED boarding time (218 to 92 minutes, p<.01). Inter-departmental communication was enhanced: the percentage of IHT in which consult services were notified upon patient arrival improved from 38% to 91% (p < .05); additionally, a NICU attending discussed the care plan with the ED attending before the patient arrived in 64% of transfers post-intervention compared to 10% pre-intervention (p < .05). Timeliness/effectiveness improvements were noted in the reduction of time to anticoagulation reversal, though this was not statistically significant (217 to 165 minutes, p = 0.53). There was no significant difference in blood pressure control (86% to 80% with adequate BP control, p = 1). Decreased mortality was noted, though this was not statistically significant (27% to 14%, p = 0.26) The implementation of a large-scale multi-modal intervention significantly improved process and communication metrics associated with IHT for patients with nontraumatic ICH and resulted in a trend towards improvement in the timeliness and effectiveness of clinical care. Future research should explore the relationship between process improvement and more rare clinical outcomes, as our study was not powered to detect such differences. Our approach may be scalable to other critical care entities and achieve greater impact on the overall safety and quality threats related to the IHT process at large.