Continuous Cloud-Based Early Warning Score Surveillance to Improve the Safety of Acutely Ill Hospitalized Patients

Introduction: This study sought to evaluate the impact of changes made to the process of continually screening hospitalized patients for decompensation. Methods: Patients admitted to hospital wards were screened using a cloud-based early warning score (modified National Early Warning Score [mNEWS]). Patient with mNEWS >= 7 triggered a structured response. Outcomes of this quality improvement study during the intervention period from February through August 2018 (1741 patients) were compared with a control population (1,610 patients) during the same months of 2017. Results: The intervention group improved the time to the first lactate order within 24 hours of mNEWS >= 7 (p < .001), the primary outcome, compared with the control group. There was no significant improvement in time to intensive care unit (ICU) transfer, ICU length of stay (LOS), or hospital mortality. Among patients with a lactate ordered within 24 hours, there was a 47% reduction of in-hospital mortality (odds ratio 0.53, 95% confidence interval 0.3-0.89, p = .02) and a 4.7 day reduction in hospital LOS (p < .001) for intervention versus control cohorts. Conclusions: Cloud-based electronic surveillance can result in earlier detection of clinical decompensation. This intervention resulted in lower hospital LOS and mortality among patients with early detection of and intervention for clinical decompensation.