A Remote Surveillance Platform to Monitor General Care Ward Surgical Patients for Acute Physiologic Deterioration

Abstract

BACKGROUND: The traditional paradigm of hospital surgical ward care consists of episodic bedside visits by providers with periodic perusals of the patient’s electronic health record (EHR). Vital signs and laboratory results are directly pushed to the EHR but not to providers themselves. Results that require intervention may not be recognized for hours. Remote surveillance programs continuously monitor electronic data and provide automatic alerts that can be routed to multidisciplinary providers. Such programs have not been explored in surgical general care wards. METHODS: We performed a quality improvement observational study of otolaryngology and ophthalmology patients on a general care ward from October 2017 to March 2019 during nighttime hours (17:00–07:00). The study was initiated due to the loss of on-site anesthesiology resources that historically helped respond to acute physiologic deterioration events. We implemented a remote surveillance software program to continuously monitor patients for severe vital signs and laboratory abnormalities and automatically alert the ward team and a remote critical care anesthesiology team. The primary end point was the true positive rate, defined as the proportion of alerts that were associated with a downstream action that changed the care of the patient. This was determined using systematic chart review. The secondary end point, as a measure of alarm fatigue, was the average number of alerts per clinician shift. RESULTS: The software monitored 3926 hospital visits and analyzed 1,560,999 vitals signs and 16,635 laboratories. It generated 151 alerts, averaging 2.6 alerts per week. Of these, 143 (94.7%) were numerically accurate and 8 (5.3%) were inaccurate. Hypoxemia with oxygen saturation <88% was the most common etiology (92, 63%) followed by tachycardia >130 beats per minute (19, 13.3%). Among the accurate alerts, 133 (88.1%) were true positives with an associated clinical action. Actions included a change in management 113 (67.7%), new diagnostic test 26 (15.6%), change in discharge planning 20 (12.0%), and change in level of care to the intensive care unit (ICU) 8 (4.8%). As a measure of alarm fatigue, there were 0.4 alerts per clinician shift. CONCLUSIONS: In a surgical general care ward, a remote surveillance software program that continually and automatically monitors physiologic data streams from the EHR and alerts multidisciplinary providers for severe derangements provided highly actionable alarms at a rate that is unlikely to cause alarm fatigue. Such programs are feasible and could be used to change the paradigm of monitoring.

Opportunities of AI-powered applications in anesthesiology to enhance patient safety

“…50 Automating the processing of multiple monitors using remote surveillance is feasible and can provide highly actionable information. 51 While such systems would require complex AI models, and those remain to be developed for intraoperative use, the benefits for patient safety will be substantial.…”

Systems Anesthesiology: Integrating Insights From Diverse Disciplines to Improve Perioperative Care

“…36 An ML algorithm developed using a retrospective cohort in a UK hospital system could detect 42% of cardiac arrests or unplanned ICU admissions up to 48 hours in advance, 37 while a remote surveillance program monitoring vital sign and laboratory abnormalities in otolaryngology and ophthalmology general care ward patients generated actionable alarms at a rate unlikely to cause alarm fatigue. 38 Another group used an automated warning system for early detection of severe postpartum hemorrhage in an obstetric population. 39 While emerging studies will likely continue to show the benefits of remote monitoring systems, systems anesthesiologists need to position themselves as part of collaborative multidisciplinary teams receiving notifications of patients who are becoming critically ill outside of the ICU setting.…”

Remote monitoring in the use of extracorporeal membrane oxygenation and acute mechanical circulatory support

“…Recent publications using the AlertWatch platform have shown promise for how these solutions can impact acute care. Safavi et al [22 ▪ ] utilized a version of the software to perform night hour remote surveillance and alerting on severe vital sign and lab abnormalities for postoperative patients on a ward without on-site intensivist coverage. In nearly 4000 hospital visits over 18 months, the observational study demonstrated a relatively low-alarm rate (2.6 alerts per week) with a majority of alarms (67%) leading to changes in management.…”