Application of Machine Learning Models to Biomedical and Information System Signals From Critically Ill Adults.

BACKGROUND: Machine learning (ML)-derived notifications for impending episodes of hemodynamic instability and respiratory failure events are interesting because they can alert physicians in time to intervene before these complications occur. RESEARCH QUESTION: Do ML alerts, telemedicine system (TS)-generated alerts, or biomedical monitors (BMs) have superior performance for predicting episodes of intubation or administration of vasopressors? STUDY DESIGN AND METHODS: An ML algorithm was trained to predict intubation and vasopressor initiation events among critically ill adults. Its performance was compared with BM alarms and TS alerts. RESULTS: ML notifications were substantially more accurate and precise, with 50-fold lower alarm burden than TS alerts for predicting vasopressor initiation and intubation events. ML notifications of internal validation cohorts demonstrated similar performance for independent academic medical center external validation and COVID-19 cohorts. Characteristics were also measured for a control group of recent patients that validated event detection methods and compared TS alert and BM alarm performance. The TS test characteristics were substantially better, with 10-fold less alarm burden than BM alarms. The accuracy of ML alerts (0.87-0.94) was in the range of other clinically actionable tests; the accuracy of TS (0.28-0.53) and BM (0.019-0.028) alerts were not. Overall test performance (F scores) for ML notifications were more than fivefold higher than for TS alerts, which were higher than those of BM alarms. INTERPRETATION: ML-derived notifications for clinically actioned hemodynamic instability and respiratory failure events represent an advance because the magnitude of the differences of accuracy, precision, misclassification rate, and pre-event lead time is large enough to allow more proactive care and has markedly lower frequency and interruption of bedside physician work flows.

COVID-19 and the Transformation of Intensive Care Unit Telemedicine.

The concept of telecritical care has evolved over several decades. ICU Telemedicine providers using both the hub-and-spoke ICU telemedicine center and consultative service delivery models offered their services during the COVID-19 pandemic. Telemedicine center responses were more efficient, timely, and widely used than those of the consultative model. Bedside nurses, physicians, nurse practitioners, physician assistants, and respiratory therapists incorporated the use of ICU telemedicine tools into their practices and more frequently requested critical care specialist telemedicine support.

Describing advanced practice provider roles within critical care teams with Tele-ICUs: Exemplars from seven US health systems.

Telehealth is an acknowledged strategy to meet patient healthcare needs. In critical care settings, Tele-ICU's are expanding to deliver clinical services across a diverse spectrum of critically ill patients. The expansion of telehealth provides increased opportunities for advanced practice providers including advanced practice nurses and physician assistants; however, limited information on roles and models of care for advanced practice providers in telehealth exist. This article reviews current and evolving roles for advanced practice providers in telehealth in acute and critical care settings across 7 healthcare systems in the United States. The health system exemplars described in this article identify the important role of advanced practice providers in providing patient care oversight and in improving outcomes for acute and critically ill patients. As telehealth continues to expand, additional opportunities will lead to novel roles for advanced practice providers in the field of telehealth to assist with patient care management for subacute, acute, and critically ill patients.

Battling Intensivist Burnout: A Role for Workload Management.

Burnout syndrome is an increasingly common phenomenon that threatens our critical care workforce and the well-being of its members. Burnout syndrome can be conceived of as a workforce manifestation of chronic workload and workforce capacity imbalance. This study explores the role of workload management tools that can address workload as a complement to the resilience-based countermeasures that seek to increase worker capacity. We were able to use step-wise increments in the volume of documentation-related tasks that occurred at the time of electronic health record (EHR) implementation to investigate the relation of workload and burnout. Specialty-specific increases in the prevalence of self-reported burnout during the era of EHR adoption were compared with increases of the length of documentation created by the corresponding specialists observed prior to and following EHR implementation; a robust direct association was reported. To connect ICU workload to burnout, the number of tasks performed was extracted from the EHR, and we measured the average time that our ICU team members required to complete these tasks. Our ICU workforce efficiency was calculated as the ratio of mandatory task time to scheduled time. Comparing this ratio vs a well-established industrial standard for equipment efficiency made us realize that our average workload seemed excessive and placed our staff at risk of burnout syndrome. It is difficult to conceive that our resilience-based countermeasures to prevent and treat burnout would not be more effective when combined with measures that reduce the time our staff members spend on mandatory ICU tasks.