The role of big data, risk prediction, simulation, and centralization for emergency vascular problems: Lessons learned and future directions

Abstract

Vascular specialists remain in high demand in current practice and commonly oversee care delivery for a variety of clinical emergencies. Accordingly, the contemporary vascular surgeon must be facile with treating a spectrum of problems, including a complex, heterogeneous group of acute arteriovenous thromboembolic and bleeding diatheses. It has been documented previously that there are substantial current workforce limitations placing constraints on vascular surgical care provision. Moreover, with the aging at-risk population, there remains a considerable national urgency to improve timely diagnoses, specialty consultation, and appropriate transfer of patients to centers of excellence capable of providing a comprehensive compendium of emergency vascular services. Clinical decision aids, simulation training, and regionalization of nonelective vascular problems are all strategies that have been increasingly recognized to address these service gaps. Notably, clinical research in vascular surgery has traditionally focused on identification of patient- and procedure-related factors that influence outcomes by using resource-intensive causal inference methodology. By comparison, large data sets have only more recently been recognized to be a valuable tool that can provide heuristic algorithms to address more complex health care problems. Such data can be manipulated to generate clinical risk scores and decision aids, as well as robust outcome descriptions, which stand to inform stakeholders regarding best practice. The purpose of this review was to provide a robust overview of the lessons derived from the application of big data, risk prediction, and simulation in the management of vascular emergencies.