Randomized controlled trial of high fidelity patient simulators compared to actor patients in a pandemic influenza drill scenario.

Abstract

During disaster drills hospitals traditionally use actor victims. This has been criticized for underestimating true provider resource burden during surges; however, robotic patient simulators may better approximate the challenges of actual patient care. This study quantifies the disparity between the times required to resuscitate simulators and actors during a drill and compares the times required to perform procedures on simulator patients to published values for real patients. A randomized controlled trial was conducted during an influenza disaster drill. Twelve severe influenza cases were developed for inclusion in the study. Case scenarios were randomized to either human actor patients or simulator patients for drill integration. Clinical staff participating in the drill were blinded to the study objectives. The study was recorded by trained videographers and independently scored using a standardized form by two blinded attending physicians. All critical actions took longer to perform on simulator patients compared to actor patients. The median time to provide a definitive airway (8.9min vs. 3.2min, p=0.013), to initiate vasopressors through a central line (17.4min vs. 5.2min, p=0.01) and time to disposition (16.9min vs. 5.2min, p=0.01) were all significantly longer on simulator patients. Agreement between video reviewers was excellent, ranging between 0.95 and 1 for individual domain scores. Times required to perform procedures on simulators were similar to published results on real-world patients. Patient actors underestimate resource utilization in drills. Integration of high fidelity simulator patients is one way institutions can create more realistic challenges and better evaluate disaster scenario preparedness.