In-situ Simulation Use for Rapid Implementation and Process Improvement of COVID-19 Airway Management

Introduction: The coronavirus disease 2019 (COVID-19) pandemic presents unique challenges to frontline healthcare workers. In order to safely care for patients new processes, such as a plan for the airway management of a patient with COVID-19, must be implemented and disseminated in a rapid fashion. The use of in-situ simulation has been used to assist in latent problem identification as part of a Plan-Do-Study-Act cycle. Additionally, simulation is an effective means for training teams to perform high-risk procedures before engaging in the actual procedure. This educational advance seeks to use and study in-situ simulation as a means to rapidly implement a process for airway management in patients with COVID-19.Methods: Using an airway algorithm developed by the authors, we designed an in-situ simulation scenario to train physicians, nurses, and respiratory therapists in best practices for airway management of patients with COVID-19. Physician participants were surveyed using a five-point Likert scale with regard to their comfort level with various aspects of the airway algorithm both before and after the simulation in a retrospective fashion. Additionally, we obtained feedback from all participants and used it to refine the airway algorithm.Results: Over a two-week period, 93 physicians participated in the simulation. We received 81 responses to the survey (87%), which showed that the average level of comfort with personal protective equipment procedures increased significantly from 2.94 (95% confidence interval, 2.71-3.17) to 4.36 (4.24-4.48), a difference of 1.42 (1.20-1.63, p < 0.001). There was a significant increase in average comfort level in understanding the physician role with scores increasing from 3.51 (3.26-3.77) to 4.55 (2.71-3.17), a difference of 1.04 (0.82-1.25, p < 0.001). There was also increased comfort in performing procedural tasks such as intubation, from 3.08 (2.80-3.35) to 4.38 (4.23-4.52) after the simulation, a difference of 1.30 po