The effect of emergency manuals on team performance during two different simulated perioperative crises: A prospective, randomized controlled trial

Abstract

Study objectiveWhether having an emergency manual (EM) available for use during perioperative crises enhances or detracts from team performance, especially for multi-factorial diagnostic situations that do not explicitly match a chapter of the EM. DesignA simulation-based, prospective randomized trial based upon two perioperative crises, one involving a patient with a transfusion reaction for which the EM contains a specific chapter, and the other involving a patient with refractory hypotension progressing into septic shock for which the EM does not have a specific chapter. Setting52 regularly scheduled 6-h courses at the Center for Medical Simulation in Boston, Massachusetts, USA. Study group304 US-trained practicing anesthesiologists. InterventionsThe absence or presence of the EM during the simulation case. MeasurementsTeams were rated in the following categories: primary underlying diagnosis, fluid resuscitation, treatment of primary diagnosis, cardiac arrest management, overall crisis management, and (if applicable) EM usage. Also, raters recorded free-text ‘field notes’ about the usage-patterns and perceived utility of the EM. Using these ‘field notes’ and a two-stage, inductively revised procedure, two independent reviewers examined a subset of case videos for action analysis. Main resultsPerformance ratings for a total of 51 teams and 95 simulations were included in the final analysis. No effect on performance was demonstrated with providing the EM in either the refractory hypotension/septic shock case or the transfusion reaction case, with the exception of the PEA arrest category. In the subset of simulations in which resuscitation from PEA arrest performance could be evaluated, EM availability was associated with an adjusted mean 1.3 point (99% confidence interval [CI]: 0.2, 2.4) improvement in performance in the transfusion reaction case (p = 0.004), but only an adjusted mean 0.2 point (99% CI, −0.7, 1.1) improvement in the refractory hypotension/septic shock case (p = 0.530) (p for interaction = 0.069). Analysis of actions found that when available, the EM was usually used, but often not until after cardiac arrest had occurred. In some cases, teams persisted with incorrect diagnoses and treatments irrespective of the presence or absence of an EM. ConclusionsProviding an EM did not affect team performance in areas like diagnosis, treatment, fluid resuscitation, communication, and teamwork in management of perioperative crises such as transfusion reaction where an explicit chapter in the EM exists and refractory hypotension / septic shock where an explicit chapter does not exist. A suggestion of improved cardiac arrest resuscitation with the availability of an EM was found, but should be interpreted with caution given a limited sample size. Observed actions using the EM demonstrated that only about half of the teams used the EM to any substantive degree and most used it relatively late in the crisis. By observation, the EM appeared to be helpful in about half of the cases and did not, by itself, deter from appropriate management.