Journal Logo

Empirical Investigations

An Unexpected Diagnosis: Simulation Reveals Unanticipated Deficiencies in Resident Physician Dysrhythmia Knowledge

Spanos, Stephanie L. MD; Patterson, Mary MD

Author Information
Simulation in Healthcare: The Journal of the Society for Simulation in Healthcare: February 2010 - Volume 5 - Issue 1 - p 21-23
doi: 10.1097/SIH.0b013e3181b2c526
  • Free


Emergency medicine entails “extreme variety, task complexity, uncertainty, unpredictability, multitasking, and production pressures that introduce high risk for error.”1 Rapid, decisive stabilizing interventions are required of the emergency department resuscitation team. In academic hospitals, the resuscitation teams may include inexperienced staff at multiple levels of training. In an effort to minimize medical errors due to team member inexperience and ineffective communication, the development of computerized decision support tools is logical as we look to devise ways to improve clinical decision making and communication. Computerized decision support has been developed to assist with data incorporation and processing and found to be a favorable adjunct to patient care in other hospital environments.2 There is limited experience in the emergency department with computerized decision support thus far.

We initiated a study to compare the resuscitation decision support techniques of the traditional code book method (books and pamphlets carried by physicians and present in the resuscitation room) versus a computerized decision support system developed by Pediatric Emergency Medicine Software. Phoenix (PEMSoft, San Francisco, CA), the pediatric critical care resuscitation software, is a length, age, and weight-based resuscitation guide, providing instantaneous pediatric drug dosing, equipment sizing, and Pediatric Advanced Life Support (PALS) protocol algorithms for pediatric medical emergencies. Users choose an algorithm that details each step needed for the resuscitation and also provides useful reminders of dangers involved with each step. As this is a new program, we sought to evaluate the feasibility and efficacy of the program in the clinical environment by implementing its use during resuscitations in the patient simulator.

Patient simulators have been used for the evaluation of physician competence in knowledge and procedural abilities.3,4 Girzadas et al3 used the patient simulator to assess their emergency medicine residents and differentiate novice from experienced physicians. McLaughlin et al4 proposed that the patient simulator be used to evaluate their emergency medicine residents for accomplishing Accreditation Council for Graduate Medical Education (ACGME) core competencies. Kobayashi et al5 also tested their new emergency department preparedness and team skills using a patient simulator. Similarly, this study would use high-fidelity simulation to evaluate the new technology of Phoenix resuscitation software. The Institutional Review Board at Cincinnati Children's Hospital approved this study.


Resident physicians at our urban, tertiary care pediatric hospital routinely participate in scheduled resuscitation simulations. Teams complete two resuscitation scenarios per visit. The resident physician groups range in size from two to six members, but more often, three or four members are present for each team. These resident physicians have previously completed PALS and/or Advanced Cardiac Life Support (ACLS) courses at the beginning of residency training. The teams consist of residents from all levels and include training programs for pediatrics, emergency medicine, combined internal medicine and pediatrics, and pediatric combined specialties (psychiatry and rehabilitation).

The study is a prospective time series design with pediatric resident physicians completing a series of resuscitations using the traditional code book method, followed by a series of resuscitations using the computerized decision support system Phoenix. We chose to study a high-risk medical emergency with definable time intervals and necessary interventions: pulseless ventricular tachycardia or pulseless ventricular fibrillation. Both scenarios require the same interventions. The time from dysrhythmia to the time of defibrillation and cardiopulmonary resuscitation (CPR) are easily measured intervals during resuscitation and also clinically important results. Previous research demonstrated that with each minute delay from the time of cardiac arrest with ventricular fibrillation to defibrillation, the probability of survival decreases 7%–10% (Heart and Stroke Foundation of Canada. Position Statements: Public Access to Automated External Defibrillators (AEDs); 2005. Available at: Studies have shown that less than 5% of patients survive if the time from cardiac arrest to defibrillation is greater than 12 minutes. Research has also demonstrated that effective CPR until defibrillation improves survival; however, defibrillation is necessary for successful resuscitation. Therefore, recognition of these dysrhythmias, followed by emergent defibrillation during ventricular tachycardia or fibrillation, is an important competency for resident physicians.

The primary outcome measure for the study was time from pulseless ventricular tachycardia or fibrillation to defibrillation, and secondary outcome measures were time to CPR, time to epinephrine administration, dose of epinephrine given, and dose of defibrillation in joules. The study was initially randomized (computerized list random generation) in two ways: (1) performance of the studied resuscitation as the first or second simulation and (2) the dysrhythmia was either pulseless ventricular fibrillation or pulseless ventricular tachycardia.


After 18 studied resuscitation scenarios, we noticed a deficiency in the resident physicians' ability to recognize ventricular fibrillation. Six pulseless ventricular tachycardia resuscitations had been simulated, and all six teams successfully identified the dysrhythmia and responded accordingly (Fig. 1). Twelve pulseless ventricular fibrillation resuscitations had been simulated, but only six teams successfully identified the dysrhythmia and responded accordingly. Some teams only recognized the rhythm after it was present for several minutes on the monitor. Instead of recognizing the dysrhythmia as pulseless ventricular fibrillation, the resident physicians often identified the rhythm as “Asystole,” “pulseless electrical activity” (PEA), and “Asystole PEA.” Often the resident physicians alternated between identifying the rhythm as “PEA” and “Asystole.” The treatment algorithms for PEA and Asystole are similar to one another, but neither PEA nor asystole require defibrillation. So, misidentification of the ventricular fibrillation rhythm would likely be fatal in a clinical environment.

Figure 1.:
Ventricular fibrillation as seen on trauma bay monitor.

In addition to the difficulty with rhythm recognition, implementation of resuscitation was variable (Table 1). Using SPSS (Chicago, IL), the mean, standard deviation, and P value were calculated on the time intervals to intervention. The time to defibrillation was affected by the timeliness of rhythm recognition, knowledge of defibrillation necessity, and inexperience with defibrillator use. Interestingly, the participants involved in ventricular fibrillation simulations quickly administered CPR, whereas two groups in the ventricular tachycardia resuscitations neglected CPR for 4–5 minutes. Both resuscitation groups administered epinephrine with equal efficiency and used the appropriate dose. Delivered joules for defibrillation were correct with all groups except one resuscitation of ventricular tachycardia and one resuscitation of ventricular fibrillation.

Table 1:


Although the data presented represent a small sample size, the disparity in rhythm recognition of ventricular fibrillation is concerning. The goal of the study was to compare decision support tools, which requires correct dysrhythmia recognition as a prerequisite. Because of concerns regarding lack of recognition of ventricular fibrillation, accomplishing study goals, and time pressures to complete enrollment, the study has since been modified such that only pulseless ventricular tachycardia is performed in resuscitation scenarios. Teaching is provided after resuscitations regarding the rhythms of ventricular fibrillation, ventricular tachycardia, PEA, and asystole, along with guidance on appropriate management. In addition, the residency director has been notified regarding the deficit in rhythm recognition and the ongoing remediation efforts at the simulation center.

Shilkofski et al6 studied pediatric resident physician resuscitation of unstable supraventricular tachycardia. They found that in 20% of scenarios, the rhythm was misidentified. Additionally, they found numerous delays in appropriate treatment; henceforth, they failed to meet American Heart Association guidelines. Nelson et al7 studied pediatric resident physician resuscitation performance during simulated pediatric cardiopulmonary arrest. They found that 86% of resident physicians used a cognitive aid during the resuscitation. However, 27% of those using the PALS aid chose the wrong algorithm to follow on their cognitive aid and thus delivered medications and electricity incorrectly. Fifty-four percent of residents in this study did not perform CPR on the pulseless patient in the time from dysrhythmia to initiation of cognitive aid use, and despite the recommendation on the cognitive aid, only 3% then initiated chest compressions. These two studies combined with the initial results of our study suggest that despite training in PALS and/or ACLS courses, pediatric resident physicians may not be proficient in dysrhythmia recognition and resuscitation. The underlying etiology of this deficiency may be related to the relative infrequency of ventricular dysrhythmias and/or the residents' discomfort in using defibrillators. Simulation is an excellent and important tool for diagnosing these deficiencies in a safe environment, providing educational interventions, and practicing rhythm recognition and appropriate interventions. Given the diagnostic value of simulation in this context and the critical nature of this deficiency, it may be useful to assess and remediate the competencies associated with dysrhythmia recognition and management before high-risk clinical rotations such as critical care and emergency medicine. Future studies could evaluate the learning retention of PALS and ACLS courses, the impact of educational interventions, and further define deficiencies and the likely etiology of these deficiencies in dysrhythmia recognition and treatment among pediatric residents.


We would like to acknowledge the research team: Mike Moyer, MS, EMT-P; Brian Pio, EMT-P; Tom LeMaster, RN, EMT-P; Tiffany Perkins, RN, BSN; Jennifer Manos, RN; and Jerome Bauer, RN.


1.Biros M, Adams J, Wears R. Errors in emergency medicine: a call to action. Acad Emerg Med 2000;7:1173–1174.
2.Morris A. Developing and implementing computerized protocols for standardization of clinical decisions. Ann Int Med 2000;132:373–383.
3.Girzadas D, Smith-Coggins R, Harter PM, et al. High-fidelity simulation: assessing resident competency in patient care. Acad Emerg Med 2006;13(suppl 1):190–191.
4.McLaughlin S, Doezema D, Sklar DP. Human simulation in emergency medicine training: a model curriculum. Acad Emerg Med 2002;9:1310–1318.
5.Kobayashi L, Shapiro MJ, Sucov A, et al. Portable advanced medical simulation for new emergency department testing and orientation. Acad Emerg Med 2006;13:691–695.
6.Shilkofski N, Nelson KL, Hunt EA. Recognition and treatment of unstable supraventricular tachycardia by pediatric residents in a simulation scenario. Simul Healthc 2008;3:4.
7.Nelson K, Shilkofski N, Haggerty J, et al. Cognitive aids do not prompt initiation of cardiopulmonary resuscitation in simulated pediatric cardiopulmonary arrest. Simul Healthc 2007;2:54.

Decision Support; Simulation; Dysrhythmia; Resident Physician; Shared Mental Model; PALS

© 2010 Society for Simulation in Healthcare