BRIEF REPORT
Improving Teamwork and Communication in Trauma Care Through In Situ Simulations Daniel Miller, MD, Cameron Crandall, MD, Charles Washington, III, MD, and Steven McLaughlin, MD
Abstract Objectives: Teamwork and communication often play a role in adverse clinical events. Due to the multidisciplinary and time-sensitive nature of trauma care, the effects of teamwork and communication can be especially pronounced in the treatment of the acutely injured patient. Our hypothesis was that an in situ trauma simulation (ISTS) program (simulating traumas in the trauma bay with all members of the trauma team) could be implemented in an emergency department (ED) and that this would improve teamwork and communication measured in the clinical setting. Methods: This was an observational study of the effect of an ISTS program on teamwork and communication during trauma care. The authors observed a convenience sample of 39 trauma activations. Cases were selected by their presenting to the resuscitation bay of a Level I trauma center between 09:00 and 16:00, Monday through Thursday, during the study period. Teamwork and communication were measured using the previously validated Clinical Teamwork Scale (CTS). The observers were three Trauma Nursing Core Course certified RNs trained on the CTS by observing simulated and actual trauma cases and following each of these cases with a discussion of appropriate CTS scores with two certified Advanced Trauma Life Support instructors ⁄ emergency physicians. Cases observed for measurement were scored in four phases: 1) preintervention phase (baseline); 2) didactic-only intervention, the phase following a lecture series on teamwork and communication in trauma care; 3) ISTS phase, real trauma cases scored during period when weekly ISTSs were performed; and 4) potential decay phase, observations following the discontinuation of the ISTSs. Multirater agreement was assessed with Krippendorf’s alpha coefficient; agreement was excellent (mean agreement = 0.92). Nonparametric procedures (Kruskal-Wallis) were used to test the hypothesis that the scores observed during the various phases were different and to compare each individual phase to baseline scores. Results: The ISTS program was implemented and achieved regular participation of all components of our trauma team. Data were collected on 39 cases. The scores for 11 of 14 measures improved from the baseline to the didactic phase, and the mean and median scores of all CTS component measures were greatest during the ISTS phase. When each phase was compared to baseline scores, using the baseline as a control, there were no significant differences seen during the didactic or the decay phases, but 12 of the 14 measures showed significant improvements from the baseline to the simulation phase. However, when the Kruskal-Wallis test was used to test for differences across all phases, only overall communication showed a significant difference. During the potential decay phase, the scores for every measure returned to baseline phase values. Conclusions: This study shows that an ISTS program can be implemented with participation from all members of a multidisciplinary trauma team in the ED of a Level I trauma center. While teamwork and communication in the clinical setting were improved during the ISTS program, this effect was not sustained after ISTS were stopped. ACADEMIC EMERGENCY MEDICINE 2012; 19:608–612 ª 2012 by the Society for Academic Emergency Medicine
From the Department of Emergency Medicine, University of New Mexico, Albuquerque, NM. Received July 1, 2011; revisions received September 13, October 21, and December 1, 2011; accepted December 30, 2011. Presented at the Council of Emergency Medicine Residency Directors Academic Assembly, San Diego, CA, 2010. Funded by the University of New Mexico, Scholarship in Education Allocations Committee Grant. The authors have no disclosures or conflicts of interest to report. Supervising Editor: James Miner, MD. Address for correspondence and reprints: Daniel Miller, MD; e-mail:
[email protected].
608
ISSN 1069-6563 PII ISSN 1069-6563583
L
imitations on resident work hours, an increased focus on decreasing medical error, desire for improved patient care, and the advance of simulation technology have led to widespread use of simulation in residency training programs. Teamwork and communication skills have been areas of particular interest for improvement since groups such as the Institute of Medicine and the National Health Service identified them as major factors in adverse clinical events.1,2 The use of simulation for medical education has good theoretical grounding, simulation has been shown to improve
ª 2012 by the Society for Academic Emergency Medicine doi: 10.1111/j.1553-2712.2012.01354.x
ACADEMIC EMERGENCY MEDICINE • May 2012, Vol. 19, No. 5
•
www.aemj.org
performance of motor tasks, and previous studies reported findings suggestive of improved teamwork and communication after simulation.3–5 However, there remains a paucity of empirical evidence to support the use of simulation specifically for teamwork and communication training.6 We hypothesized that performing simulations of trauma codes in the actual trauma bay with participation from all members of the trauma team would lead to improved teamwork and communication during real trauma activations as measured by the Clinical Teamwork Scale (CTS).7 The goals of this investigation were to determine whether an in situ simulation program could be feasibly instituted in the emergency department (ED) of a Level I trauma center and to determine if such a program could improve teamwork and communication during the care of trauma patients. METHODS Study Design We performed an observational pre-, during, and postintervention comparison study on the effects of an in situ trauma simulation (ISTS) program on teamwork and communication observed among the trauma team during actual trauma activations. The study received institutional review board approval with waiver of informed consent. Study Setting and Population We conducted this study in the trauma resuscitation area of a Level I trauma center ⁄ academic tertiary care center with over 80,000 ED visits and 2,149 trauma activations in 2010. Study participants were multidisciplinary and comprised all members of the trauma team: trauma surgery attendings and residents, emergency physician attendings and residents, ED nurses, technicians, pharmacists, clerks, and respiratory therapists. We used the same method to recruit participants for both the simulation and the observational portions of the study: a page was sent out over our trauma activation paging system and the resulting ad hoc team would run the simulation or be observed running a real trauma, respectively. We selected cases for data collection as a convenience sample of those presenting between 0900 and 1600 Monday through Thursday when our observers were available for data collection. Study Protocol We measured teamwork and communication skills using the CTS, a 14-item scale with a range from 0 (unacceptable) to 10 (perfect) that measures five domains of clinical teamwork: communication, situational awareness, decision-making, role responsibility, and patient friendliness.7 In a previous study it was found to have substantial agreement (j = 0.78), score concordance among raters (Kendall coefficient = 0.95), and excellent inter-rater reliability (interclass correlation coefficient = 0.98).7 Our observers were three Trauma Nursing Core Course certified RNs trained on use of the CTS first by reviewing the original paper’s descriptive anchors of each measure (Table 1 of the original manuscript7). These nurses were further trained
609
by observing the same eight simulated and actual trauma cases that were concurrently observed by two Advanced Trauma Life Support (ATLS) instructors who work as ED attending physicians in our Level I trauma center. Each of these cases was followed by a discussion of appropriate CTS scores between the observers and the ATLS instructors. Following this observer training, CTS measurements were made on a total of 39 observed real trauma activations during four sequential phases: 1) preintervention phase (January 19, 2010, through April 16, 2010)— observers collected baseline data by scoring trauma activations before any interventions had been performed; 2) didactic-only phase (May 11, 2010, through August 12, 2010)—we scored trauma activations for a period following a lecture and discussion on teamwork and communication in trauma care that was given to all departments that participate in trauma activations during the first week of the phase; 3) ISTS phase—our observers scored real trauma activations during the time period concurrent with the performance of weekly ISTSs (August 15, 2010, through October 8, 2010); and 4) potential decay phase—beginning 1 week after the simulation program had been discontinued, we scored trauma activations for a period of 4 weeks (October 18, 2010, through November 12, 2010). The didactic-only phase was a standardized lecture that specified the responsibilities for each individual member of the trauma team and the order in which tasks should be carried out and assigned a specific position in the resuscitation area for each team member. The lecture was given at the departmental meeting for every involved department. The ISTS intervention was an 8-week period of once-weekly trauma simulations performed in our trauma resuscitation area. We conducted simulations using a SimMan 3G (Laerdal, Wappingers Falls, NY). Each simulation team included an emergency medicine attending physician and senior resident; a trauma surgery attending physician, senior resident, and junior resident; and an ED nurse, a technician, a clerk, a respiratory therapist, and a pharmacist. We designed simulations with graduated difficulty highlighting increasing importance of teamwork and communication. The first case was a patient with blunt abdominal trauma in obvious shock with an intact airway and a Focused Assessment with Sonography for Trauma examination positive for intraperitoneal free fluid. The final case was a patient with penetrating chest injury who arrived without intravenous access and required advanced airway management, tube thoracostomy, and pericardiocentesis for stabilization. Each ISTS began with the simulated patient’s arrival via EMS and concluded when either all critical actions had been performed, or the team decided to take the patient out of the ED. Sessions lasted between 15 and 30 minutes and were followed by a 15 to 20 minute debriefing session led by an attending emergency physician with simulation training. Debriefings focused primarily on aspects of teamwork and communication rather than the specific medical and surgical issues of trauma care. We estimate that, of all potential trauma team members, 80% were reached by the didactic intervention and 25% were reached with the ISTS intervention.
6.4 (2.5)
5.7 (2.3)
5.7 (1.6)
4.4 (1.7)
7.0 (1.5)
6.9 (1.6)
7.1 (1.4)
6.7 (1.5)
6.6 (1.6)
5.0 (4.0–7.0)
5.3 (3.0–8.0)
6.0 (3.8–8.0)
6.0 (4.0–7.0)
4.0 (3.0–6.0)
7.0 (6.0–8.0)
7.0 (6.0–8.0)
7.5 (6.0–8.0)
7.3 (6.5–8.0)
7.0 (6.0–8.0)
7.0 (5.0–8.0)
Overall communication Orient new members Transparent thinking Directed communication Closed loop communication Overall situational awareness Resource allocation Decision-making
Prioritization
Role responsibility Role clarity
7.2 (1.3)
7.5 (6.5–8.0)
8.0 (5.0–8.0)
8.0 (6.5–8.0)
8.0 (7.0–8.0)
CTS = clinical teamwork scale; IQR = interquartile range. *Kruskal-Wallis test comparing phase to baseline. Column graph of mean score across four study phases. àKruskal-Wallis test for differences across all phases.
6.7 (1.6)
7.0 (5.0–8.0)
8.0 (5.0–8.3)
8.0 (6.0–8.7)
8.0 (4.5–8.7)
8.0 (6.0–8.0)
8.0 (5.0–8.0)
5.0 (5.0–7.0)
7.0 (5.0–7.0)
7.0 (5.0–7.0)
7.0 (5.0–7.0)
7.0 (5.0–8.0)
7.3 (7.0–8.0)
Median (IQR)
7.2 (1.2)
7.4 (1.6)
7.0 (1.8)
7.1 (1.9)
7.4 (1.7)
7.2 (2.0)
7.2 (1.3)
7.2 (1.5)
5.8 (1.9)
6.2 (1.6)
6.1 (1.6)
5.8 (2.1)
6.3 (1.6)
7.0 (1.5)
Mean (SD)
p*
0.822
0.161
0.337
0.301
0.484
0.562
0.623
0.720
0.049
0.336
0.728
0.626
0.147
0.183
8.0 (8.0–8.0)
8.0 (8.0–8.7)
8.3 (8.0–8.7)
8.0 (8.0–8.7)
8.5 (8.0–8.7)
8.5 (8.0–8.7)
8.0 (8.0–8.3)
8.0 (8.0–8.3)
6.3 (6.0–7.0)
7.5 (7.0–8.0)
8.0 (8.0–8.5)
7.8 (7.6–8.0)
8.0 (8.0–8.0)
8.0 (8.0–8.0)
Median (IQR)
8.1 (0.5)
8.2 (0.5)
8.2 (0.7)
8.2 (0.5)
8.4 (0.4)
8.4 (0.4)
8.2 (0.4)
8.2 (0.4)
6.4 (0.8)
7.4 (0.7)
7.9 (1.0)
7.8 (0.3)
7.8 (0.4)
7.8 (0.6)
Mean (SD)
In situ simulation p*
0.069
0.018
0.017
0.015
0.015
0.015
0.037
0.035
0.024
0.011
0.021
0.165
0.003
0.020
8.0 (6.0–8.0)
7.0 (5.0–8.0)
8.0 (4.0–8.0)
8.0 (4.0–8.0)
7.0 (6.5–8.0)
7.0 (5.6–8.0)
7.0 (6.5–8.0)
7.0 (6.5–8.0)
5.0 (5.0–6.0)
5.0 (4.5–7.0)
8.0 (4.0–8.0)
6.5 (4.0–9.0)
7.0 (4.5–7.0)
7.0 (5.0–8.0)
Median (IQR)
6.8 (2.4)
6.2 (2.2)
6.2 (2.5)
6.2 (2.5)
6.9 (1.2)
6.9 (1.2)
7.1 (0.9)
7.1 (0.9)
5.2 (1.5)
5.6 (1.9)
6.4 (2.2)
6.5 (3.5)
6.0 (1.9)
6.4 (1.8)
Mean (SD)
Decay p*
0.903
0.747
0.968
0.968
0.600
0.687
0.968
0.899
0.384
0.968
0.480
0.472
0.407
0.779
Graph
pà
0.102
0.050
0.044
0.051
0.038
0.047
0.069
0.066
0.093
0.032
0.042
0.169
0.012
0.059
•
7.2 (1.3)
6.1 (1.6)
Didactic
Phase
Miller et al.
Performance as leader ⁄ helper Patient friendliness
5.3 (1.9)
6.0 (5.0–7.5)
Teamwork
Mean (SD)
Median (IOR)
CTS Component Measure
Baseline
Table 1 Clinical Teamwork Scores by Study Phase
610 IMPROVING TEAMWORK AND COMMUNICATION WITH IN SITU SIMULATIONS
ACADEMIC EMERGENCY MEDICINE • May 2012, Vol. 19, No. 5
•
www.aemj.org
Data Analysis All three observers were present for 10 of the 39 measured trauma activations. We used these cases to measure inter-rater agreement using Krippendorff’s alpha, and agreement was excellent (mean agreement of 0.92 with the lowest value being 0.85 for overall decisionmaking).8 We calculated median and interquartile range (IQR) scores for each of the CTS component measures and compared the distribution of the scores across the four phases of the study. Using nonparametric testing procedures (Kruskal-Wallis), we made two principal comparisons. In the first comparison, we analyzed whether the scores observed in the didactic phase, ISTS, or decay phase differed significantly from those of the baseline phase. In this comparison, the baseline phase was treated as a control. In our second analysis, we tested the hypothesis that the scores observed differed across phases. For all analyses, we used SAS software (PROC NPAR1WAY, version 9.3, SAS Institute, Cary, NC). We used a two-tailed Type I error rate of 5% to determine statistical significance. During the study design phase to estimate our needed sample size, we used the following: assuming a baseline CTS score of 5 with a standard deviation of 1.6, we required 12 observations in each group to achieve an 80% power of detecting a 2-point change in CTS score between the ISTS and nonsimulation groups. RESULTS We collected data on a total of 39 trauma activations out of approximately 260 trauma activations that presented during the observation periods. The scores for 11 of 14 measures improved from the baseline to the didactic phase and the mean and median scores of all CTS component measures were greatest during the ISTS phase. In our primary comparison, we contrasted the scores for each of the experimental observation phases (didactic, ISTS, and decay) to the first phase (baseline). When comparing the didactic phase to baseline, only closed-loop communication was significantly different. In comparing the ISTS phase to baseline, nearly all of the communication component measures were significantly improved over baseline, with only the orient new members and patient friendliness component measures not significantly improved from baseline. The decay phase scores were similar to the baseline scores across all measures. In the analysis of our secondary hypothesis (scores differed across phases), only overall communication differed significantly. These data and analyses are displayed in Table 1. The mean time to application of cardiac monitor decreased from 158 seconds at baseline to 142 seconds during the didactic phase and 129 seconds during the simulation intervention phase. This change was not statistically significant. DISCUSSION Previous work has shown promising findings of improved teamwork in teams trained with simulation.5 Our study observed an improvement in teamwork and communication during our simulation intervention
611
phase, but we did not observe a postintervention benefit. We suspected that by minimizing the differences between the setting of our intervention and the setting of its intended effect, we could achieve a more robust change in behavior, so we brought the simulation environment to the location where patient care occurs. The majority of component measures on the CTS increased from the baseline to the didactic phase, and all measures were at their highest during the ISTS phase. It is important to note that this observed effect was lost once the intervention was stopped, suggesting that for simulation’s beneficial effects on teamwork and communication to continue, the simulations must also be continued. In addition to the quantitative results, we note the following qualitative observations. We were able to implement an in situ simulation program in a busy urban ED with excellent participation from all members of the trauma team. Each case required 2 hours of faculty time for set-up, simulation, debriefing, and takedown. Participation in simulations appeared to be facilitated by timing simulations during hours of the day and days of the week that are typically less busy (approximately 10:00 on midweek days). The use of moulage appeared to enhance enthusiasm and interest, and learner feedback confirmed this observation. Members of the team such as nursing and ED technicians expressed an appreciation of being able to discuss teamwork and communication issues with openness during our debriefing sessions that they seldom enjoyed with real patient encounters. These same team members reported that debriefing sessions offered an opportunity to bring to light commonly encountered but rarely mentioned issues (for instance, when multiple physicians issue verbal orders for pain medication on the same patient, it is unclear which should be carried out). Our ISTS program required 36 instructor hours to reach an estimated 25% of trauma team members whereas our didactic intervention required 6 instructor hours to reach an estimated 80% of trauma team members. Further analysis based on individual institution’s needs and available resources are necessary to determine whether the benefit of an ISTS program warrants this increased cost. LIMITATIONS When designing the study we were confronted with the fact that, due to hourly changes in the composition of our trauma team, the study of a control versus an experimental group would create insurmountable logistical difficulties. However, by granting the following assumptions we were able to design a theoretically sound study. The first assumption is based on an observation; our trauma team is not a static group, so to study static groups artificially created to form a control and an experimental arm would be to study a system that does not mirror the reality of trauma care where teams are typically formed in an ad hoc manner and change from one case to the next. Our second assumption was that teamwork and communication, with respect to our trauma teams, are properties that emerge from the interaction of the group as a whole
612
Miller et al.
•
IMPROVING TEAMWORK AND COMMUNICATION WITH IN SITU SIMULATIONS
more so than from the particular individuals that comprise the group. While a single strong personality can have pronounced effects on teamwork and communication, our trauma team is comprised of no fewer than eight health care workers, so we believe that the variability of individual team members has much less of an effect on teamwork and communication than does the overall culture of interaction among the team. Thus, if teamwork and communication are characteristics that emerge from the interactions of the team as a whole, then continuity should exist in these characteristics despite a discontinuity of the individual members comprising the team. Last, we noted that if we were intervening at the level of the team, our measurement should reflect this by looking at the team level rather than the individual level. The CTS measures the team as a whole, allowing us to not have to measure or control for individual team members. In comparing each study phase to baseline, we observed significantly improved scores in 12 of 14 CTS measures during the ISTS phase, but when we tested for differences between all phases, only overall communication appeared significantly different. The loss of significance in the secondary analysis could be due to insufficient power to detect a difference, or it may reflect an actual lack of difference between the ISTS phase and the other phases. Our data were collected as a convenience sample. Because the members of our trauma team all work evenings, nights, and weekends, in addition to the weekday periods that were sampled, this sampling method is unlikely to have affected the composition of our trauma team. However, the types of trauma cases that were sampled may have been affected by not observing during nights and weekends. Additionally, this was a single-center study. Further study is thus necessary to determine whether our findings can be generalized to other centers and to trauma care performed during off hours. Further research with larger sample sizes and more patient-oriented outcomes may be needed to fully assess the effect of an ISTS program on patients in the setting of trauma care. CONCLUSIONS We were able to deploy an in situ trauma simulation program into the clinical environment with participation of all services involved in trauma resuscitations in a busy academic urban ED. Improvements in all component measures of teamwork and communication were
seen during the in situ simulation intervention phase, but only overall communication appeared significantly different when we tested for differences across all phases. All observed benefit appeared to decline after the simulation program was stopped. This study shows that while teamwork and communication in the clinical setting may be improved during an in situ simulation program, these potential benefits are lost if the simulation program is not continued. The authors acknowledge Diane Coday-Ramsay, Lisa Danes, Deborah Ruff, Stephen Lu, Luke Esquibel, the Basic Advance Trauma Computer Assisted Virtual Experience (BATCAVE) and the Scholarship in Education Allocations Committee (SEAC)
References 1. Kohn LT, Corrigan JM, Donaldson MS. To Err is Human: Building a Safer Health System. Washington DC: Institute of Medicine, National Academies Press, 2000. 2. Chief Medical Officer. An Organisation with a Memory. Report of an Expert Group on Learning from Adverse Events in the NHS. London, UK: HMSO, 2000. 3. Rabih AC, DeRubertis BG, Lin SC, et al. Simulation improves resident performance in catheter-based intervention. Results of a randomized, controlled study. Ann Surg. 2006; 244:343–52. 4. Seymour NE, Gallagher AG, Roman SA. Virtual reality training improves operating room performance: results of a randomized, double-blinded study. Ann Surg. 2002; 236:458–63. 5. Shapiro MJ, Morey JC, Small SD, et al. Simulation based teamwork training for emergency department staff: does it improve clinical team performance when added to an existing didactic teamwork curriculum? Qual Saf Health Care. 2004; 13:417–21. 6. Issenberg SB, McGaghie CM, Petrusa ER, Gordon DL, Scalese RJ. Features and uses of high fidelity medical simulators that lead to effective learning: a BEME systematic review. Med Teacher. 2005; 27:10– 28. 7. Guise JM, Deering SH, Kanki BG, et al. Validation of a tool to measure and promote clinical teamwork. Simul Healthc. 2008; 3:217–23. 8. Hayes AF, Krippendorff K. Answering the call for a standard reliability measure for coding data. Comm Meth Measures. 2007; 1:77–89.
