by Stephen D. Pratt, MD and Benjamin P. Sachs, MB
In recent years, the medical community has reached a near-consensus that team training and Crew Resource Management (CRM) techniques can improve patient safety. However, the most effective way to teach and implement these concepts is much less clear. Options include high-fidelity simulation and classroom-based systems.(1) Some degree of simulation, or “practice,” is necessary in any teamwork training curriculum. For instance, students may enact a scenario that teaches them to use appropriate, closed-loop communication (this is when the sender initiates communication, the receiver confirms that the communication has been heard and repeats the content, and the sender verifies the accuracy of that content). Alternatively, they may practice conflict resolution or other strategies to improve the culture of safety. However, classroom-based team training does not require an expensive, high-fidelity simulated environment. In contrast, this technique consists of lectures, instructional vignettes or videos, cases reviews, interactive problem-solving exercises, question-and-answer sessions, and examinations to test knowledge.
In our judgment, classroom-based team training with low-level simulation is the most effective way to implement such programs in today’s environment, particularly given the high cost, both in money and manpower, of high-fidelity simulation.
There are two sets of goals when bringing CRM concepts to medicine. The first, and easiest to achieve, is to teach the necessary knowledge, skills, and attitudes (KSAs). The techniques to accomplish this are well known. Knowledge can be taught by didactic lectures, case-based teaching, on-line or print texts, interactive scenarios, or other similar techniques. It can be assessed by pre- and post-tests. Both class-based and simulator systems can effectively teach the basic KSAs. Class-based training has been shown to positively influence team attitudes (2) and skills.(3)
When used as described above, class-based training offers three advantages in this teaching phase. First, it does not need an expensive, specialized environment. The purchase price of a high-fidelity obstetrics simulator can easily reach into tens or even hundreds of thousands of dollars for the most realistic models. In addition, staff must be employed to operate and maintain the simulator. The purchase and operating costs lead directly to high team-training implementation costs. At our institution, the direct cost of a 1-day session at the obstetrics simulator is $950 for attending physicians and $150 for nurses. The direct costs for us to train our staff of more than 200 people would have been more than $85,000 for just the teaching phase! This does not include training residents, unit coordinators, scrub technicians, or others who are part of the team, but who are frequently not included in the simulator-based scenarios. Alternatively, training our entire staff in CRM techniques using a 4-hour classroom course required small costs for copying educational materials, the manpower costs to get trainers out of their clinical environments (a total of about 100 person-hours), the costs of meals we provided at the 4-hour sessions, and other small miscellaneous expenses.
Gaba argues that high-fidelity simulation is superior to other educational methods, contending that the “realistic” environment of the simulator improves the learning environment and increases the opportunity to practice the KSAs. However, this has never been demonstrated. Several authors have suggested that low-level simulation, as could be used in a classroom, is likely to be as effective in teaching teamwork techniques as the costly high-fidelity variety.(4-6) Gaba has even raised concerns that simulators may be relegated to simply practicing difficult or infrequently performed tasks.(7) Even if simulation does improve the initial learning of the KSAs, one must wonder whether the purported advantages of the simulated environment are worth the large expense.
The second advantage of classroom-based education is that it allows more staff to be trained at once. Up to 40 people per session have been effectively trained in one model (2), which comports with our experience as well. Finally, class-based teaching is easy to schedule as part of the staff orientation process, as it does not require staff to leave their environment in order to get to the simulator. Thus, new staff can quickly and efficiently be taught the KSAs they must know in order to work on a unit steeped in CRM concepts.
The second goal of team training is to successfully transfer the KSAs to the clinical arena and to improve patient safety. This is the more important and the more difficult task. It is unlikely that any single intervention, either classroom or simulator-based, is likely to effect a permanent change in the culture of medical care. The post-teaching implementation phase consists of several steps. First, the teamwork concepts are rolled out in stages. This allows staff to integrate them into their practice as a group and to practice them one at a time. The roll-out plan should be carefully scheduled in order to keep staff from being overwhelmed by the multiple changes. For instance, simply having team meetings each shift may be the goal for several weeks. This can be followed by other team behaviors (closed loop communication, task assistance, cross-monitoring, etc). Coaching is the second step in the implementation process.
Teamwork champions are placed on the unit to coach the behaviors as they are rolled out. This reinforces the KSAs learned during the teaching phase and helps to translate them into clinical practice. The coaching effort may add cost, as the coaches might have to be compensated for their time spent on the unit, but these funds can be taken from monies saved on the teaching phase. Finally, encouragement and feedback, especially from leadership, are essential to the implementation process. This can come in the form of sharing teamwork success stories or in sharing data collected from coaches on the success of the implementation process. Based on our experience, this implementation process takes months. It requires constant reinforcement of the KSAs learned in the classroom and ongoing coaching and mentoring by clinical champions committed to the process. These champions ensure that structured meetings occur throughout the day to update the teams’ situation awareness of the every changing environment. They teach and coach CRM concepts such as cross-monitoring, conflict resolution, effective communication, and others in real time and in the environment in which they are needed.
The difficulties in implementation of team training highlight one of the differences between medicine and aviation (see related Perspective and interview); namely, the “flight” is not always well defined in medicine. Procedure-based units such as the operating room or the GI suite are easily compared to aviation. They have planned procedures, with a scheduled time, expected course, and projected end point. They have a well-defined “cockpit” crew. These units can make use of many CRM concepts such as “pre-flight” briefings, checklists, and structured moments (“time outs”) for error reduction. In these types of units, practicing the pre-procedure processes in a simulated environment might be helpful in reducing error (although one must still question whether costly, high-fidelity simulation is necessary).
However, in most medical environments (eg, the emergency department, labor and delivery, medical wards, ICUs), no such schedule exists. These environments require that teamwork concepts be applied to the management of the entire unit, not simply to the patient or procedure, in an effort to reduce, mitigate, and effectively respond to adverse events. Staff must learn to develop situation awareness for and cross-monitoring of multiple caregivers, caring for multiple patients with multiple medical issues in a dynamic, unscheduled environment. One cannot have a pre-determined checklist in the emergency department because it is not possible to know what the emergency will be, when it will arrive, or what resources it will require. Rather, team training in these environments must create a structure and a culture that encourage the tenets of teamwork. Teamwork has to be more fluid and able to adapt to the many “take-offs” and “landings” and mid-air crises that occur without a defined start or end. We believe that high-fidelity simulation is highly limited in its ability to effectively teach this CRM-based unit management, partly because it has overemphasized crisis management (ie, the management of major adverse events instead of the management of the resources, workflow, and teamwork on a unit). Although this could be changed, it would require a significant philosophical shift in the way simulators are run. In addition, it would simply be cost prohibitive to build simulated environments with multiple patients and many caregivers.
In contrast, class-based team training has been shown to change behavior and improve patient outcomes in the clinical arena. Morey and colleagues demonstrated improved teamwork behaviors and clinic outcomes in emergency departments where staff underwent MedTeams training.(3) The group at Vanderbilt has found that class-based team training has lead to dramatic improvements in several clinical outcomes (personal communication with Dr. Rhea Seddon). In our own experience over the last 5 years, we have seen a 25% overall reduction in adverse obstetric events (53% reduction among women who deliver before 37 weeks) and a 52% reduction in malpractice claims and observation cases leading to a $4.9 million dollar reduction in reserve dollars allocated. Although proving that these outcomes were directly caused by our classroom-based CRM program is difficult, this was the major intervention over this period, and we believe that causality is highly likely.
Ultimately the best way to teach and implement team training may be through a combination of classroom training to teach the KSAs and simulator-based training to practice crisis management. However, given the current costs and as-yet unproven benefits of high-fidelity simulation, we believe that classroom-based training, followed by intensive coaching during implementation, is the most effective method of bringing CRM concepts to most medical environments.
Stephen D. Pratt, MD
Instructor in Anesthesia, Harvard Medical School
Director, Obstetric Anesthesia, Beth Israel Deaconess Medical Center
Benjamin P. Sachs, MB, BS
H. Rosenfield Professor, Harvard Medical School
Professor, Harvard School of Public Health
Chief of Obstetrics and Gynecology, Beth Israel Deaconess Medical Center
Back to Top
1. Baker DP, Gustafson S, Beaubien JM, Salas E, Barach P. Medical team training programs in health care. In: Advances in Patient Safety: From Research to Implementation. Volumes 1-4. Rockville, MD: Agency for Healthcare Research and Quality; February 2005. AHRQ Publication Nos. 050021 (1-4). Available at:
http://www.ahrq.gov/downloads/pub/advances/vol4/Baker.pdf. Accessed February 21, 2006.
2. Grogan EL, Stiles RA, France DJ, et al. The impact of aviation-based teamwork training on the attitudes of health-care professionals. J Am Coll Surg. 2004; 199:843-848.
go to PubMed ]
3. Morey JC, Simon R, Jay GD, et al. Error reduction and performance improvement in the emergency department through formal teamwork training: Evaluation results of the MedTeams project. Health Serv Res. 2002;37:1553-1581.
go to PubMed ]
4. Salas E, Bowers CA, Rhodenizer L. It is not how much you have but how you use it: toward a rational use of simulation to support aviation training. Int J Aviation Psych. 1998;8:197-208.
5. Beaubien JM, Baker DP. The use of simulation for training teamwork skills in health care: how low can you go? Qual Saf Health Care. 2004;13(suppl 1):i51-i56.
go to PubMed ]
6. Salas E, Wilson KA, Burke CS, Bowers CA. Myths about crew resource management training. Ergonomics in Design. Fall 2002;Q4:20-24.
7. Gaba DM. The future vision of simulation in health care. Qual Saf Health Care. 2004;13(suppl 1):i2-i10.
go to PubMed ]