Using Simulation to Identify and Resolve Threats to Patient Safety

Background:

The simulation-based team training used in commercial aviation can provide healthcare professionals with guidance on improving patient safety.

Objective:

To show how in situ simulation can identify latent environmental threats to patient safety. Study Design: Case study.

Methods:

This in situ simulation took place at a large Midwestern hospital in January 2007. It involved a patient with chest pain and hypotension that required cardiac catheterization. The simulation had 2 phases: emergency department and catheterization laboratory. Materials included a patient manikin, a high-definition camcorder, and software for annotating the video in real time. Props (eg, simulated electrocardiogram results, chest x-rays) were used. A Master Scenario Event List was used to orchestrate the entire simulation event.

Results:

Three latent environmental threats to patient safety were identified: procedures for transporting patients between the 2 units, for managing the handoff process, and for organizing the cardiac catheterization process. These were not training issues, but were due to poorly developed or nonexistent procedures that affected the performance of all healthcare teams on those units every working day. The threats were identified by the simulation participants (along with their supervisors) during the post-simulation debriefing as being sufficiently common and dangerous to warrant further review and remedy.

Conclusion:

By conducting our simulations in the actual environment of care, using intact teams of healthcare professionals who practiced their actual technologies and work processes during the simulation, we could identify latent environmental threats to patient safety that could never be explored in an artificial laboratory environment.

(Am J Manag Care. 2010;16(6):e145-e150)

Simulation scenarios that are appropriately designed and performed allow decision makers in healthcare organizations to: n Understand systems, procedures, and protocols that are inhibiting their healthcare professionals from giving the best patient care possible. n Understand and appreciate what is happening in other areas of the healthcare organization so that solutions can be developed that work for the entire organization. n Provide a test bed for newly developed processes and procedures, newly designed or remodeled departments, or any significant change in the healthcare organization before real patients come into contact with those areas.

Since the publication of To Err Is Human,¹ healthcare professionals have increasingly looked to commercial aviation for guidance on improving patient safety. One of the most widely adopted approaches is simulation-based team training (SBT), which merges the human factors content from crew resource management (CRM) training² with the skills practice, performance assessment, and debriefing methods from Line Oriented Flight Training (LOFT).³ Many SBT programs in healthcare take place in simulation laboratories that are designed to replicate operating rooms or labor and delivery suites.^4,5 By contrast in situ simulation takes place on actual patient care units.⁶ This approach has proven quite useful in identifying latent environmental threats to patient safety,7 such as poorly designed equipment, clinical care protocols, and patient handoffs that cannot be explored in the laboratory.

In situ simulation is a unique team training and risk assessment tool that is based on principles of commercial aviation’s Advanced Qualification Program (AQP). Briefly stated, the AQP is a voluntary alternative to the traditional “one-size-fits-all” airline pilot training guidelines that are mandated under Part 121 of the Code of Federal Regulations. The AQP allows participating air carriers some degree of latitude in tailoring their training curricula and retraining schedules to better suit their unique operational needs. In exchange for this freedom, the Federal Aviation Administration requires that participating carriers conduct a rigorous training needs analysis, iteratively revise their training curricula and retraining schedules based on actual pilot crew performance data, and train CRM skills and technical skills together throughout the entire curriculum.^8,9

At the most basic level, the AQP is a data-driven approach to training. Measures of technical and CRM performance are collected immediately prior to the pilot crews’ first LOFT exercise (ie, “first-look” assessments) and again during subsequent LOFT simulations throughout the multiday training and assessment exercise. By systematically comparing the crews’ first-look performance with their subsequent performance, airline safety managers can track how well each crew has improved. They also can search for performance trends to identify likely revisions for the next year’s training curriculum.

Often, the first-look performance assessments suggest safety issues that are not directly related to training, such as conflicting standard operating procedures, dangerous approach patterns, or technical malfunctions. These nontraining issues are further investigated and resolved, because doing so positively impacts all pilot crews within the airline, not just the particular crew in whom the problem was first identified. In this way, the AQP becomes a vehicle for continuously diagnosing and improving the airline’s overall flight safety program, not just its pilot crew training curriculum.

In many ways, our AQP-derived in situ simulation approach is similar to SBT, which was originally developed in the military aviation community.^10,11 In both cases, critical skills are identified via a detailed needs analysis; the simulation events serve as the training curriculum; simulation scenarios are designed to reflect real-world “near miss” events; the teams’ performance is observed and evaluated in real time by trained instructors; and the simulation is followed by a debriefing that uses videotaped examples of the team’s own performance to diagnose areas for improvement.

That being said, some key differences between our in situ simulation approach and more traditional SBT methods bear mentioning here. As the name implies, SBT is first and foremost a training intervention. The entire purpose of the simulation event is to improve the participants’ performance on some set of predetermined skills. Not surprisingly, observed performance problems tend to be limited to various types of individual—or team—errors such as poor decision making, leadership, or communication.

Although training is one use of in situ simulation, it is by no means its primary use. Given the large and heterogeneous nature of healthcare teams—which often involve 10 or more doctors, nurses, anesthesiologists, technicians, pharmacists, and the like—it is unlikely that any hospital could ever hope to effectively train all their healthcare professionals using costly SBT methods. The healthcare industry stands in stark contrast to the commercial aviation industry, in which every pilot receives a mix of classroom training and SBT annually. Given the relative lack of training resources in healthcare, we primarily use in situ simulation as an organizational diagnostic and risk assessment tool. By conducting our simulations in the actual environment of care, using intact teams of healthcare professionals, and by having the participants practice using their actual technologies and work processes during the simulation, we are able to identify latent environmental threats to patient safety that could never be explored in an artificial laboratory environment. Examples include inadequate supervision, maladaptive elements of organizational culture, well-intentioned (but poorly implemented) work processes and procedures, and equipment that does not work as intended. Because these factors typically affect all healthcare professionals within the hospital—not just those who participate in the simulation event—in situ simulation, much like the first-look assessments in aviation, can identify and resolve latent threats⁷ to patient safety that have potentially far-reaching effects throughout the organization.

Another major difference between our in situ simulation approach and traditional SBT methods happens after the simulation event has finished. After the simulation event is concluded, team members are typically debriefed about their strengths and areas for improvement. The debriefing may be led by either a team or the instructor, but all debriefings generally use videotaped examples of the team’s own performance to help focus the discussion.¹² Because our previous experience has shown that team-led debriefings often quickly deteriorate into heated arguments about “who was wrong” rather than “what went wrong,” we draw heavily on United Air Lines’ AQP debriefing approach, in which an expert facilitator helps the team members to diagnose their own performance.¹³

During the debriefing, our facilitator speaks very little by design. He or she uses the Rogerian method¹⁴ to help participants self-reflect on what happened, to diagnose their own strengths and weaknesses, and to identify remedial actions going forward. When the facilitator does speak, the comments are typically limited to 3 main functions: (1) establishing the ground rule that the discussion should focus on learning from mistakes rather than assigning blame; (2) stimulating team discussion by asking open-ended questions (eg, “What was going on here?”) about specific behavioral events that occurred during the simulation; and (3) offering examples from his or her own personal experience and asking the team to comment on how these ideas might apply in this situation. As noted earlier, in many cases, the team identifies latent environmental threats to patient safety such as poorly designed work processes, as well as recommendations for resolving them. In addition, after the debriefing is completed, the facilitator presents high-level findings to the hospital’s senior management so that they can take action to resolve these issues. Taken together, these factors make in situ simulation a uniquely innovative extension of more traditional SBT approaches.

In the following sections, we present a case study involving an in situ simulation at a large Midwestern hospital. The case involved a patient who was experiencing chest pain and hypotension that required cardiac catheterization. After describing the simulation and debriefing methods, we discuss the latent environmental threats to patient safety that were identified, along with subsequent actions taken to correct those issues.

METHODS

Participants

During the emergency department (ED) phase of the simulation, the participants included 1 confederate researcher playing the grandson of the patient, 1 triage nurse, 3 ED nurses, 1 ED resident, 1 ED attending physician, 1 x-ray technician, and 1 pharmacist. The catheterization laboratory phase included 5 catheterization lab nurses, 1 cardiac fellow, and 1 cardiologist.

Materials

The materials included a lifelike patient manikin for simulating the patient’s physiologic condition, a high-definition camcorder for recording the participants’ behavior, and commercial off-the-shelf software for annotating the video in real time. A variety of props, including simulated electrocardiogram (ECG) results and chest x-rays, were used to enhance the simulation’s realism. Finally, a Master Scenario Event List (MSEL) was used to orchestrate the entire simulation event. These materials are described in greater detail below.

Patient Manikin. The patient was simulated using a HAL manikin, which was manufactured by Gaumard Scientific (Miami, FL). A PC laptop with a wireless connection was used to control the manikin’s heart rate, blood pressure, respiration, lung sounds, and reaction to simulated drugs (both oral and injectable) that would normally be administered to heart attack patients. The manikin was operated by a member of our simulation team (a licensed emergency medical technician) to ensure that the manikin’s response would closely mimic that of an actual patient presenting with an acute myocardial infarction.

Audiovisual Equipment. The simulation was recorded using a Sony Handycam high-definition camcorder. The camcorder was connected to an Apple MacBook Pro laptop computer via a standard FireWire cable. The MacBook was running Studiocode video analysis software, which allowed the simulation team to bookmark examples of particularly effective and ineffective team performance (in real time) for subsequent review during the debriefing. After the simulation, these video exemplars were then displayed to the participants using an InFocus portable projector.

Props. The ECG was a realistic ECG printout that was modeled after an actual ST-segment elevation myocardial infarction patient. It was handed to the healthcare team upon request. A case-appropriate chest x-ray was provided. It too was handed to the healthcare team upon request. Realistic lab results were developed prior to the simulation and entered into the hospital’s electronic medical record system so that the simulation participants could search for the results just as they would on any other patient.

Master Scenario Event List. We created an MSEL with 8 interrelated events. Each event included a specific trigger that activated the event, a list of technical and teamwork behaviors that would likely be expected of most healthcare teams, and typical “distractor” conditions to make the event more realistic. In the first event, the patient presents to the ED complaining of chest pain, but there is only 1 healthcare provider in the room. In the second event, a second healthcare provider enters the room, and the 2 of them must work together to jointly diagnose the patient’s condition. In the third event, the team expands and the patient’s condition worsens considerably, requiring a transfer to the cardiac catheterization laboratory. In the fourth event, the formal patient handoff is made to the cardiac catheterization team. The MSEL continues through 4 additional events (some of which may be skipped, depending on the team’s decisions) until the catheterization is performed and the patient stabilizes. The MSEL is integral to providing a structure for the scenario to keep the team focused on the task at hand, but how the team members choose to perform that task (ie, their decision-making process) is entirely up to them.

Design and Procedure

All of the simulation participants were volunteers who knew that a simulation was going to be performed that day, but they did not know any of the specific simulation details. Prior to the simulation proper, all participants were allowed to interact with the manikin and had the opportunity to ask questions about its capabilities and limitations. The simulation began approximately 30 minutes after this familiarization session had ended.

The scenario formally began with the presentation to the ED of an intermittently unresponsive, 55-year-old male patient who had been complaining of chest pain (event 1). The ED team transported the patient to the catheterization laboratory after obtaining an ECG and chest x-ray (event 4). During the transport to the catheterization lab, the patient experienced increased chest pain and developed hypotension (event 5). Once in the catheterization lab, confusion ensued regarding team member roles (event 6). However, once roles were defined and the team began working together, the catheterization was completed and the patient’s blockage was removed (event 8).

The simulation scenario involved both the ED and cardiac catheterization departments. During the postsimulation debriefing session, the simulation team and the healthcare professionals watched the scenario video and discussed their own behavior. They also identified a number of latent environmental threats that had impacted their performance. Based on those observations, the hospital staff took steps to address the threats and improve patient safety. The simulation team conducted follow-up interviews that confirmed the subsequent systems solutions and results presented here. All personally identifying information have been removed to preserve confidentiality.

RESULTS

Because this is a case study, all of the results are qualitative in nature. For ease of interpretation, we present the results in a chronological narrative.

Emergency Department Admission

The simulated patient (hereafter referred to as the patient) was a 55-year-old male who was brought to the ED in a wheelchair by his grandson (a confederate researcher). The grandson reported to the intake nurse that his grandfather had called because he was experiencing severe chest pain. The grandson also reported that the patient had suffered a previous heart attack and was diabetic. At the time of intake, the patient was intermittently unresponsive. The intake nurse called a colleague to page the resuscitation team to room 1. She also told the grandson to go to the registration desk. As the overhead page went out to the ED, the grandson began to wheel his grandfather to registration.

Once the patient arrived at registration and the grandson began registering his grandfather, the nurse came and decided to take the patient into the ED resuscitation room. Once in the room, she briefed the assembling team about the patient’s chest pain and diabetes. She also told them that the patient had a very low pulse and was lapsing in and out of consciousness. An attending emergency physician, a resident emergency physician, and several nurses attended to the patient, who remained unresponsive. The emergency resident reported that the patient was breathing and had a regular heart rate.

The patient regained consciousness and reported nausea and pain in the center of his chest. The nurse ordered an ECG and began setting up for intravenous (IV) delivery. Upon further evaluation by the resident, the patient reported pain in his chest that radiated down his arm and up to his jaw, shortness of breath, sweatiness, nausea, vomiting, and lightheadedness. He described his pain as an 8 out of 10 in severity and unchanged since its onset. The patient stated that he had not taken any aspirin at home. The resident asked the nurse to give the patient aspirin and morphine. Blood pressure was 90/61 mm Hg. Another nurse called cardiology and faxed up the ECG. The nurse announced that 2 IVs had been placed, and despite a saline infusion, blood pressure was dropping and was at 49/31 mm Hg. At this point, the resident asked for a bolus to be given and the patient was given dopamine and heparin. Blood pressure changed to 70/30 mm Hg. The nurse faxed the ECG to cardiology a second time. While they waited for a response from the cardiologist, the resident called for a chest x-ray. Around this point, the patient again reported sharp pain in his chest. The attending physician discussed the patient’s status with the cardiologist, who agreed to take the patient once the chest x-ray was complete.

At this point, the patient’s blood pressure was 71/41 mm Hg. The ED staff prepared the patient for transport while the resident spoke with the patient about the catheterization procedure and obtained his consent. A transport team of 2 nurses moved the patient from the ED to the catheterization lab. The team took the main elevator from the basement-level ED up to the second floor catheterization lab. On the way, the patient reported increased pain, so the transport nurse administered an additional dose of morphine.

First Environmental Threat to Safety. Transporting the patient via the main elevator.

Observed Consequence. Transporting the patient can take several minutes, especially when the transport team has to wait for a free elevator. The transport team can only access the main elevator if other healthcare personnel, patients, and visitors are not already occupying it. Additionally, using the main elevator may result in several stops on floors between the ED and the catheterization lab. Therefore, the wait for the elevator and the extra stops can add several minutes to the transportation time, delaying further patient care.

System Solution. There is now a key for the service elevator at the ED front desk. In an emergent situation, the nurses can use this key to access the service elevator rather than using the main elevator.

Reported Result. In urgent situations, nurses avoid waiting for an elevator and can bypass other floors on the way to the catheterization lab, saving valuable time.

Entrance to the Catheterization Laboratory

The patient’s blood pressure dropped to 49/29 mm Hg as the transport team and patient entered the catheterization lab. The ED nurse helped to prepare the patient for the move from the transport cart to the bed while she briefed the catheterization team on the patient’s status. The ED nurse reiterated that the patient had not been given a beta-blocker because his blood pressure was low. The ED nurse touched the catheterization resident on the arm to ensure that he heard her. As the catheterization team assumed the care of the patient, the ED nurse looked uncomfortable. She quietly asked if anything was needed from her and then quickly left the room.

Second Environmental Threat to Safety. Difficult communication between ED and catheterization laboratory personnel during the handoff.

Observed Consequence. The ED nurse was unsure of where to stand and to whom to give the briefing, because this was not the ED nurse’s typical team or room. Given the uncertainty, the ED nurse may not have provided all the necessary information to the proper individuals before leaving the catheterization lab to return to the ED.

System Solution. A new process was implemented to better manage the handoff. The ED nurse now asks for the appropriate person to whom to give the patient status report. The catheterization team now identifies 1 person ahead of time to take the report from the ED nurse. This person is either the lead nurse or the first nurse in the room; however, the person who takes this role stands near the door to receive the briefing.

Reported Result. The ED nurse now knows immediately to whom he or she should give the report because that individual is located near the door. The ED nurse is able to brief the catheterization lab nurse quickly and appropriately, then leave the room in a timely manner.

Patient Care in the Catheterization Laboratory

Once the patient was in the catheterization lab, 9 healthcare workers entered the room to assist with patient care. The room quickly became loud, and individual team members were initially unaware of their specific role assignments and responsibilities. The doctor explained the procedure to the patient, and several healthcare workers were beginning various preparatory procedures on the patient. Meanwhile, additional healthcare workers continued to be briefed on the patient status as they arrived.

A sterile field was established and the procedure began. Once the procedure began, the room quieted and was more organized around the patient. Arterial access was achieved, preparations were completed, and balloon angioplasty was accomplished, which stabilized the patient’s blood pressure.

Third Environmental Threat to Safety. Undefined team member roles as the catheterization team came together in the room.

Observed Consequence. There was potential for delay in beginning the procedure because the healthcare team did not have a shared mental model for role assignment and task priority.

System Solution. The team identified roles and prioritized tasks for each individual in the room. For example, the catheterization lab nurse taking the patient briefing now stands at the foot of the bed, in the same spot each time, and the ED transport team knows where to look in order to give a report to the appropriate person.

Reported Result. Rather than waiting for all healthcare personnel to arrive in the room to begin working, the catheterization team members are now able to complete specified preparatory procedures based on their role assignment and task priority. Once the patient and all healthcare members are in the room, the catheterization procedure can begin without delay.

CONCLUSION In this article, we have shown how a single in situ simulation at a large, Midwestern hospital’s ED and cardiology catheterization units helped to identify 3 latent environmental threats to patient safety. These threats—procedures for transporting patients between 2 units, for managing the handoff process, and for organizing the cardiac catheterization process—were not training issues. They were due to poorly developed (or altogether lacking) procedures that affected the performance of all healthcare teams on those units every working day. These threats just happened to be observed during this simulation and were identified by the simulation participants (along with their supervisors) during the postsimulation debriefing as being sufficiently common and dangerous to warrant further review and remedy.

We, as the simulation team, recreated an extremely common event using commercially available tools. As the simulation unfolded, we identified behavioral examples that we thought were noteworthy. Subsequently we asked the participants to review, comment on, and suggest the remedies they thought would have the greatest impact. We cannot emphasize enough that the problems were diagnosed and remedied by the participants themselves. The in situ simulation and Rogerian debriefing processes merely provided a vehicle for helping these issues come to the surface. It was all a matter of listening to the participants, who are in the best position to know what works and what does not. These insights, coupled with their commitment to improving patient care and safety, led the healthcare workers at this hospital to generate the solutions and results presented here.

Author Affiliations: From the College of Aviation (WRH, BMB-S, AMG, KO-K, ACG, WF), Western Michigan University, Kalamazoo, MI; Aptima, Inc (JMB), Woburn, MA; and the Department of Emergency Medicine (RLL), Michigan State University / Kalamazoo Center for Medical Studies, Kalamazoo, MI.

Funding Source: Funding for this research was provided by a grant from the Michigan Economic Development Corporation; Battle Creek Unlimited, and The Forest Park Foundation.

Author Disclosure: Dr Hamman and Ms Beaudin-Seiler are board members and part-time employees of the Patient Safety Organization One and report owning stock in the company. Dr Hamman also reports having received grants from the Picker Foundation and having grants pending from the Agency for Healthcare Research and Quality. The other authors (JMB, AMG, KO-K, ACG, RWF, RLL) report no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article. The views presented in this paper are those of the authors and do not necessarily represent those of their sponsoring organizations.

Authorship Information: Concept and design (WRH, BMB-S, JMB, AMG, RWF, RLL); acquisition of data (WRH, BMB-S, JMB, AMG, KO-K, RWF); analysis and interpretation of data (WRH, BMB-S, JMB, AMG, RWF); drafting of the manuscript (WRH, BMB-S, JMB, AMG, KO-K, ACG); critical revision of the manuscript for important intellectual content (WRH, BMB-S, JMB, AMG, KO-K, ACG, RWF, RLL); statistical analysis (WRH); obtaining funding (WRH); administrative, technical, or logistic support (BMB-S, JMB, RLL); and supervision (WRH, BMB-S, RWF, RLL).

Address correspondence to: William R. Hamman, MD, PhD, College of Aviation, Western Michigan University, 237 North Helmer Rd, Battle Creek, MI 49037. E-mail: william.hamman@wmich.edu.

1. Kohn LT, Corrigan JM, Donaldson MS, eds. Committee on Quality of Health Care in America. To Err Is Human. Building a Safer Health System. Washington, DC: National Academy Press; 1999.

2. Helmreich RL, Merritt AC, Wilhelm JA. The evolution of crew resource management training in commercial aviation. Int J Aviation Psychol. 1999;9(1):19-32.

3. Butler RE. LOFT: full mission simulation as crew resource management training. In: Wiener EL, Kanki BG, Helmreich RL, eds. Cockpit Resource Management. San Diego, CA: Academic Press; 1993:231-259.

4. Gaba D, Howard S, Fish K, Smith B, Sowb Y. Simulation-based training in anesthesia crisis resource management (ACRM): a decade of experience. Simulation & Gaming. 2001;32:175-193.

5. Davies JM. Medical applications of crew resource management. In: Salas E, Bowers CA, Edens E, eds. Improving Teamwork in Organizations: Applications of Resource Management Training. Mahwah, NJ: Erlbaum; 2001:265-281.

6. Hamman WR, Rutherford W, Fuqua W, et al. In situ simulation: moving simulation to new levels of realism within healthcare organizations. In: Patankar MS, Ma M, eds. Proceedings of the 3rd Safety Across High Consequence Industries (SAHI) Conference. Vol 1. St. Louis, MO: St. Louis University Press; 2007:64-74.

7. Reason J. Human Error. New York, NY: Cambridge University Press; 1990.

8. Federal Aviation Administration. Advanced Qualification Program. Washington, DC: 1991. Advisory Circular 120-54.

9. Federal Aviation Administration. Special federal aviation regulation 58--advanced qualification program. Federal Register. Vol 55. Washington, DC: National Archives and Records Administration; 1990:40262-40278.

10. Oser R, Cannon-Bowers JA, Salas E, Dwyer D. Enhancing human performance in technology-rich environments: guidelines for scenariobased training. Human/Technology Interaction in Complex Systems. 1999;9:175-202.

11. Prince C, Salas E. Training and research for teamwork in the military aircrew. In: Wiener EL, Kanki BG, Helmreich RL, eds. Cockpit Resource Management. San Diego, CA: Academic Press; 1993: 337-366.

12. Beaubien JM, Baker DP. Post-training feedback: the relative effectiveness of team- versus instructor-led debriefings. Proceedings of the 47th Annual Meeting of the Human Factors and Ergonomics Society. Santa Monica, CA: Human Factors and Ergonomics Society; 2003:2033-2036.

13. Dismukes RK, McDonnell LK, Jobe KK, Smith GM. What is facilitation and why use it? In: Dismukes RK, Smith GM, eds. Facilitation and Debriefing in Aviation Training and Operations. Burlington, VT: Ashgate; 2000:1-12.

14. Rogers C. Freedom to Learn. Columbus, OH: Charles E. Merrill; 1969.

Download PDF: Using Simulation to Identify and Resolve Threats to Patient Safety