Use of simulation models when developing and testing hospital evacuation plans: a tool for improving emergency preparedness.

BACKGROUND: In recent decades, analyses of hospitals evacuations have generated valuable knowledge. Unfortunately, these evacuation case studies often lack crucial details and policies that would be helpful in evacuation preparedness. The aim of this study was to use a simulation model to illustrate how it can aid emergency planners in the development, testing, and revising of hospitals evacuation plans. This study includes evacuation exercises at two emergency hospitals in Region Stockholm, Sweden. METHODS: A scientifically validated simulation system for "table top" exercises was used for interactive training of hospital medical staff, prehospital staff and collaborating agencies. All participants acted in their usual professionals' roles. The exercises were run in real-time and mirrored actual hospital resources with the aid of moveable magnetic symbols illustrating patients, staff and transport, presented on whiteboards. During the exercises, observers and independent instructors documented actions taken and post-exercise surveys were conducted to obtain reactions and compare results. RESULTS: The simulation system allowed the emergency planner to test the whole evacuation process, making it possible to train and evaluate the important functions of management, coordination, and communication. Post-exercise surveys explored participants perception of the exercises. Analysis of open-ended questions included areas for improvement and resulted in five main categories: (1) management and liaison; (2) communication; (3) logistics; (4) medical care and patient prioritisation; and (5) resource utilisation. CONCLUSIONS: This study has shown that "table top" exercises using a validated simulation system can serve to guide emergency planners when developing evacuation plans, procedures, and protocols as well in training of all medical staff. The system also served to train adaptive thinking, leadership, communication, and clarification of critical functions.

A method for detailed determination of hospital surge capacity: a prerequisite for optimal preparedness for mass-casualty incidents.

BACKGROUND: Defined goals for hospitals' ability to handle mass-casualty incidents (MCI) are a prerequisite for optimal planning as well as training, and also as base for quality assurance and improvement. This requires methods to test individual hospitals in sufficient detail to numerically determine surge capacity for different components of the hospitals. Few such methods have so far been available. The aim of the present study was with the use of a simulation model well proven and validated for training to determine capacity-limiting factors in a number of hospitals, identify how these factors were related to each other and also possible measures for improvement of capacity. MATERIALS AND METHODS: As simulation tool was used the MACSIM® system, since many years used for training in the international MRMI courses and also successfully used in a pilot study of surge capacity in a major hospital. This study included 6 tests in three different hospitals, in some before and after re-organisation, and in some both during office- and non-office hours. RESULTS: The primary capacity-limiting factor in all hospitals was the capacity to handle severely injured patients (major trauma) in the emergency department. The load of such patients followed in all the tests a characteristic pattern with "peaks" corresponding to ambulances return after re-loading. Already the first peak exceeded the hospitals capacity for major trauma, and the following peaks caused waiting times for such patients leading to preventable mortality according to the patient-data provided by the system. This emphasises the need of an immediate and efficient coordination of the distribution of casualties between hospitals. The load on surgery came in all tests later, permitting either clearing of occupied theatres (office hours) or mobilising staff (non-office hours) sufficient for all casualties requiring immediate surgery. The final capacity-limiting factors in all tests was the access to intensive care, which also limited the capacity for surgery. On a scale 1-10, participating staff evaluated the accuracy of the methodology for test of surge capacity to MD 8 (IQR 2), for improvement of disaster plans to MD 9 (IQR 2) and for simultaneous training to MD 9 (IQR 3). CONCLUSIONS: With a simulation system including patient data with a sufficient degree of detail, it was possible to identify and also numerically determine the critical capacity-limiting factors in the different phases of the hospital response to MCI, to serve as a base for planning, training, quality control and also necessary improvement to rise surge capacity of the individual hospital.

Assessment of the accuracy of the Medical Response to Major Incidents (MRMI) course for interactive training of the response to major incidents and disasters.

BACKGROUND AND AIMS: The benefit of simulation models for interactive training of the response to major incidents and disasters has been increasingly recognized during recent years and a variety of such models have been reported. However, reviews of this literature show that the majority of these reports have been characterized by significant limitations regarding validation of the accuracy of the training related to given objectives. In this study, precourse and postcourse self-assessment surveys related to the specific training objectives, as an established method for curriculum validation, were used to validate the accuracy of a course in Medical Response to Major Incidents (MRMI) developed and organized by an international group of experts under the auspices of the European Society for Trauma and Emergency Surgery. METHODS: The studied course was an interactive course, where all trainees acted in their normal roles during two full-day simulation exercises with real time and with simultaneous training of the whole chain of response: scene, transport, the different functions in the hospital, communication, coordination, and command. The key component of the system was a bank of magnetized casualty cards, giving all information normally available as a base for decisions on triage and primary management. All treatments were indicated with attachments on the cards and consumed time and resources as in reality. The trainees' performance was recorded according to prepared protocols and a measurable result of the response could be registered. This study was based on five MRMI courses in four different countries with altogether 235 participants from 23 different countries. In addition to conventional course evaluations and recording of the performance during the 2 exercise days, the trainees' perceived competencies related to the specific objectives of the training for different categories of staff were registered on a floating scale 1-10 in self-assessment protocols immediately before and after the course. The results were compared as an indicator of to which extent the training fulfilled the given objectives. These objectives were set by an experienced international faculty and based on experiences from recent major incident and disasters. RESULTS: Comparison of precourse and postcourse self-assessments of the trainees' perceived knowledge and skills related to the given objectives for the training showed a significant increase in all the registered parameters for all categories of participating staff. The average increase was for prehospital staff 74 percent (p<0.001), hospital staff 65 percent (p<0.001), and staff in coordinating/administrative functions 81 percent (p<0.001). CONCLUSIONS: The significant differences in the trainees' self-assessment of perceived competencies between the precourse and postcourse surveys indicated that the methodology in the studied course model accurately responded to the specific objectives for the different categories of staff.

Comparative study of physiological and anatomical triage in major incidents using a new simulation model.

OBJECTIVES: To develop and evaluate a simulation model making it possible to evaluate the accuracy and efficiency of different triage methods; to compare the results of physiological and anatomical triage performed by medical staff with different levels of skills with the use of this model. DESIGN AND OUTCOME MEASURES: A simulation model was created based on patient cards giving sufficient physiological data as a base for physiological triage and anatomical data as description of findings at exposure, providing a base for anatomical triage. Three groups with different skills in disaster medicine, nurse students (n = 23), ambulance nurses (n = 20), and surgeons (n = 30), performed triage based on the patient cards. The outcome was given as potential avoidable mortality. The results of the triage for the two methods were compared to the result of the same triage performed by an expert group. RESULTS: Differences in triage: Within the groups, the difference between the two triage methods was only significant for the surgeons (p < 0.001), who had a better result using the anatomical triage. For the "physiological triage," there were no significant differences between the three groups. Regarding the results for the "anatomical triage," there were significant differences between both the nurse students and the surgeons (p < 0.001) and the ambulance nurses and the surgeons (p < 0.05). Results in distribution of patients and potential avoidable mortality: Within the groups, the difference between the two methods was significant for all the groups (nurse students, p < 0.01; ambulance nurses, p < 0.01; and surgeons, p < 0.001). They all had a better outcome with anatomical triage (nurse students, 6.1 percent; ambulance nurses, 6.1 percent; and surgeons 19.5 percent less mortality than physiological triage). The group that made the best outcome from physiological triage was the ambulance nurses who had a significantly better result than both nurse students (p < 0.01) and surgeons (p < 0.001). The mean mortality rate for ambulance nurses was 31.1 percent, nurse students, 37 percent, and surgeons was 38.1 percent. Regarding the anatomical triage, there was a significant difference (p < 0.001) between the groups of nurse students (30.9 percent mortality) and surgeons (18.9 percent mortality). The differences between the rest of the groups were also significant (p < 0.05). CONCLUSIONS: The model developed for this study made it possible to compare different methods of triage and also triage performed by staff of different levels of training and experience. Anatomical triage for all test groups in this study gave significantly better results than physiological triage regarding calculated outcome and this difference increased with increasing experience.