The Use of In-Situ Simulation to Improve Safety in the Plastic Surgery Office: A Feasibility Study

Design

With approval as an exempted study from the Beth Israel Deaconess Medical Center investigational review board, we conducted a needs assessment, planned pre- and postsurveys, designed scenarios, and structured debriefings to meet the needs of the office-based personnel and to conform to existing regulations of several accrediting agencies.

Office setting and personnel

The in-situ simulations were conducted in 2 free-standing office-based surgical centers in the Boston area, both of which specialized in plastic surgery procedures. Each facility performed about 1000 procedures per year. These procedures were performed using both general anesthesia and deep sedation as well as conscious sedation techniques. Both centers had previously received nationally recognized accreditation and had incorporated various safety equipment, protocols, and standards as outlined by these accrediting agencies. All medications and equipment used in the scenarios, for both routine and emergency situations, were the property of the office and remained in their normal locations for all of the scenarios. The only exception to the usual routine was the presence of a human patient simulator (Laerdal SimMan, Stavanger, Norway) with its monitor/emulator in place of the usual vital signs monitor. In one case, the automated external defibrillator was altered to allow the conductive pads to clip onto the mannequin during defibrillation.

Office-based personnel consisted of reception staff, billing staff, managers, nurses, medical assistants, and surgeons. The level of training in emergency procedures was commensurate with each individual's professional training, although the lay office staff was current in their certification of basic life support (BLS).

Needs assessment

A specific, office-based Needs Assessment was performed during 2 initial visits to the facility prior to the in-situ simulation session.²⁷ To maintain the overall confidentiality of the staff as well as the secrecy of the nature of the simulations, only 1 surgeon and 1 nurse were involved in the initial visit. The general outline of the Needs Assessment form as well as specific requests from the office personnel helped to shape the eventual scope, topics, and debriefing focus for the simulation exercises, as shown in Table 1. A second visit was performed to convey a general schedule of events and to allow the simulation and audiovisual experts to confirm the location of the electrical outlets, camera positions, debriefing rooms, and projection equipment.

Simulation design

Several scenarios were developed to address perceived knowledge deficits and to strengthen team dynamics during complex medical emergencies. These scenarios were based on both the general requirements determined by the Needs Assessment as well as the specific suggestions during the previsits. The scenarios were crafted to represent technically challenging tasks to the medical personnel as well as require urgent actions from the administrative staff members.²⁸ In one scenario, for example, the patient experienced malignant hyperthermia (MH) during administration of a general anesthetic, as shown in Table 2. In another scenario, a surgeon and a surgical nurse were asked to excise a lipoma model and close a wound using subcuticular sutures. The lipoma model was conceived and built by our simulation experts and validated by local surgeons. In both scenarios, an actor represents the patient when checking in to the receptionist and while in the preoperative room.

Table 2. Sample scenario template: malignant hyperthermia

Once in the procedure room, the patient was represented by the human patient simulator. The scenario in the procedure room was designed to consist of several stages, but the transition to each stage as well as alterations in the scenario could be managed by the simulation experts and were reflective of the actions performed by the office personnel. In general, each scenario lasted 20 minutes. The scenarios ended when sufficient therapeutic maneuvers had been performed or when the outside emergency medical teams had arrived to assume care for the patient.

Each scenario was followed by a structured debriefing. The debriefings were attended by all of the office personnel and had several overall goals (Table 3). All of the participants were asked to contribute to the discussion, which dealt with key concepts of Crisis Resource Management (CRM), video review of the scenario, issues of communication, and the role of each participant in maintaining the culture of safety in the office. At the end of the training session, all participants gave both verbal and written comments about the strengths and weaknesses of the simulator sessions.

Evaluation

While individual performances were not assessed and no ratings were given, each debriefing did focus on examples of individual, team, and systemic problems that could impede rapid resolution of a crisis. Using video review of the scenarios, the debriefer could isolate specific instances of communication to either demonstrate useful examples or to encourage the group to suggest improvements in the choice of words or phrases. For example, team actions could be compared with existing ACLS (Advanced Cardiovascular Life Support) algorithms. The performance of certain tasks such as mask ventilation or chest compressions could be assessed for proper rate and chest excursion. Systemic issues such as the appropriate emergency telephone numbers, adequate written protocols, and explicit ingress and egress routes could be discussed in the context of the recorded scenario. Finally, a comparison between the performance of the office team could be made with existing benchmarks and specific recommendations of accrediting agency regulations.

Prior to the exposure to the 2 simulation exercises, all participants were administered the abbreviated version of the AHRQ Medical Office Survey on Patient Safety.²⁶ This nationally validated survey tool is specifically designed to measure the culture of patient safety in medical offices from the perspectives of providers and staff. This survey is intended to measure to which extent the organization's culture emphasizes patient safety, teamwork, discussion of mistakes, and continuous improvement and learning. The same survey was then readministered after the completion of the 2 simulation exercises following the debriefing session.

Statistical analysis of survey results

Responses from the survey were presented as frequencies and percentages. For the matched pairs, McNemar tests or Stuart-Maxwell tests were performed to examine the differences in participant survey responses before and after simulation sessions. The proportional differences among all respondents were compared using Fisher exact tests. All statistical tests were 2-sided, with a type I error of 0.05. A P value of less than 0.05 was considered to be statistically significant. All statistical analyses were performed with SAS version 9.3 (SAS Institute, North Carolina).

Overview

A total of 16 personnel in 2 plastic surgery offices participated in the half-day, in-situ simulation training exercise. After a brief introduction to the proposed schedule and a tour into the procedure room to meet a “healthy” mannequin on the operating table, the participants completed a safety culture survey and decided whether to allow for video recording of the sessions. The standardized (confederate) patient for the first scenario then arrived at the reception area and proceeded with the check in. The patient changed clothes and walked to the preprocedure area. One of the office nurses completed a screening questionnaire and physical examination. The surgeon and anesthesiologist then made their individual introductions, explanations and completed their own paperwork with the patient. The standardized (actor) patient then walked into the procedure room and a similarly dressed mannequin assumed the role of patient. After appropriate monitors were applied, the patient was prepped and draped for the procedure. A time out was performed. The specific stages of the scenario were then performed.

Survey outcomes

Survey results were obtained from all participants who were administered the abbreviated version of the AHRQ Medical Office Survey on Patient Safety, both prior to case scenarios and after the completion of the 2 scenarios (Table 2) and the debriefing session (Table 3). The pre- and postintervention responses showed a statistically significant difference in a few categories, as shown in Table 4. Specifically, the participant survey indicated greater awareness of patient safety issues following participation in simulation and debriefing exercises in 3 areas (P < 0.05): the need to change processes if there is a recognized patient safety issue (100% vs 75%), openness to ideas about improving office processes (100% vs 88%), and the need to discuss ways to prevent errors from recurring (88% vs 62%). However, the rest of the survey responses did not show statistically significant differences between pre- and postintervention.

Clinical outcomes

Two scenarios were demonstrated during the simulation exercise. The first was a toxic reaction to a local anesthetic injection, which resulted in ventricular tachycardia and pulseless electrical activity. The second case was a reaction to general anesthetics known as MH.²⁹ In both cases, the anesthesiologist and surgeon needed to establish good communication and to orchestrate a larger response to the crisis. Clinical outcomes were deliberately varied and related to the speed of the response to ventricular tachycardia and the ability to assess the efficacy of therapy and the potential side effects from therapeutic intervention. For example, repeated doses of amiodarone would result in hypotension, whereas early cardioversion would lead to pulseless electrical activity. Thus, the participants were forced to reassess and try alternative therapies to better manage the patient. In the case of the MH patient, the vital signs were allowed to deteriorate until a full dose of 5 mg/kg of dantrolene was given. The timely delivery of such a large dose required the coordinated effort of between 3 to 8 people. Any delay in assembling the office staff to help with resuscitation resulted in longer periods of hypotension.

Team-based processes

Team-based processes were evaluated according to the central principles of CRM as described by Gaba et al.^30,31 Principles of Role Clarity, Communication, Personnel Support, Resources, and Global Assessment were outlined and described before the video debriefing.³² During the debriefing, the participants were encouraged to categorize the clinical elements and actions in terms of these principles of CRM and to suggest alternatives or improvements in words or actions. In all of the scenarios, there were issues relating to the recognition of a crisis, organization of a team leader, setting of priorities, and the provision of timely feedback. In some cases, the use of an algorithm (eg, ACLS) was used to compare actions against an accepted standard. In other cases, the algorithms were faithfully followed but one or several participants became overwhelmed by the number of clinical tasks and were not assisted by other, less busy members of the team. Individual skills and behaviors were reviewed to ensure that all participants stayed within their own level of training and scope of practice.^33,34

Systems-based issues

As team behaviors became bogged down, there appeared systems issues that contributed to the frustration and ineffectiveness of the participants. In the MH case, for example, there was a lack of large syringes and large bore needles in the accompanying kit. This lack of an adequate system slowed the rate at which dantrolene was able to be solubilized and administered to the patient. Also in the MH case, the lack of routine temperature monitoring frustrated the anesthesia provider and this lack of monitored temperature reading delayed the recognition of MH. While an increase in temperature is usually a late sign of MH, the participant ignored several early signs such as tachycardia, hypercarbia, and acidosis. Several minutes elapsed before the emergency was declared, and the appropriate emergency medical systems were activated.

Once activated, the medical systems worked with varying degrees of effectiveness. A prepared kit for the treatment of MH was already in the procedure room and there were enough vials of dantrolene, but not enough syringes to reconstitute and administer the medication. Nurses were unfamiliar with the medication, the diluent, and the initial and total doses. As a result, the anesthesia provider had to guide the nurses but, in turn, failed to recognize electrocardiographic signs of hyperkalemia and of ventricular ectopy. Several of the office staff left to procure more dantrolene, but the operating room personnel were unaware of their whereabouts or their progress. In general, the scenario suffered for lack of an event manager who was able to fully coordinate multiple activities.

During the debriefing, the office staff was able to identify further flaws in the systemic response to the emergency. After dialing 9-1-1, for example, the ambulance would arrive to the parking lot without knowing how to proceed to the procedure room. Even an enhanced ambulance emergency response system, such as the one that was available, does not provide directions within the building. The staff suggested that one of the reception staff would need to be in the parking lot to receive the ambulance and to direct the medical personnel. A route that would accommodate the width of a wheeled stretcher with a patient was devised. The need for periodic updates from the event manager in the procedure room to all the staff was emphasized, and the importance of the nonmedical personnel was reinforced.

Generating recommendations

Following the scenarios and after a debriefing session that stressed the importance of leadership, communication, situational awareness, and elements of resuscitation, a number of specific recommendations were made to improve the process of crisis management (Table 5). These 5 recommendations came from open-ended questions in the follow-up survey given to the participants 1 month after the exercise. Three of the comments were centered on emergency algorithms that were discussed in the postscenario debriefings. Two of the comments related to continuation of deliberate practice of simulation and communication exercises. The speculated root-cause of these comments may have been derived from the group's performance in the simulation exercises and subsequent reflection on ways to improve.

All of the participants recognized the requirements for a coordinated manager of the medical emergency and for ready access to the written algorithms or protocols to help to plan for such an event. Most felt that their knowledge of ACLS was “rusty” and welcomed a refresher course or a series of simulations to periodically practice these resuscitative skills. While doctors and nurses were current with respect to ACLS certifications, the nonmedical office staff were not all current with BLS and felt that regular practice of BLS would be essential to be comfortable during a medical crisis.

In addition to the regular practice of appropriate ACLS and BLS algorithms, the office-based group discussed several other improvements in their systems-based approach to a medical emergency (Table 4). The office staff requested a regular review of the items in the code cart with a brief description of how they work and under what situation one might use them. The staff further asked for repeated simulation exercises to demonstrate other anticipated office-based problems. Finally, the office voiced a need to practice effective and closed-loop communication on a regular and nonurgent basis, to make it a habitual part of the working environment.