Interprofessional and interdisciplinary simulation-based training leads to safe sedation procedures in the emergency department

The first step of the education concept is independent study of a hand-out that has to be completed before the simulation, providing theoretical information about sedation, the ED sedation concept and underlying guidelines. The aim of this step is to familiarise participants with the general concept and to ensure a baseline knowledge about the topic. With regards to content, our concept is based on the guidelines of the American College of Emergency Physicians Clinical Policies Subcommittee [15]. Because the discussion about the best medication used for sedation procedures in the ED (ketamine, midazolam, propofol, fentanyl) is still on-going [16‐18], we deliberately did not dictate any specific medication, but demonstrated the advantages and disadvantages of all available medication and provide a standardized framework for the safe conduction of sedation, regardless of the specific medication used.

The training day, as a second step, starts with a short theoretical repetition of the concept and focuses on questions of the participants. This step was introduced to provide participants with the opportunity to clarify any questions that might have arisen during step 1 and to lower the threshold to engage in the simulation activities of step 3.

The third step consists of three simulated scenarios with supervision, interposed with airway skills training. This step is conducted to practice the concepts introduced during the first two steps and to provide participants with the opportunity to train in sedation as well as to increase their familiarity with complications that might arise.

The case scenarios were practiced in real work surroundings with an interprofessional real life team consisting of physicians and nurses. All simulation cases are practiced with a computer-enhanced mannequin simulator (SimMan 3G, Laerdal Medical, Norway). According to our concept, a sedation team consists of at least two trained staff members, including at least one physician. After each case scenario, medical issues as well as human factors were debriefed using video-assistance by an interprofessional (nurse/physician) and interdisciplinary (emergency medicine/anaesthesia) instructor team. Each instructor (6 in total) had a minimum of 1 year of experience in simulation/debriefing and was trained in a simulation instructor course of several days duration. One supervising instructor was present at all training courses in order to ensure continuity in teaching techniques, as well as in the medical contents.

A special focus of our simulation scenarios was on the prevention of possible complications and on emergency procedures. We emphasised the importance and implementation of capnography, as recommended in procedural sedation guidelines and research [19, 20]. The training session included the use of a structured pre-sedation checklist (available on request), including a team timeout, as well as the development and documentation of emergency strategies.

In the airway skill training, we formed small groups of 2–3 participants. Each group was trained using an Adult Airway Management Trainer Manikin (Laerdal, Norway). The airway skills course focussed on bag-valve ventilation techniques (single and two-handed approach), with simple tools such as nasopharyngeal and oropharyngeal airway devices, including insertion techniques, indications and pitfalls. As a rescue tool, the insertion of an laryngeal mask was taught. We did not teach orotracheal intubation in order to lower participants’ anxiety and - most importantly - because the necessary routine to master orotracheal intubation safely would be very difficult to ensure and maintain in our ED setting. All airway tools introduced during training, as well as the necessary medication, are now provided in the ED in a special sedation procedure box and all participants have been familiarised with the equipment provided in the ED through the training.

In the simulation based team training courses, we used three standardised scenarios (available upon request). In each scenario, the focus is on a special medical topic frequently encountered in our ED (e.g., luxated shoulder), as well as on crisis resource management (CRM). For crisis resource management, we used the CRM principles as conceptual framework, as outlined by Gaba and Rall [21‐24]. Regarding medical factors, we focused on the main complications that arise in sedation procedures (apnoea, circulatory problems and potentially obstructed airways).

Before and after the simulation-based training course, we used a custom questionnaire to assess confidence and familiarity with the sedation concept, awareness of emergency procedures, knowledge of the sedation medication and knowledge of CRM principles. The acceptance of the specified statements had to be indicated on a 11-point Likert scale ranging from 0 (completely disagree) to 10 (completely agree). All measurement instruments are available upon request.

As the last and fourth step, the simulation-based training day was followed by a separate individual practical learning experience in the department of anaesthesia, where participants further practiced bag-ventilation and the use of the simple airway tools they were familiarized with during training.

Because of the suspected infrequency of sedation procedures and subsequent possible lack of regular experience, we conduct monthly refresher trainings of 3 h duration for all trained staff.

Starting with the roll-out of the program, it is now mandatory to conduct procedural sedation in our ED in accordance to the trained standards. Every sedation procedure is under on-going evaluation through a questionnaire (available upon request) and the documentation in the patients chart. General factors (type of intervention, administered medication, timeline of the intervention and ER visit as well as reported adverse events) are routinely recorded in the patient chart. Overall satisfaction of the patient, as well as pain experienced during the procedure are monitored. The perceived usefulness of the sedation training for the actual sedation team is further assessed. Additionally, we screened the charts of all patients who underwent a sedation retrospectively for adverse events. To evaluate the usefulness of the sedation program for ED workflow, we compared age, gender, the American Association of Anaesthesiologists - scale (ASA), time to procedure (defined as time until procedure was begun), time for procedure (defined as duration of procedure) and time to discharge in patients sedated for the reduction of a luxated shoulder by the trained ED team in 2015 with the patients treated by anaesthesia teams in 2014.

The study protocol was assessed by the Ethics Committee of the Canton Bern, Switzerland (Req-2016-00134) and was classified as quality control investigation. Therefore no informed patient consent is necessary according to Swiss law.

We compared participants by age group, years of working experience, gender and profession by means of t-test (age, experience) and Fishers exact test (gender) as appropriate. Self-assessed knowledge and confidence before and after the training were compared using paired samples Student’s t tests. Time to procedure, time for procedure and time to discharge between patients sedated by the ED team and those cared for by teams of the Department of Anaesthesia were compared by means of Mann–Whitney-U test for unrelated samples, their ASA-classification by means of a univariate ANOVA. All calculations were performed in SPSS Statistics 21 (IBM Coorp.). A p-value of less than 0.05 was considered statistically significant. We calculated Cohen’s d as effect size for all statistically significant tests in unrelated samples and d _z for related samples.

We found a highly significant increase in self-efficacy from before to after training in all participants, as well as in all subgroups (all p < 0.01, d _z for all participants = 1.82, Table 2). A subgroup analysis found no significant difference in the training effect between experienced and inexperienced participants (p = 0.745), or between physicians and nurses (p = 0.36). The training effect depended on the level of training and profession and is shown in Fig. 2.

We found that emergency situational knowledge in self-assessment as well as knowledge of medication for sedation significantly increased from before to after the training course, both overall and in groups by experience and profession (all p < 0.01, d _z for all participants = 1.65 respectively d _z for all participants = 1.89, Table 2).

Twenty-one participants (42 %) indicated that they had not used CRM-guidelines at all before the training course. After the training courses, participants were asked specifically about the CRM principles they used during the training and were planning to use regularly. The most frequent free text answers were “Call for help early” and “Communicate effectively; Closed loop communication”. As with the previous questions, there was a significant increase in self-assessed knowledge in all subgroups (all p < 0.01, d _z for all participants = 1.49, Table 2) without significant difference in subgroups for experience (p = 0.257) and profession (p = 0.703).

In 2015, the department of emergency medicine conducted a total of 43 sedations of which 38 are fully documented. The average time to the procedure requiring sedation was 144 min (SD = 146 min), the average time to discharge was 387 min (SD = 255 min). Of a total of 43 patients sedated in the ED in 2015, four patients encountered minor complications that could be managed by the ED team: one patient required an oropharyngeal airway, one patient reported hallucinations, one patient remembered minor pain during the procedure and one patient vomited after a sedation. We did not encounter complications requiring the intervention of an anaesthetist, advanced airway management or hospitalization of the patient. All patients reported, that they would again agree to the form of sedation chosen. All staff reported that the simulation training described above was helpful to them for the current procedure.

The most common reason for sedation in the ED in 2015 was a luxated shoulder (n = 19). In comparison to 2014, when all sedations in the ED were conducted by the department of anaesthesiology, patients requiring sedation for luxated shoulders were of equal age and had comparable comorbidities as assessed by American Association of Anaesthesiologists Scale (ASA scale) (Table 3). While we did not find significant differences between patients treated in the ED and those treated by the Department of anaesthesiology regarding the patient’s time to discharge or the time required for the procedure, the time to the procedure was significantly faster in patients sedated by the ED team (p = 0.002; see Table 3 and Fig. 3, d = 0.88).

It has been previously suggested that increased self-efficacy is not necessarily linked to better objective performance, and this has been discussed in the literature [36, 37]. In contrast to this, self-efficacy becomes a self-fulfilling prophecy by actually increasing the chances of success in a given task [38]: humans are generally better in tasks we feel proficient in. In accordance with this conception of self-efficacy, Schubert et al. found that good performance was related to high levels of self-reported feelings of preparedness [39].

The limited number of procedures performed during the first year after implementation emphasises the need of a structured training as well as the need of regular refresher trainings. It is reassuring that all complications encountered during the real-life sedations could be mastered by the trained ED staff without harm to the patient. In patients with a luxated shoulder, the time to procedure in sedations done by the ED staff was shorter than the time to procedure when a anaesthesia team was needed. This is an important step on the way to minimize the time to reduction in order to protect patients from secondary damage. It remains unclear, why the time to discharge was not significantly different. A significant reduction in time to discharge could contribute to an improved patient flow in a ED and further improve patient satisfaction.

One limitation concerns the fact that we only measured self-efficacy and self-accessed knowledge after the training day and did not test objectively acquired knowledge. We instead included real-life outcome investigation, as our training, similar to other similar programs, aims to focus on team-work to improve patient outcome in contrast to classic teaching forms like class-room teaching [12]. Another possibility of outcome assessment would have been the video-assisted observation of actual performance, using objective behavioural marker systems. Unfortunately, this was not possible in our setting, because assessment including video-recordings is extremely resource demanding and sensitive regarding privacy. We would however argue that patient outcome is the clinically more relevant measure.

Second, the sample size of patients in this study is small. This further emphasises the need for standardised procedures and structured and repetitive training, since the personal experience of staff is likely limited as a consequence.