Effects of a pilot simulation training on interdisciplinary closed-loop communication and time to critical actions at an emergency centre in Burundi

Abstract

Introduction

Emergency medicine simulation is an effective training modality in both high and low resource settings. We describe the authors’ experiences conducting a four-week interdisciplinary, in situ, simulation training series at an emergency centre in Burundi.

Methods

Training emphasized effective closed loop communication, early airway, breathing, and circulation assessment, as well as time to vital signs, IV placement, and oxygen administration when appropriate. Six doctor-nursing teams participated in four training sessions as well as pre- and post-test simulation cases which were graded by an independent evaluator.

Results

The training resulted in a statistically significant improvement in closed loop communication as well as notable narrowing in standard deviation of times to critical actions after the training intervention. Although the small sample size and large variation of data limited their statistical significance, these results may indicate a short-term benefit towards early assessment, management and team communication when simulating management of critical patients. Additionally, we found that interdisciplinary, in situ simulation was a safe and likely beneficial option for training and team building in a resource-limited emergency centre.

Conclusion

While this pilot study establishes feasibility of low-cost, interdisciplinary emergency simulation training in resource-limited settings, further research is needed to establish educational effectiveness on quality-of-care measures and its generalizability to other contexts prior to implementing similar trainings.

African relevance.

• •Study was performed in an African emergency centre with Burundian doctors and nurses.
• •There is a dire need for training from experienced trainers especially in areas across the continent with nacent EM systems.
• •Simulation training is a method being increasingly used in emergency care training throughout Africa.
• •The study addresses logistical challenges and benefits of developing and implementing a simulation training series in a resource limited emergency centre.

Alt-text: Unlabelled box

Introduction

Simulation training has been shown to be an effective and efficient means of Emergency Medicine education in both high- and low-resource settings [[1], [2], [3], [4], [5]]. It allows clinicians to apply theoretical concepts, critical thinking, and kinaesthetic skills to managing diverse patient presentations in a low-risk, moderate-fidelity environment. It also helps improve team dynamics, such as closed loop communication during stressful situations, which has been linked to improved patient care [[6], [7], [8]].

Although simulation often targets a specific learner group, interdisciplinary approaches have also proven effective [9]. Rather than focusing solely on one group such as medical students or trainees, interprofessional simulation enhances team dynamics, communication, and role clarity. It helps participants understand other team members’ perspectives during debriefs, strengthening coordination. A study by Ayensa-Arano et al. showed that simulations involving doctors and nurses improved teamwork and empowered nurses [10]. In this study, we explore the effects and authors’ perceptions of interprofessional simulation training involving emergency nurses, the emergency doctor, and non-emergency doctors covering overnight shifts.

While simulation is often conducted in a dedicated lab, in situ simulation—performed within patient care areas—has been shown to be effective and safe [[11], [12], [13]]. Labs provide structured training environments and dedicated equipment but pose challenges in limited-resource settings. When space and staffing are constrained, allocating exclusive areas and staff for simulation is often unfeasible. Removing staff from clinical duties may also be unsafe. Furthermore, simulation labs may reduce realism by excluding challenges such as limited equipment, staff task shifting, and spatial management. Given these barriers, we explore the benefits and limitations of in situ simulation at a referral emergency centre in Burundi.

Our simulation cases and debriefs emphasised rapid vital sign assessment, IV placement, oxygen administration, closed loop communication, and a standardised Airway-Breathing-Circulation-Disability-Exposure (ABCDE) approach for critical patient stabilisation. Closed loop communication, previously identified as an area needing improvement at this site, was emphasised in evaluations and debriefs. A study of paediatric resuscitations found that closed loop communication reduced task completion time threefold [14].

This study aims to (1) describe a novel interprofessional in situ simulation training for nurses and doctors at a resource-limited emergency centre in Burundi and (2) evaluate its impact on teamwork and quality of care measures including patient assessment, stabilisation, and communication using standardised simulation cases. To our knowledge, this is the first study assessing the educational feasibility of interdisciplinary in situ emergency care simulation in this setting.

Methods

Setting

The study was conducted at a 308-bed rural referral hospital in Burundi, with a seven-bed emergency centre designated for triage, stabilisation of patients with abnormal vital signs, critical internal transfers, and referral cases. The emergency centre is staffed by a dedicated medical officer during weekdays and by general surgery or internal medicine registrars overnight. Eleven nurses provide 24-hour coverage, with two to four present at any given time.

Intervention design

A four-week programme was developed with four simulation cases for doctor-nurse teams, focusing on the ABCDE approach and closed loop communication. Each nursing team completed all four cases. Sessions were held when staff were available and clinical load allowed for a ten-minute session including debrief. The design of simulation cases was guided by local author’s experience and cases were adapted to locally relevant pathology and available diagnostic and treatment resources. Cases were designed and reviewed by an emergency medicine consultant and registrar. Case format was derived from Emergency Medicine Simulation Education Research of Canada (EM-SERC).

Cases were designed to focus on the first ten minutes of managing critically ill patients. While the ABCDE survey ideally takes under five minutes, scenarios lasted up to ten minutes to allow for completion of all critical actions. Both paediatric and adult cases were included, covering common and rare presentations (e.g., traumatic pneumothorax, cerebral malaria, typhoid bowel perforation, asthma exacerbation, myxoedema coma, rheumatic heart disease). Debriefs included ABCDE application review, participant feedback, and targeted coaching. Skills such as airway positioning, suctioning, BVM ventilation, chest compressions, needle thoracostomy, GCS/AVPU assessment, and trauma examinations were reinforced. Clinical considerations were discussed and questions addressed. While formal rubrics were used for pre- and post-tests, training sessions followed a more flexible competency framework. Fig. 1 outlines the simulation flow.

Fig. 1.

Flow diagram of simulation session.

A cross-over design was used for pre- and post-test cases (e.g., Teams A, B, and C used Case 1 for pre-test and Case 6 for post-test, and vice versa for Team D, E, and F), helping control for case difficulty. Doctor-Doctornurse teams remained consistent to control for experience (Supplementary Appendix). Nurses had no prior training on the evaluated measures.

At the study’s conclusion, the local emergency provider was trained to independently design and lead simulations using on-site resources including a simulation mannequin, iPad, and a simulation case database, supporting sustainability.

Participants

Participants included 11 nurses and six doctors (one full-time emergency doctor and five general surgery registrars covering overnight). Teams consisted of one doctor and two nurses (except Team F, which had one nurse). All nurses participated in pre/post-tests and training sessions, except for one on leave. While all doctors completed pre/post-tests, only the full-time emergency doctordoctor completed all training cases due to scheduling constraints. This reflected real-life team dynamics where rotating coverage prevents full team consistency.

Doctor-nurse pairs were kept consistent for evaluations, though nurse pairing varied during training due to shift overlaps (Fig. 2).

Fig. 2.

Participant study flow diagram.

Data collection

A single evaluator observed all pre/post-test sessions and timestamped critical actions. A three-point scale assessed task performance: 0 (not done), 1 (done), 2 (done well and communicated). Closed loop communication was defined by the nurse repeating and acknowledging verbal orders. A rapid hatching method was used to quantify this: a “/” for an order given, and a “\” if acknowledged, forming an “X.” This allowed calculation of the percentage of completed closed loop communications. Orders were not predetermined, ensuring unbiased assessment.

Materials

Training utilised a donated Laerdal Resusci Anne QCPR mannequin capable of pulse palpation, chest compressions, and ventilation chest rise. An iPad was used to display vital signs, X-rays, ultrasound images, and ECGs. All other available emergency centre equipment was used, without opening medications.

Data analysis

All data analysis was performed using STATA software. We used paired t-tests to compare the proportion of closed loop communication, time to vital signs, time to IV placement, time to oxygen administration, and time to case completion before and after the training intervention. Wilcoxon signed-rank tests were used to compare airway, breathing, and circulation assessments, as well as role assignment, before and after the intervention.

This study was approved by the Frank Ogden School of Medicine at Hope Africa University Scientific Research Ethics Committee, Bujumbura, Burundi (Study REF. No. 2022/022), and the University of Southern California Institutional Review Board approved the study via ceded review to the above committee (#UP-23-00741).

Results

Analysis of closed loop communication revealed a statistically significant mean increase in the percentage of closed loop communication, from a mean of 9.45% (SD = 10.42) before the intervention to 52.33% (SD = 33.90) after the intervention. This yielded a gross improvement of 42.88%, 95% CI [-74.72, -11.04], t(5) = -3.46, p = .018.

We observed a decrease in time to first vital signs, from a mean pre-test time of 135.5 s (SD = 119.3) to a mean post-test time of 74.5 s (SD = 21.48). The mean difference was 61 s, 95% CI [-54.70, 176.70], t(5) = 1.36, p = .23. There was a decrease in time to IV placement, with a mean pre-test time of 248.17 s (SD = 161.72) and a mean post-test time of 134.17 s (SD = 43.1). The mean difference was 114 s, 95% CI [-81.15, 309.15], t(5) = 1.5, p = .19. Time to oxygen administration also decreased, from a mean pre-test time of 275 s (SD = 185.14) to 192.67 s (SD = 145.80). Two-sample t-test analysis showed a mean difference of 82 s, 95% CI [-295.41, 460.08], t(4) = 0.61, p = .58.

Time to case completion—defined as the time until the last doctor’s order was completed—also improved, though not statistically significantly. The mean pre-test time to completion was 522.67 s (SD = 305.37), compared to 462.67 s (SD = 20.58) post-test. Two-sample t-test analysis showed a mean difference of 60 s, 95% CI [-257.77, 377.77], t(5) = 0.49, p = .65.

Regarding performance in role assignment and airway, breathing, and circulation assessment, there was high variability. Airway and breathing assessments improved in two teams, worsened in one, and remained unchanged in three; p = 1.00 for both. Circulation assessment worsened in two out of six teams and did not improve in any; p = .50. However, role assignment improved in four teams, worsened in one, and remained unchanged in one; p = .38 (Fig. 3, Fig. 4 and Table 1).

Fig. 3.

Changes in time to vital signs, IV placement, oxygen administration and time to case completion.

Fig. 4.

Percentage of closed loop communications before and after simulation training intervention.

Table 1.

Each team’s change in performance of role assignment, breathing assessment, circulation assessment, and airway assessment during the post-test case compared to the pre-test case.

Role Assignment
Team	Post-Test Change in Performance
A	Improved
B	Improved
C	Improved
D	Improved
E	Worsened
F	No Change
Breathing Assessment
Team	Post-Test Change in Performance
A	Worsened
B	No Change
C	No Change
D	Improved
E	No Change
F	Improved
Circulation Assessment
Team	Post-Test Change in Performance
A	No Change
B	Worsened
C	No Change
D	No Change
E	Worsened
F	No Change
Airway Assessment
Team	Post-Test Change in Performance
A	Worsened
B	No Change
C	No Change
D	Improved
E	No Change
F	Improved

Discussion

This section discusses our quantitative findings, which are limited by sample size and study design (as noted in the limitations). We also reflect on the authors’ practical experiences in organising the simulation sessions.

This small pilot study of an interdisciplinary simulation training demonstrated a statistically significant improvement in closed loop communication and a non-statistically significant short-term improvement in time to critical action completion. A notable reduction in standard deviations across most measures post-intervention suggests a more organised and consistent approach to managing critically ill patients.

The improvement was likely driven by the participation of nursing staff in the training sessions, as only one doctor attended. This highlights the essential role of nurses in interdisciplinary training. Despite 5 out of 6 doctors not participating in the training, improvements in communication and trends towards better task performance were observed. In many low-resource settings, nurses are often the first to assess and stabilise patients. Practising rapid assessment, IV access, and oxygen delivery alongside the ABCDE approach likely enhances critical care. While the interdisciplinary nature of training reflects real-world dynamics and was well received, limited doctor participation limits our ability to determine the added value of interdisciplinary versus nurse-focused training.

We maintained consistent doctor-nurse teams to control for experience levels and used a cross-over evaluation method to balance test case complexity. However, due to scheduling limitations, nurse team compositions varied during training sessions. This mirrors real-life scheduling challenges and may support scalability of the training model.

Conducting simulation within the emergency centre enhanced realism. Nurses had to locate and prepare equipment in their usual setting, uncovering real barriers to effective resuscitation. For example, beds needed to be repositioned for airway access, calculation and dosing errors occurred during paediatric simulations, and equipment such as suction cables or intraosseous kits were either missing or misplaced. These issues were addressed in real time thanks to the in-situ approach.

From an organisational standpoint, in situ simulation offered several advantages. It was low cost—the only specific requirements were a mannequin and a tablet for displaying vitals and test results. All other materials were sourced from existing emergency department stock (without opening supplies). Moreover, it required no special scheduling; cases were conducted when beds were free and staff were available. As Martin et al. noted, effective in situ simulation should be tailored to the resources and needs of each centre [15].

A local champion is crucial for success. Two authors (PB and CW), both local emergency doctors, helped tailor case content to local pathology and encouraged nurse participation. They also led debriefs with appropriate clinical context. By the end of the training, both nurses and the weekday emergency doctor expressed interest in continuing simulation education. Subsequently, the weekday doctor received training in case design, facilitation, and debriefing, and led several sessions independently, supporting sustainability.

Closed loop communication is often evaluated via video analysis [[16], [17], [18]]. However, in low-resource settings without video equipment, we propose a simple hatching method to rapidly assess communication, as described in the methods section. Though future validation is needed, it enables quick feedback in post-case debriefs. Initially, we aimed to track a larger number of actions (per Appendix A), but practical constraints led us to narrow our focus to key metrics.

By the end of the intervention, the weekday emergency doctor was more confident during both simulated and real resuscitations, frequently requesting closed loop communication. Nurses were also observed repeating instructions during clinical care. Although these behaviours may be subject to the observer effect and attrition is possible, we are optimistic that improved team dynamics will persist.

Limitations include small sample size, group variability, and limited doctor involvement, restricting the generalisability of our findings. Subjective assessments (e.g. ABC evaluations, role assignment) may carry evaluator bias. Training and test cases were not externally validated, and the oxygen metric was only applicable to one case, yielding six datapoints. We did not measure participant confidence or long-term skill retention. Future use of a structured framework like the Kirkpatrick model may improve evaluation of training outcomes across multiple dimensions.

As noted, limited doctor participation weakens our ability to comment on the value of interdisciplinary versus nurse-specific training. One team had only one nurse, potentially affecting results. These issues reflect real-world challenges of inclusive training. The evaluation tool was not externally validated, and only one evaluator—a senior emergency medicine consultant—was available. Finally, this was a small, single-centre study in Burundi and may not be generalisable to other low-resource settings.

In conclusion, this small, single-centre pilot study demonstrated the feasibility of a low-cost, in situ, interdisciplinary simulation training programme with improved closed loop communication at a Burundian emergency centre. While the author’s had a favourable experience using interdisciplinary training sessions, comparing these to simulation targeting a specific cadre of learner in future studies would be valuable. We recommend further evaluation of the benefits and challenges of in situ simulation compared to use of a dedicated simulation lab as well as evaluation of long-term knowledge and skill retention from similar simulation training.

Dissemination of results

Results from this study was shared with the local emergency centre doctor who will distribute it in an anonymous manner to participating staff through an informal presentation at an upcoming departmental meeting.

Declaration of generative AI and AI-assisted technologies in the writing process

During the preparation of this work the author(s) used GPT-4o in order to standardize terminology and assist with optimizing word count. After using this tool/service, the author(s) reviewed and edited the content as needed and take(s) full responsibility for the content of the publication.

CRediT authorship contribution statement

Eric Arnold: Conceptualization, Methodology, Formal analysis, Data curation, Writing – original draft, Writing – review & editing, Project administration. Patricial Bulakali: Writing – review & editing. Elizabeth Burner: Formal analysis, Software, Visualization. Taylor Burkholder: Methodology, Writing – review & editing. Carlan Wendler: Conceptualization, Methodology, Investigation, Resources, Writing – review & editing, Supervision.

Declaration of competing interest

The authors declared no conflicts of interest.