Enhancing paediatric resuscitation team performance: targeted simulation-based team leader training

Abstract

Effective team leadership is linked to improved resuscitation outcomes. Previous studies have focused primarily on trainee performance and simulation-based outcomes. We hypothesised that a targeted simulation-based educational intervention for experienced physicians focusing on specific process and communication goals would result in improved performance during actual resuscitations. We conducted an observational pilot study evaluating specific process metrics during clinical resuscitations before and after a 1-hour training intervention for paediatric emergency medicine (PEM) supervising physicians using rapid cycle deliberate practice simulation-based training. Videos of clinical resuscitations from before and after the intervention were retrospectively reviewed to assess time to patient transfer to emergency department stretcher, time to primary assessment and time to team leader summary statement. Between March and July 2018, 21/38 of PEM supervising physicians participated in a training session. After the intervention period, clinical resuscitation teams showed significant improvements in targeted process metrics: transfer of patient within 1 min (79% vs 100%, p=0.03), assessment completed within 3 min (28% vs 75%, p=0.01) and summary statement within 5 min (50% to 85%, p=0.03). Brief, focused simulation-based team leader training can improve the teamwork and communication performance of experienced clinicians during clinical resuscitations.

Effective team leadership has been linked to increased guideline adherence and better patient outcomes during trauma resuscitations.¹ American Heart Association resuscitation courses introduce team leadership skills, discuss crisis resource management, task delegation, effective communication and efficient team dynamics.^2–8 Courses such as Advanced Cardiovascular Life Support, Advanced Trauma Life Support, Paediatric Advanced Life Support (PALS) and Team Strategies and Tools to Enhance Performance and Patient Safety focus on the introduction of teamwork and communication skills for less experienced providers, rather than experienced provider skill refinement.^{2 9}

Studies examining the impact of teamwork and communication training have focused on less experienced providers, such as residents and medical students.^{3 10–13} Published team leadership tools have assessed trainee physician performance in teamwork/communication across a variety of clinical settings.^{10 11} Hunziker et al examined medical students receiving either teamwork/communication or technical skill instruction. Four months postintervention, the teamwork instruction group performed better in time to start cardiopulmonary resuscitation (CPR), hands-on CPR time and leadership utterances during simulated resuscitations.¹² Gregg et al showed long-term resident performance improvement during trauma resuscitations after immediate, individualised feedback.¹⁰ Trainee-led teams with more teamwork/communication errors had increased patient treatment errors during trauma resuscitations.¹⁴ Rapid-cycle deliberate practice (RCDP) simulation has been linked to a greater impact on resuscitation team performance than traditional simulation when reassessed in a simulated setting.¹⁵ To date, no studies have focused on the impact of teamwork/communication training interventions for experienced physicians on actual resuscitations.

An observational pilot study was conducted to determine if an educational intervention for experienced paediatric emergency medicine (PEM) physicians, which included establishing process and communication goals and practice using RCDP simulation, improved key process metrics in clinical resuscitations.

Methods

Setting

The study was conducted in the academic paediatric emergency department (PED) of a tertiary care, free-standing children’s hospital with approximately 51 000 annual visits. The median resuscitation patient age was 6.1 years (IQR 2.2–14.8 years). The hospital is not a trauma centre, and, therefore, the majority of resuscitations are medically focused (eg, respiratory failure, arrhythmias, cardiac arrest and seizure). Resuscitations are conducted in the primary resuscitation room, if available. On average, 20 resuscitations occur in the room per month. Paediatric resuscitations are managed by the PEM emergency department (ED) team, led by a supervising physician (either an attending who has completed a 3-year subspecialty fellowship or fellow who is in the process of completing the training), with team members consisting of PEM nurses, respiratory therapists and residents (paediatric or emergency medicine), all based in the PED.

Preintervention quality review

During an ongoing quality improvement (QI) programme, all clinical resuscitations in the primary resuscitation room were recorded and stored on a centralised server between December 2017 and February 2018. One member of our QI team, a PED clinical nurse specialist and author (EB), retrospectively reviewed all recordings and extracted process and communication performance data for ongoing quality assurance unrelated to our study. No identifying patient or team information was collected. Time zero was defined as when the patient entered the room. Times to evaluations and interventions were measured from time zero during video review using a stopwatch and documented in resuscitation performance notes. Resuscitations performed outside of the resuscitation room, those with inadequate video or audio quality, and videos that began with patient care in progress were not included for analysis. Extracted video data were reviewed retrospectively for our study. Initial reviews identified care delays and teamwork and communication gaps. An intervention targeting team leaders was designed based on these identified needs.

Process metrics

Analysis of data extracted from preintervention resuscitations identified the following areas for improvement: time to start evaluation/care, time to primary evaluation of patient and time to summary statement. Based on author consensus, the following goal metrics were established: transfer to ED stretcher within 1 min, verbalisation complete primary assessment within 3 min and delivery of summary statement by team leader within 5 min.

Intervention

In March 2018, we implemented a voluntary, standardised, 1-hour simulation-based team leader training session, immediately preceding a required interprofessional simulation. Intervention participants were supervising PEM physicians only. Four sessions were conducted over a 5-month period, each with a different participant group.

Each session included three to six supervising physicians and two facilitators. We reviewed team leadership best practice characteristics and shared preintervention clinical data. Best practices incorporated into the session included:

1. Role identification and assignment.

2. Verbally prompting patient transfer to ED bed.

3. Verbally prompting initiation of the primary survey.

4. Summary statement with working diagnosis and priorities.

5. Maintenance of a calm environment.

6. Closed-loop communication.

Two facilitators, experts in PEM, teamwork and simulation, then employed RCDP with participants rotating through team roles during a low fidelity paediatric seizure scenario. Consistent with RCDP, the scenario proceeded until a performance goal was missed. The resuscitation was stopped by the facilitator; the gap and goal were discussed; and the scenario was restarted from the beginning. The simulated resuscitation ended when all goals were achieved. Immediately following the intervention, participants continued skill practice in 2-hour-long interprofessional simulated resuscitations.

Postintervention quality review

Postintervention process metric data were collected from resuscitation video performance notes extracted retrospectively by EB for QI purposes of resuscitations occurring from August to December 2018. The same inclusion and exclusion criteria and methodologies for calculating time metrics were applied.

Statistical analysis

Process metrics were categorised as completed or not completed within the target time period and were analysed using χ² tests.

Results

Preintervention recordings of 14 clinical resuscitations were analysed. During the intervention period, March–July 2018, 21/38 (55%) of PEM supervising physicians attended the team leader training sessions. Postintervention, 20 clinical resuscitation videos were analysed. Clinical process metric data are presented in table 1 with an increase in completion for all observed.

Table 1.

Clinical process metric data from clinical resuscitations in a single PED

Performance metric	Preimplementation, n (%) (n=14)	Postimplementation, n (%) (n=20)	P value
Transfer to ED stretcher within 1 min of arrival	11 (79)	20 (100)	0.03
Primary assessment completed within 3 min	4 (28)	15 (75)	0.01
Summary statement within 5 min	7 (50)	17 (85)	0.03

ED, emergency department; PED, paediatric emergency department.

Discussion

This pilot study was designed to evaluate the impact of targeted team leader simulation-based training on resuscitation team performance. Our brief intervention resulted in key clinical process metric improvements in actual patient resuscitations, when assessed across all resuscitation teams. Only approximately half of individual team leaders participated in the intervention. Team performance was not analysed for individual team leader participation. During postimplementation, standardised positioning of personnel within the resuscitation room based on roles was emphasised, but no other additional interventions or structural changes were implemented to improve team performance. Notably, team improvement occurred despite the inclusion of a new academic year in the postimplementation period, when less experienced physician trainees joined resuscitation teams as the primary patient assessment provider.

Targeted team leader training, emphasising crisis resource management, has been previously linked to improved communication and guideline adherence for trainees in a simulated setting.³ Trainee leadership coaching has shown improved CPR and team leader performance during simulated resuscitations over technical skill instruction.¹² This study demonstrates translation of improved resuscitation team performance for specific metrics following team leader training for experienced providers, beyond the simulated setting and into the clinical environment.

All PEM physicians in this study were PALS certified prior to the study. Clinical resuscitation team performance improvement suggests there is benefit from targeted advanced team leader training for supervising physicians beyond the foundational leadership skills presented during certification courses, taught during graduate medical education or acquired through clinical practice. Additionally, team performance may increase without all team leaders attending training sessions, with key concepts potentially being spread through other channels.

Limitations

This study was conducted in a single academic PED. Although individual team leaders received training, clinical resuscitation metrics were tracked in aggregate. The proportion of clinical resuscitations led by team leaders who had attended training cannot be determined because identifying information was not recorded. A major limitation is that this study used data collected for QI purposes by one video reviewer, and intrarater reliability was not measured. The number of videos excluded for inadequate audio or visual quality, conducted in rooms without video capability, or that began after the patient arrival was not captured.

Analysis focused on process metrics, not clinical outcomes due to the infrequent nature of resuscitations and wide variety of diagnoses encountered. Process metrics were selected based on identification of recurrent team performance gaps during preimplementation quality review as key steps necessary during early resuscitation phases to advance care. While team performance improved postimplementation, the observational nature of this study makes it difficult to attribute observed improvement directly to the intervention. Although there were no other concurrent resuscitation team changes beyond an increased emphasis on physical placement within the room based on role, team performance improvement may have been due to team member awareness of team leader goals. This study did not determine which goals or process metrics should be generalised across institutions. Future multicentre studies may allow for further evaluation of the impact of advanced team leader training and team performance on clinical outcomes.

Conclusions

Low-fidelity, simulation-based targeted team leader training is an additional approach that can enhance resuscitation team performance, improving clinical process metrics. Experienced physicians can benefit from focused process and communication training, beyond content presented in standard resuscitation courses.