Code Team Structure and Training in the Pediatric Resuscitation Quality International Collaborative

Abstract

Objectives:

Code team structure and training for pediatric in-hospital cardiac arrest are variable. There are no data on the optimal structure of a resuscitation team. The objective of this study is to characterize the structure and training of pediatric code teams in sites participating in the Pediatric Resuscitation Quality Collaborative.

Methods:

From May to July 2017, an anonymous voluntary survey was distributed to 18 sites in the international Pediatric Resuscitation Quality Collaborative. The survey content was developed by the study investigators and iteratively adapted by consensus. Descriptive statistics were calculated.

Results:

All sites have a designated code team and hospital-wide code team activation system. Code team composition varies greatly across sites, with teams consisting of 3 to 17 members. Preassigned roles for code team members before the event occur at 78% of sites. A step stool and backboard are used during resuscitations in 89% of surveyed sites. Cardiopulmonary resuscitation (CPR) feedback is used by 72% of the sites. Of those sites that use CPR feedback, all use an audiovisual feedback device incorporated into the defibrillator and 54% use a CPR coach. Multidisciplinary and simulation- based code team training is conducted by 67% of institutions.

Conclusions:

Code team structure, equipment, and training vary widely in a survey of international children’s hospitals. The variations in team composition, role assignments, equipment, and training described in this article will be used to facilitate future studies regarding the impact of structure and training of code teams on team performance and patient outcomes.

Each year, thousands of children worldwide suffer from cardiopulmonary arrest and require cardiopulmonary resuscitation (CPR)^1–3 with variability in the team composition, training, and response.⁴ International Consensus on Cardiopulmonary Resuscitation guidelines on CPR quality benchmarks of chest compression fraction, depth, and rate⁵ are difficult to achieve despite continuous monitoring and feedback during in-hospital cardiac arrest.^6,7 Improving outcomes related to pediatric CPR require a multidisciplinary resuscitation team (ie, code team) effort with implementation of best practices and evaluation of areas for improvement.⁸ Reports describing code team structure, training, and equipment in pediatric hospitals are limited.⁹ Because data on the current state of code team structure and training are so limited, the implementation of new guidelines and recommendations for resuscitation efforts may prove challenging. The objective of this study is to characterize the variability of structure and training of pediatric code teams outside of the intensive care unit across an international resuscitation collaborative of children’s hospitals (Fig. 1).

FIGURE 1.

Map and background information about the actively participating sites in the collaborative that responded to the survey.

METHODS

This international multicenter observational study was approved by the Cincinnati Children’s Hospital Medical Center Institutional Review Board. From May to July 2017, an anonymous voluntary Research Electronic Data Capture¹⁰ survey was distributed to the 18 sites that are actively participating in the international Pediatric Resuscitation Quality (pediRES-Q) Collaborative. The pediRES-Q Collaborative (ClinicalTrials.gov, NCT02708134) is a large multi- center international pediatric resuscitation quality improvement network. The survey content was developed by the study investigators and iteratively adapted by consensus of the authors of this article (Appendix, http://links.lww.com/PEC/A378). Collaborative site primary investigators are the local expert at their center regarding local resuscitation practices and submitted 1 completed questionnaire per institution. Descriptive statistics were calculated.

RESULTS

Code Team Composition and Code Team Member Roles

Responding institutions were asked to report the number and occupation of code team members, any preassigned roles, and the code team structure during nights and weekends (Table 1). All sites have a designated code team and hospital-wide code team activation system. Forty-four percent (8/18) of the code teams relieve the general ward team immediately upon arrival, whereas 56% (10/18) conduct the resuscitation including some of the ward providers. Only 2 of 18 sites have a different structure on nights and weekends. Overall, 89% (16/18) of sites have a designated rapid response team that is distinct from their code team.

TABLE 1.

Code Team Structure for All Sites (n = 18)

		United States (n = 13)	Non–United States (n = 5)	Total Sites (n = 18)
Team Composition	Median total members (range)	10.0 (4–17)	8.0 (3–10)	8.0 (3–17)
	Median nurses (range)	2.0 (1–5)	3.0 (1–5)	3.0 (1–5)
	Median physicians (range)	3.0 (1–7)	3.0 (2–5)	3.0 (1–7)
	Median respiratory therapists (range)	1.0 (1–2)	0.0 (0–2)	1.0 (0–2)
Preassigned Roles	No. teams with preassigned roles, n (%)	10 (77)	4 (80)	14 (78)
	Roles that are preassigned*	n = 10	n = 4	n = 14
	Leader, n (%)	9 (90)	4 (100)	13 (93)
	Airway, n (%)	10 (100)	4 (100)	14 (100)
	Medication nurse, n (%)	8 (80)	3 (75)	11 (78)
	Defibrillator, n (%)	6 (60)	1 (25)	7 (50)
	CPR coach, n (%)	6 (60)	0 (0)	6 (43)
	Access, n (%)	5 (50)	2 (50)	7 (50)
	Documentation, n (%))	5 (50)	3 (75)	8 (57)
	Parental support, n (%)	7 (70)	2 (50)	9 (64)
	Compressors, n (%)	4 (40)	2 (50)	6 (43)
Replacement of Floor Team	No. institutions that relieve members from the floor when code team arrives
	All relieved, n (%)	5 (39)	3 (60)	8 (44)
	Some relieved, n (%)	8 (61)	2 (40)	10 (56)
	Not relieved, n (%)	0 (0)	0 (0)	0 (0)
Changes to Team Structure	No. code teams with different night structure, n (%)	1 (8)	1 (20)	2 (11)
	No. code teams with different weekend structure, n (%)	1 (8)	1 (20)	2 (11)
	No. sites with other teams that respond to hospital-wide events (ie, rapid response team), n (%)	12 (93)	4 (80)	16 (89)

^*Can be multiple answers.

Code team composition varies greatly across sites, with teams consisting of 3 to 17 members (median, 8 members; interquartile range, 6–12). Overall, 78% (14/18) of hospitals have preassigned roles for code team members before the code event. Within the sites that preassign roles, the most commonly assigned roles in decreasing frequency are as follows: airway provider (100%), team leader (93%), medication administration nurse (78%), parental support (64%), and documentation (57%). Less commonly preassigned roles include defibrillator manager (50%), venous or intraosseous access provider (50%), CPR coach (43%), and chest compressors (43%). The team leader is a physician only at 85% of the centers who preidentify a team leader, with the remaining 15% having dual physician and nurse team leaders. Physicians who respond to code events include the pediatric intensive care unit (PICU) attending (61% of centers), PICU fellow (78%), emergency department (ED) attending (6%), and other attending (33%), which included anesthesia or cardiology. Sixty-one percent of centers have residents on their hospital-wide code team. In terms of nurse responders, 94% of centers have PICU nurses responding with some centers having ED (17%) and cardiology (33%) nurses responding as well. Other nonclinician members of the code team include a chaplain (39%), social worker (61%), pharmacist (56%), and at a few centers others including nurse supervisors, paramedics, radiology, child-life specialists, and security.

Code Team Equipment

Respondents were asked about code team equipment used during a code response (Table 2). A step-stool and backboard are used during resuscitations in 89% (16/18) of surveyed sites. Almost all sites (17/18) have code carts present on the pediatric general care wards in addition to the ICUs and ED containing medications, defibrillator, and equipment. Cardiopulmonary resuscitation feedback is used by 72% of the sites. Of those sites that use CPR feedback, all use a feedback device incorporated into the defibrillator. The second most common CPR feedback used is end-tidal CO2 monitoring, followed by intra-arterial blood pressure monitoring, and lastly a code member assigned as CPR coach. Ninety-four percent (17/18) of surveyed sites use either automatic external defibrillators (AEDs) or manual defibrillators during a code response. About half of the sites use both devices depending on location.

TABLE 2.

Code Team Equipment for All Sites (n = 18)

		United States (n = 13)	Non–United States (n = 5)	Total Sites (n = 18)
AEDs and Defibrillators	No. sites where code cart remains local, n (%)	13 (100)	4 (80)	17 (94)
	No. teams who use a stool and backboard for compressions, n (%)	13 (100)	3 (60)	16 (89)
	No. teams who use AEDs, defibrillators, or both outside the ICU, n (%)	13 (100)	4 (80)	17 (94)
	No. teams who use only AEDs outside the ICU, n (%)	1 (8)	0 (0)	1 (6)
	No. teams who use only manual defibrillators outside the ICU, n (%)	5 (39)	1 (20)	6 (35)
	No. teams with both AEDs and defibrillators outside the ICU based upon location, n (%)	7 (54)	3 (75)	10 (59)
CPR Feedback	No. teams who use CPR feedback, n (%)	11 (85)	2 (40)	13 (72)
	Type of CPR feedback used	n = 11	n = 2	n = 13
	Device, n (%)	11 (100)	2 (100)	13 (100)
	CPR coach, n (%)	7 (64)	0 (0)	7 (54)
	End tidal CO2, n (%)	9 (82)	2 (100)	11 (85)
	Intra-arterial blood pressure monitoring, n (%)	8 (73)	1 (50)	9 (69)

Code team response times are tracked at 67% (12/18) of centers. The median response time at tracked centers is 180 seconds (inter-quartile range, 120–216 seconds). Code teams are activated multiple ways at the responding centers including overhead (44%), pager (89%), and phone (11%). The code team responds to the inpatient setting at all centers with response to clinics at 94% (17/18), emergency department 61% (11/18), public areas 100% (18/18), and the operating room at 33% (6/18).

Code Team Training

Respondents were asked about the presence, frequency, duration, and type of code team training (Table 3). Code team training is conducted by 67% (12/18) of institutions surveyed. Overall, 62% of surveyed US institutions and 80% of non-US institutions provide code team training. All institutions describe the training as multidisciplinary and simulation-based with 83% conducting code team training in situ and 58% conducting code team training in a simulation laboratory. The median duration of code team training is 53 minutes with a range of 15 to 180 minutes. Slightly more than one third of US sites conduct training at more frequent intervals but of shorter duration than the training provided in non-US sites. United States sites report code team training either weekly (13%), biweekly (25%), or monthly (38%). At the non-US sites, half report monthly training and half report quarterly training. United States sites conduct training for a median duration of 27.5 minutes (range, 15–120 minutes), whereas the median duration of training reported by non-US sites is 76 minutes (range, 60–180 minutes).

TABLE 3.

Description of Code Team Training for Programs Who Have Scheduled Training (n = 12) or Host Debriefings (n = 18)

		United States (n = 8)	Non–United States (n = 4)	Combined Sites (n = 12)
Training Frequency	No. teams who conduct code team training, n (%)	8 (62)	4 (80)	12 (67)
	Multidisciplinary training (n)	8 (100)	4 (100)	12 (100)
	Weekly, n (%)	1 (13)	0 (0)	1 (8)
	Biweekly, n (%)	2 (25)	0 (0)	2 (17)
	Monthly, n (%)	3 (38)	2 (50)	5 (42)
	Quarterly, n (%)	2 (25)	2 (50)	4 (33)
	Median duration of code team training, in minutes (range)	27.5 (15–120)	76 (60–180)	52.5 (15–180)
Training Style	Lecture, n (%)	0 (0)	0 (0)	0 (0)
	Simulation, n (%)	8 (100)	4 (100)	12 (100)
	Simulation, in situ,* n (%)	6 (75)	4 (100)	10 (83)
	Simulation, in laboratory,* n (%)	4 (50)	3 (75)	7 (58)
Training Occurs	During patient care hours,* n (%)	8 (100)	4 (100)	12 (100)
	Non-patient care hours,* n (%)	2 (25)	0 (0)	2 (17)
Structured debriefings after hospital-wide code events (n = 18) *
Debriefings	Yes, percent hot debriefs, n (%)*	10 (83)	4 (80)	14 (82)
	Yes, percent cold debriefs, n (%)*	8 (67)	4 (80)	12 (71)
	No	1 (8)	1 (20)	2 (12)

^*Can be multiple answers.

Differences Between US and Non-US Sites

Teams from the US sites have a median of 10 members (range, 4–17) as compared with the non-US sites with a median of 8 team members (range, 3–10). In the United States, respiratory therapists are always present on the team with a median of 1 (range, 1–2); however, their presence is mixed outside of the United States with a median of 0 (range, 0–2) for non-US sites. Preassigned roles are present at 10 of the US sites and 4 of non-US sites. In the United States, 60% of sites use a CPR coach and 60% use a defibrillator provider, with 0% of the non-US sites using a CPR coach and only 25% assigning a defibrillator provider. Commonly assigned roles include the following: airway (100% for both US and non-US sites), team leader (90% US vs 100% non-US), and medication administration nurse (80% US vs 75% non-US). Compressors are assigned at 40% of US and 50% of non-US sites.

Code carts with medications and equipment are present on the ward in all US sites and 80% of non-US sites. Step stools and backboards were used during CPR by all of the surveyed US sites versus 60% of non-US sites. Some type of CPR feedback (via device, coach, end-tidal CO₂ or intra-arterial pressure monitoring) was used by 85% of US sites but only 40% of non-US sites.

DISCUSSION

This article serves as the largest and most detailed description of the landscape of pediatric code teams in terms of team composition, equipment, and training. Pediatric code team composition, role assignment, equipment, and training are highly variable across 18 sites participating in a voluntary international pediatric resuscitation quality improvement collaborative. This variability is noteworthy because there are limited studies evaluating code team structure and training and will likely have an effect on the implementation of any recommendations or guidelines. Without understanding the limitations of team structure, the challenges of implementing new elements to resuscitation cannot be fully understood. The collaborative is comprised of teams that have an active interest and are collecting data and sharing best practices to improve resuscitation outcomes. In theory, this group might be more attentive to the structure of their team and where their teams’ strengths and weaknesses may be. Although this survey certainly does not represent all pediatric resuscitation team structures, it highlights that there is a lot of variability, even among highly motivated and engaged teams.

This variability is important to examine in light of the American Heart Association consensus recommendations to optimize outcomes after in-hospital cardiac arrest. Their recommended best practices include an established code team, defibrillator, and code cart within 2 minutes of any patient, CPR feedback via device or provider, training of staff including simulation-based mock codes, and plans for routine debriefing.¹¹ In agreement with the American Heart Association recommendations, all sites have a designated code team and hospital-wide code team activation system. The roles of event leader, airway, and medication administration nurse are commonly preassigned among surveyed sites, whereas the roles of CPR coach, defibrillator, and compressors are less likely to be assigned. There are currently no recommendations or data regarding composition of an ideal code team. A point of emphasis for the collaborative is on the importance of CPR feedback via an audiovisual device. The role of CPR coach with use of CPR feed- back device is variable across institutions. Although previous studies have demonstrated a low use of audiovisual devices for CPR feedback during actual resuscitation events,⁴ real-time CPR feedback with an audiovisual device has been shown to improve chest compression fraction, depth, and rate.^6,12 In addition, CPR coaches have been shown to improve the depth of chest compressions and shorten the time to defibrillation.¹³

Simulation-based training is another recommended best practice and includes preparation before actual resuscitation attempts and debriefing after attempted resuscitations. However, only two thirds of sites report code team training, and the frequency and duration of that training is highly variable. There are insufficient published data to determine the optimal duration and frequency of code team training¹⁴; however, simulation-based training has been associated with improved team communication, organization, and performance.^9,15 In addition, some centers have seen improvement in survival rates after monthly mock codes.¹⁶ Simulation code team training experience has been associated with reduced time to initiating CPR, time to shock delivery, and improved team communication in a simulated clinical environment.¹⁵

There are several limitations to this study. First, this is a descriptive survey from a quality improvement collaborative. The nature of the survey creates bias, as the 18 organizations involved in the pedRES-Q Collaborative have demonstrated a commitment to improving CPR quality and survival. As a result, the findings may not be representative of responses from all organizations involved in pediatric resuscitation but is the largest description to date. Second, the data are self-reported with no independent verification.

Code team structure, equipment, and training varied widely in a survey of 18 hospitals in a collaborative of US and non-US children’s hospitals. In the future, the pediRES-Q Collaborative will attempt to identify characteristics of pediatric code team training and equipment that are associated with improved outcomes in in- hospital pediatric cardiac arrest.