High-quality cardiopulmonary resuscitation (CPR) increases the likelihood of survival of pediatric out-of-hospital cardiac arrest (OHCA). Maintenance of high-quality CPR during transition of care between prehospital and pediatric emergency department (PED) providers is challenging. Our objective for this initiative was to minimize pauses in compressions, in alignment with American Heart Association recommendations, for patients with OHCA during the handoffs from prehospital to PED providers. We aimed to decrease interruptions in compressions during the first 2 minutes of PED care from 17 seconds (baseline data) to 10 seconds over 12 months. Our secondary aims were to decrease the length of the longest pause in compressions to <10 seconds and eliminate encounters in which time to defibrillator pad placement was >120 seconds.
Our multidisciplinary team outlined our theory for improvement and designed interventions aimed at key drivers. Interventions included specific roles and responsibilities, CPR handoff choreography, and empowerment of frontline providers. Data were abstracted from video recordings of patients with OHCA receiving manual CPR on arrival.
We analyzed 33 encounters between March 2018 and July 2019. We decreased total interruptions from 17 to 12 seconds during the first 2 minutes and decreased the time of the longest single pause from 14 to 7 seconds. We saw a decrease in variability of time to defibrillator pad placement.
Implementation of a quality improvement initiative involving CPR transition choreography resulted in decreased interruptions in compressions and decreased variability of time to defibrillator pad placement.
Survival in pediatric out-of-hospital cardiac arrest (OHCA) is low, with published estimates of 10.2%.1 Multiple studies reveal that high-quality cardiopulmonary resuscitation (CPR) in OHCA increases the likelihood of return of spontaneous circulation and favorable neurologic outcomes.2–5 The 2015 American Heart Association (AHA) defines high-quality CPR by 5 components: (1) adequate frequency, (2) adequate depth, (3) allowing full chest recoil, (4) minimizing interruptions in compressions, and (5) avoidance of excessive ventilation.6 Animal studies demonstrate that interruptions in chest compression decrease coronary and cerebral blood flow.7 The AHA recommends compression pauses to be <10 seconds.6 Despite the evidence linking high-quality CPR to better outcomes, and ubiquitous recommendations to prioritize CPR quality, studies of pediatric OHCA consistently report that CPR fails to meet the AHA’s standards.8
Transitions in care, or handoffs, have been shown to be associated with increased risk for medical errors.9–11 CPR handoffs from prehospital to hospital providers may be especially associated with high risk for decreased quality of CPR because of shifting team members and competing clinical priorities. Interruptions in chest compressions are a common cause of poor-quality CPR and are associated with worse outcomes.7,12–15 Early defibrillation is a key component of survival for patients with ventricular fibrillation or ventricular tachycardia.16 Delays in pad placement during CPR handoffs may contribute to delays in defibrillation. Therefore, when a patient arrives to the pediatric emergency department (PED) in OHCA, it is imperative for the PED team to (1) take over CPR with minimal interruptions and (2) apply defibrillator pads quickly. To our knowledge, a detailed study of CPR handoffs from prehospital to PED providers during OHCA has not been published. Moreover, there are no studies of efforts to improve CPR handoffs.
In 2018, an interprofessional committee observed that transfer from prehospital to PED teams for patients with OHCA frequently involved inadequate compressions, prolonged pauses, and challenges with defibrillator pad application. The committee conducted a preliminary review of CPR quality during handoffs to establish baseline performance. For 12 consecutive encounters (March 2018 to August 2018) with OHCA and CPR in progress, the PED care team averaged 17 seconds of interruptions in chest compressions during the first 2-minute cycle of CPR. Eight encounters (66%) had at least 1 pause >10 seconds. The PED care team applied defibrillator pads within the first 2-minute CPR cycle for 9 encounters (75%).
Our objective for this initiative was to improve the quality of CPR for patients with OHCA during handoffs. We theorized that standardization of roles during CPR handoffs would lead to improved CPR quality, especially reducing interruptions in chest compressions. We also postulated that engaging frontline, nonphysician providers to develop a standardized approach to CPR handoffs would be associated with improved CPR quality. We specifically aimed to reduce the total duration of interruptions in chest compressions during the first 2-minute cycle of PED CPR from 17 to 10 seconds within 12 months. We set our goal time of 10 seconds on the basis of AHA guidelines that pauses in chest compressions should be <10 seconds and simulated scenarios in which the team was able to transition care with pauses in chest compressions between 7 and 10 seconds. Our secondary aims were to decrease the length of the longest pause in chest compressions during the first 2-minute cycle of PED CPR to <10 seconds and eliminate encounters in which time to defibrillator pad placement is >120 seconds.
This quality improvement (QI) initiative was conducted in the resuscitation area of a high-volume academic PED with an annual volume of 62 000 visits. The resuscitation area has ∼4500 patient encounters per year, or 7% of the total patient volume. The parent pediatric institution is a level I trauma center, with ∼85% of the pediatric admissions from a population base of 2 000 000 people. Before initiation, our institutional review board determined the project to be non–human subjects research.
In our system, emergency medical services (EMS) calls ahead for nearly all patients in OHCA, and a care team convenes in a resuscitation bay in advance. Care teams are interprofessional and include at least 10 members: 2 physicians, 4 nurses, 1 respiratory therapist, 2 patient care assistants (PCAs), and 1 paramedic. Individuals rotate on and off the resuscitation team on assignments, with team members changing multiple times each day. The PCAs are primarily responsible for chest compressions, and the paramedic is primarily responsible for defibrillator pad placement and managing the defibrillator.
Each resuscitation bay is equipped with ceiling-mounted digital video cameras and microphones that record video and audio continuously. Video recordings of patient encounters are available for review by using a proprietary software program (Live Capture; B-Line Medical, Washington, DC). The program displays time in hours, minutes, and seconds and allows for pausing, rewinding, and annotation. Patients and families provide consent for recordings used for quality assurance and peer review activities in our general consent-to-treat form.
This QI initiative was conducted by the Medical Resuscitation Committee (MRC) in collaboration with our institution’s Center for Simulation and Research. The MRC is an interprofessional committee within the Division of Emergency Medicine that is tasked with oversight of the care provided to patients who are critically ill and nontrauma patients in the resuscitation area. Formed in 2007, the history and primary activities of the MRC have been described previously.17
In July 2018, the MRC initiated a project to improve CPR quality during handoffs. The improvement team initially consisted of 4 MRC members, 2 pediatric emergency nurses, and 2 pediatric emergency medicine faculty physicians.
Theory and Key Drivers
On the basis of a review of the preliminary data, the improvement team theorized that clear roles and a standardized approach to CPR transition; a standardized approach to defibrillator pad placement; reduced cognitive load during high-stress, low-frequency events; and empowerment of frontline providers contribute to optimal handoffs. The improvement team then constructed a key driver diagram to make explicit the theory for improvement (Fig 1).
Design of the Intervention
Two members of the improvement team conducted informal interviews with several PED paramedics and PCAs. Paramedics commented that there were often conflicting priorities, which created barriers to pad placement and CPR handoff. EMS frequently gives report and removes equipment, whereas the PED team is tasked with taking over CPR, ventilation, applying monitors, receiving report, and performing an assessment. Although all of these activities are important, they cannot happen simultaneously. When important tasks are performed out of order or when one task interrupts the performance of another, the team’s stress may increase, and the quality of CPR may decrease.
Using the principle of “deference to expertise,”18 we recognized early that the PCAs and paramedics had far greater insight into the CPR handoff process and would be effective in designing an intervention. Thus, one PCA and one paramedic were added to the core improvement team. In concert with one of the emergency nurses and a faculty physician, the paramedic and PCA co-led the design and dissemination of the intervention.
On the basis of the interviews, key insights from the PCA and paramedic, and the preliminary data, the improvement team developed as the main intervention a detailed description of the responsibilities of PCAs and paramedics during CPR handoffs (Supplemental Information). The improvement team detailed the choreography and the language to cue specific actions. The team drafted an initial version of the CPR handoff choreography, and then, after the intervention was refined through group discussion, the team conducted initial tests of change through improvement team–only simulations in the resuscitation area.
The improvement team disseminated the intervention in several ways. First, the improvement team and the MRC are all care team members, enabling us both to model the CPR choreography and educate other care team members. Second, the improvement team focused educational efforts on PCAs and paramedics using the infrastructure of their monthly educational program. The improvement team first e-mailed PCAs and paramedics a PowerPoint summary of the project, the specific CPR choreography, and a video of performance of the CPR choreography during a simulation in the emergency department resuscitation area (https://vimeo.com/297368784). After reviewing all materials, each PCA and paramedic participated in a brief simulation of the CPR choreography, led by a nurse project champion. Forty-six of 48 (96%) PCAs and paramedics completed the education.
Third, the improvement team presented educational material to nurses, respiratory therapists, and pediatric emergency medicine physicians at respective meetings. The team consistently emphasized the importance of letting the PCAs and paramedics complete the CPR handoff process.
The final step was a series of in situ simulations implemented to both test and further disseminate the novel process. The MRC and our institution’s Center for Simulation and Research run an in situ simulation program originally developed in 2007.19 We completed 3 in situ simulations to test the CPR choreography before application to actual patient events. Using the regular page-based PED notification system, we gave the care team prearrival notification of a patient in cardiac arrest. Care team members were given time to prepare for patient arrival. Simulation center staff, dressed as EMS providers, arrived performing CPR on a mannequin on a transport stretcher. PED staff were expected to take over care of the simulated patient, including the novel CPR transition process. Improvement team members were present at each simulation to assess performance and debrief, focusing on CPR handoff.
Study of the Intervention
After completion of the educational efforts and simulations, assessment of the CPR handoff began on actual patient events. All patients with OHCA were eligible, with the exclusion of cases with a mechanical compression device. Eligible patients were identified by the PED nurse who codirects the MRC using the standard process. The MRC nurse notified the improvement team coleaders (E.F.H., M.K.C.T., and A.V.), who completed a structured video review for each patient. During collection of the preliminary data on CPR handoffs, members jointly reviewed several patient videos, developing consensus definitions for key data elements. Previous studies of care in the resuscitation area have revealed that video review is a highly reliable method of data collection20,21 ; therefore, we did not conduct additional reliability testing.
The primary process measure was the total duration of interruptions in chest compression per patient during the first 2 minutes of CPR. The beginning of the 2-minute period started when the EMS stretcher was parallel to the PED bed and ended 120 seconds later. Interruption in CPR was determined by video-based observation of chest compressions; pauses for ventilation were excluded. If compressions were not resumed immediately after 2 breaths delivered, the time from the end of the last ventilation to the next compression was included. If at least one ventilation did not occur during a pause, that time was included as an interruption. When an endotracheal tube was in place, all pauses were included as an interruption because AHA guidelines recommend continuous compressions when a definitive airway device is in place. We also assessed CPR pauses in this fashion when a supraglottic device was in place because, locally, we consider these devices as advanced airway devices for CPR purposes. We assessed the longest single pause in chest compressions for each patient encounter during the first 2 minutes of CPR.
Our secondary process measure was the duration from patient arrival to defibrillator pad placement. Defibrillator pad placement was defined by video-based observation of the time when both the front and back pads were in place and the cable was connected to the defibrillator. We intended to evaluate this process measure both in terms of total time to pad placement during individual patient encounters as well as adherence to defibrillator pad placement within 120 seconds in all patient encounters.
Statistical process control was used to determine if the CPR handoff intervention was associated with significant changes in processes. Annotated control charts were developed and updated. To measure the number of seconds off the chest during the first 2 minutes of CPR, the longest single pause in chest compressions, and the time to defibrillator pad placement, we constructed X charts. A moving range (MR) chart was developed to assess variation. To measure our adherence to defibrillator pad placement within 120 seconds of patient arrival, we constructed a G chart, in which each data point indicates the number of cases between failures. We used standard rules for interpretation of a Shewhart chart to determine special cause variation.22
Starting in March 2018 and ending in July 2019, 33 patients were managed in our resuscitation bays for OHCA. Our primary process measure, number of seconds off the chest during CPR, decreased from 17 to 12 seconds. The CPR choreography intervention decreased our average time off the chest, meeting special cause variation (Fig 2). After the improvement team discussed this change, namely our theory for change, the centerline was adjusted by using standard control-chart rules. This change has been sustained through 19 patients.
Our second process measure, longest single pause per patent encounter, decreased from 14 to 7 seconds (Fig 3). The CPR choreography intervention decreased our average time off the chest during the longest pause, meeting special cause variation. After the improvement team discussed the theory for change, the average centerline was decreased by using standard control-chart rules.
Last, our third process measure, time to defibrillator pad placement, remained unchanged at 92 seconds (Fig 4). The MR chart revealed a reduction in the variation in time to defibrillator pad placement after interventions (Fig 4). Although our average time to defibrillator pad placement is <120 seconds, several patient encounters before the implementation of our improvement initiative exceeded this goal. The G chart (Fig 5) revealed an increase in the number of patient encounters between events in which defibrillator pad placement took >120 seconds, with the most recent data point (July 5, 2019) revealing special cause variation. However, given the limited number of data points in this graph, we cannot be certain that this reflects a true change in process.
Pediatric OHCA is a low-frequency, high-acuity event that presents unique challenges to teams attempting to perform life-saving skills. For CPR handoffs, we theorized that engaging frontline providers to develop a standardized choreography of chest compression and pad placement during transition would be associated with improved CPR quality. We found that after testing an intervention designed to standardize CPR choreography, the quality of CPR handoffs improved, as measured by decreased time off the chest, decreased variation in the time to defibrillator pad placement, and more patient encounters between defects in pad placement within 120 seconds.
Similar QI projects have revealed that standardization can improve care in the most stressful and acute environments.23 Wurster et al23 used QI methodology to improve compliance with advanced trauma life support recommendations for the primary assessment of trauma patients. Hunt et al24 instituted a resuscitation quality bundle and demonstrated improved adherence to AHA CPR guidelines for pediatric in-hospital cardiac arrest.
We recognize that decreasing compression interruptions by 5 seconds is of unclear clinical significance. However, one case highlights the potential importance. A patient presented in cardiac arrest during our study period. The handoff choreography was used, with a total of 14 seconds off the chest during the first 2 minutes and pad placement within 86 seconds. At first rhythm check the patient was in ventricular fibrillation, had defibrillation performed, and return of spontaneous circulation was achieved. This patient had a full neurologic recovery. This patient experience, although not statically significant, does highlight the potential impact of this work.
Our goal was to reduce interruptions in CPR during the first 2 minutes to <10 seconds. Although we have reduced the total interruption time, we have not achieved our goal of 10 seconds. We feel this is due to the multiple points in transition in which interruptions occur (ie, CPR from EMS to PED provider, patient from EMS stretcher to PED bed, and pauses for front and back pad placement). It is possible to have brief pauses for each point in the transition and still have a total interruption time >10 seconds. We recognize that the most potentially clinically significant component is the length of any single pause; thus, we are encouraged by the reduction of the duration of the single longest pause achieved through this process. We are planning to conduct individual CPR provider feedback when we identify prolonged interruptions in CPR >10 seconds.
Additionally, since the implementation of our improvement efforts, we have seen a reduction in the occurrence of encounters during which defibrillator pad placement took >120 seconds, reflected in our most recent data point in Fig 4 spanning 6 months. The few data points in Fig 4 highlights the rarity of CPR in the PED coupled with our now rare prolonged time to defibrillator pad placement. We intend to continue to follow this outcome measure and evaluate individual cases in which defibrillator pad placement took >120 seconds to identify contributing factors.
This QI project involved standardization of a high-acuity, low-frequency event and was led by frontline providers thoroughly involved in the project. The process has had sustained improvement for 9 months. We used key drivers to inform our theory for improvement. Ultimately, we postulate that frontline provider buy-in and acceptance of the designed intervention provided a foundation that sustained improvement.
Our work has several limitations. It was performed at a single large pediatric center with specialized resources that may limit generalizability. Our institutional culture and resources supported successful implementation but may pose challenges in other settings. Low frequency of pediatric OHCA and poor outcomes, in addition to the multiple other factors involved in resuscitative care, limit our ability to assess the clinical significance of this work on patient outcomes.
Implementation of a QI initiative has resulted in decreased interruptions in chest compressions and decreased variability of time to defibrillator pad placement for patients in cardiac arrest. Multiple critical key drivers have propelled this improvement and have sustained performance, including defined roles and responsibilities, a standardized process for CPR transition and defibrillator pad placement, and education and empowerment of frontline providers.
Dr Hoehn conceptualized and designed the study, conceptualized and revised the choreography, and assisted with video review for data collection and analysis; Ms Cabrera-Thurman conceptualized and designed the study, conceptualized and revised the choreography, assisted with video review for data collection and analysis, and led the education with paramedics and patient care assistants; Dr Vukovic assisted with video review for data collection and completed the data analysis; Ms Oehler assisted with the chart design used for the project and completed the data analysis; Ms Frey conceptualized and designed the study and conceptualized and revised the choreography; Mr Helton conceptualized and revised the choreography, assisted with video review for data collection and analysis, and led the education with paramedics and patient care assistants; Dr Geis conceptualized and designed the study; Dr Kerrey conceptualized and designed the study, conceptualized and revised the choreography, and reviewed the data analysis; and all authors drafted and revised the complete manuscript, approved the final manuscript as submitted, and agree to be accountable for all aspects of the work.
FUNDING: No external funding.
POTENTIAL CONFLICT OF INTEREST: The authors have indicated they have no potential conflicts of interest to disclose.
FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.