Abstract
Background:
Bystander cardiopulmonary resuscitation (CPR) is a key link in the chain of survival for out-of-hospital cardiac arrest (OHCA). While bystander CPR initiated within 10 minutes is associated with improved survival in adults, this association in children is unknown. We examined the association between time to bystander CPR initiation and survival outcomes in children with OHCA.
Methods:
We conducted a retrospective cohort study in the U.S. using the Cardiac Arrest Registry to Enhance Survival by including pediatric (<18 years) patients with non-traumatic OHCA from 2013 to 2023. Time to bystander CPR was estimated from the 9-1-1 call timestamp and bystander-report of time of CPR initiation. The primary outcome was survival to hospital discharge; the secondary outcome was favorable neurological survival. Multivariable hierarchical logistic regression models evaluated the association between time to bystander CPR and survival outcomes, adjusted for demographic and cardiac arrest characteristics.
Results:
Of 10,991 pediatric OHCAs (mean[SD] age, 4.5[5.8]years; 40.9%female), 5,446(49.5%) received bystander CPR, whereas 5,545(50.5%) received CPR after first responder arrival. Median time to bystander CPR was 3.0 minutes(IQR:1.0–9.0). Overall, 1,677(15.3%) survived to discharge, and 1,420(12.9%) had favorable neurological survival. As compared with patients without bystander CPR, there was a graded inverse relationship between time to bystander CPR and survival to discharge during the first 5 minutes of initiation (0–1 minute: adjusted odds ratio [aOR],1.91[95% CI: 1.65–2.20]; 2–3 minutes: aOR,1.98[1.63–2.40]; 4–5 minutes: aOR,1.37[1.09–1.72]]). A similar pattern was observed for the outcome of favorable neurological survival, with approximately two-fold higher odds when CPR was initiated within 3 minutes and no benefit when initiated after 5 minutes.
Conclusions:
In pediatric OHCA, earlier bystander CPR initiation was associated with improved survival outcomes, with the greatest benefit observed within the first 5 minutes. These findings underscore the importance of early CPR initiation and the need for continued efforts to enhance bystander response.
Keywords: chain of survival, cardiac arrest, OHCA, children
INTRODUCTION
Early cardiopulmonary resuscitation (CPR) in out-of-hospital cardiac arrest (OHCA) increases the likelihood of survival.1 To improve early initiation of CPR, national organizations have worked to increase bystander CPR rates, emphasizing this intervention as a critical link in the chain of survival.1 While prior studies have found that bystander CPR is associated with higher survival for adults and children with OHCA,2–6 less is known about the timing of when bystander CPR is initiated in children with OHCA and survival. This may be because the estimated annual incidence of pediatric OHCA is only ~4,000 cases in the U.S., which is substantially lower than the estimated annual incidence of 350,000 in adults.7
Unlike adults, the age distribution and etiology of OHCA differ in children. Over 40% of pediatric OHCAs are in infants under 1 year of age, and respiratory arrest and drowning are common reasons for OHCA in children.3,6 As children are less likely to have a cardiac etiology of arrest, the initially detected cardiac arrest rhythm is more frequently non-shockable.2,8 Given these differences in etiology of cardiac arrest, the association between time of bystander CPR initiation and survival outcomes for pediatric OHCA may differ from that in adults, with important implications for dispatcher recognition and timeliness of response. In a recent study in adults with OHCA, bystander CPR initiated up to 8–9 minutes after OHCA onset was associated with improved survival as compared to patients without bystander CPR.9 However, as a higher proportion of pediatric OHCA is due to respiratory or drowning etiologies, and as the current paradigm of Basic Life Support training is hands-only CPR, whether this relationship between bystander CPR initiation and survival outcomes is different in pediatric OHCA remains unclear.
To address this knowledge gap, we analyzed data from the Cardiac Arrest Registry to Enhance Survival (CARES), the largest OHCA surveillance registry in the U.S. Our study examined the association between time to bystander CPR initiation and survival outcomes in pediatric OHCA, as compared to OHCAs without bystander CPR, and whether this relationship differed by age group, etiology of OHCA, and witnessed status of arrest. Findings from these analyses could inform future resuscitation strategies and training, especially if this time relationship differs from adult OHCAs.
METHODS
Data Source
This study used data from CARES, a prospective, multicenter, observational registry of OHCA in the U.S. CARES was established as a collaboration between the Centers for Disease Control and Prevention and the Emory University School of Medicine's Department of Emergency Medicine.10,11 The registry currently collects OHCA data from 33 state-based registries and community sites in 14 additional states, covering a catchment area of approximately 175 million people, or roughly 53% of the U.S. population. OHCA patients and their outcomes are identified through review of data from 9-1-1 dispatch centers, emergency medical service (EMS) agencies, and receiving hospitals. All patients with a confirmed non-traumatic OHCA, defined as apnea and pulselessness in whom resuscitation or defibrillation is attempted, are included. Data on each patient are collected using standardized international Utstein definitions,12 and CARES staff perform audits to ensure completeness.. As this study used de-identified data, it was granted a waiver of informed consent by the Saint Luke's Hospital Institutional Review Board. The data supporting the findings of this study are available upon direct request to the CARES registry.
Study Cohort
Within CARES, we identified pediatric patients (<18 years) with non-traumatic OHCA between January 1, 2013, and December 31, 2023 (Figure 1). Among 15,484 patients with available time-to-CPR data, we excluded those with a time to CPR exceeding 30 minutes as these were likely due to inaccurate data entry (n=2,433). We also excluded OHCAs witnessed by first responders or EMS (n=1,328), since the focus of our study was to assess the relationship between time to bystander CPR initiation and survival outcomes. Additionally, we excluded cases occurring in nursing homes or healthcare facilities (n=424), with missing data on survival (n=184), a negative time to CPR (n=123), and with missing data on the witnessed status of the cardiac arrest (n=1). The final study cohort consisted of 10,991 pediatric patients with OHCA.
Figure 1. Definition of the Study Cohort.
Abbreviations: CPR, cardiopulmonary resuscitation; OHCA, out-of-hospital cardiac arrest.
Independent Variables and Study Outcomes
The independent variable was time to bystander CPR, with a bystander defined as any layperson not part of the emergency 9-1-1 response team. Within CARES, this is derived from the variables, time of first CPR (defined as the time of first CPR attempt) and time of the 9-1-1 call.. Thus, time to CPR is estimated as the difference between the time of bystander CPR initiation and the time of the 9-1-1 call at the dispatch center, with the 9-1-1 call time used as a surrogate estimate for the time of cardiac arrest. In cases where a bystander does not have information on time of initiating CPR, this time is estimated by EMS personnel or is left blank.
The primary outcome was survival to hospital discharge. The secondary outcome was favorable neurological survival, defined as survival with a discharge Pediatric Cerebral Performance Category score of 1 (no neurological disability), 2 (mild disability) or 3 (moderate disability).
Statistical Analysis
For descriptive purposes, we examined the distribution of time to CPR initiation among those who received bystander CPR versus those who received CPR exclusively from a first responder or EMS (i.e., without bystander CPR). Given the large sample size, baseline characteristics between patients with and without bystander CPR were compared using standardized differences, with a standardized difference greater than 10% denoting a significant imbalance.
To examine the association between time to bystander CPR and survival to discharge, we constructed generalized linear mixed models with a binomial distribution and logit link, with EMS agency modeled as a random effect to account for the clustering of patient outcomes within sites. Given the anticipated non-linear relationship between time to CPR initiation and survival outcomes, and to facilitate clinically meaningful interpretation, we categorized time to bystander CPR into 2-minute interval groups (0–1, 2–3, 4–5, 6–7, 8–9, and ≥10 minutes) and compared them to the reference group of patients with no bystander CPR (i.e., a 7-level categorical variable). Besides adjusting for time of bystander CPR initiation, this model also adjusted for the following variables as fixed effects: pediatric age group (<1 year, 1 to <8 years, 8 to <18 years), sex, race, presumed arrest etiology (cardiac, respiratory, drowning, other), whether the arrest was witnessed, and arrest location (public versus home). We did not adjust for initially detected cardiac rhythm (shockable versus non-shockable) as it is a potential mediator of outcomes, contingent on when CPR is initiated. Nonetheless, we performed an additional analysis further adjusting for initial rhythm. To further evaluate whether the association between time to bystander CPR and outcomes was affected by the type of CPR provided, we adjusted for the type of bystander CPR (compressions only, compressions with ventilations, ventilations only, unknown, or missing).
To determine if the association between time to bystander CPR and survival to discharge varied by age group, arrest etiology, and witnessed status of the arrest, we decided, a priori, to test for an interaction between time to bystander CPR and these variables and report estimates of effect, stratified by these variables. Finally, we repeated the above analyses for the secondary outcome of favorable neurological survival.
Because the primary analysis was restricted to patients with available data on the main exposure (time to bystander CPR), those with missing time to CPR data were excluded. To assess for potential selection bias, we performed a sensitivity analysis using inverse probability weighting to evaluate whether study results were similar when accounting for a patient’ probability to be excluded due to missing data on time to CPR. A 2-tailed P value of <0.05 was considered statistically significant. All analyses were performed with SAS software, version 9.4 (SAS Institute, Cary, NC).
RESULTS
Among 10,991 pediatric patients with OHCA, 5,446 (49.5%) received bystander CPR, of which 1971 (36.2%) were provided rescue breaths in addition to chest compressions (Table S1). The median age was 1.0 year of age (interquartile range: 0.3 to 9.0), of which 5,229 (47.6%) were infants (<1 year), 2,801 (25.5%) were young children (1 to <8 years), and 2,961 (26.9%) were older children (8 to <18 years). Overall, 40.9% were female, and among those with available race and ethnicity data, 35.7% were White, 30.9% were Black, 12.2% were Hispanic, and 1.7% were Asian. The most common arrest etiology was a presumed cardiac cause (43.3%); however, 37.3% had a respiratory etiology, 10.0% drowning, and 9.4% were of other causes, including drug overdose. Most OHCAs occurred at home (85.6%). Pediatric patients with bystander CPR were more likely to be older children, of White race, and present with an initial shockable rhythm (Table 1).
Table 1: Patient Characteristics.
Comparison of patients who received bystander CPR vs. those with CPR initiated by first responder or EMS personnel.
| Total | CPR Initiated By | Absolute Standardized Difference, %* | ||
|---|---|---|---|---|
| n = 10991 | Bystander n = 5446 |
First Responder or EMS n = 5545 |
||
| Time to CPR | 50.5 | |||
| Mean ± SD | 7.8 ± 7.5 | 5.9 ± 7.2 | 9.6 ± 7.3 | |
| Median (IQR) | 5.6 (2.0, 11.0) | 3.0 (1.0, 9.0) | 8.0 (4.3, 13.0) | |
| Age, years | 12.0 | |||
| Mean ± SD | 4.5 ± 5.8 | 4.9 ± 6.0 | 4.2 ± 5.7 | |
| Median (IQR) | 1.0 (0.3, 9.0) | 1.0 (0.3, 10.0) | 0.8 (0.2, 7.0) | |
| Age Group, years | 14.3 | |||
| <1 | 5229 (47.6%) | 2400 (45.9%) | 2829 (54.1%) | |
| 1 to <8 | 2801 (25.5%) | 1449 (51.7%) | 1352 (48.3%) | |
| 8 to <18 | 2961 (26.9%) | 1597 (53.9%) | 1364 (46.1%) | |
| Sex | 4.4 | |||
| Female | 4494 (40.9%) | 2167 (48.2%) | 2327 (51.8%) | |
| Male | 6496 (59.1%) | 3279 (50.5%) | 3217 (49.5%) | |
| Missing | 1 | 1 | ||
| Race | 30.7 | |||
| Black | 3400 (30.9%) | 1386 (40.8%) | 2014 (59.2%) | |
| White | 3922 (35.7%) | 2303 (58.7%) | 1619 (41.3%) | |
| Hispanic | 1345 (12.2%) | 598 (44.5%) | 747 (55.5%) | |
| Asian | 190 (1.7%) | 95 (50.0%) | 95 (50.0%) | |
| Native | 117 (1.1%) | 52 (44.4%) | 65 (55.6%) | |
| Unknown | 2017 (18.4%) | 1012 (50.2%) | 1005 (49.8%) | |
| Arrest Location | 7.9 | |||
| Home/Residence | 9411 (85.6%) | 4587 (48.7%) | 4824 (51.3%) | |
| Public | 1580 (14.4%) | 859 (54.4%) | 721 (45.6%) | |
| Witnessed Status of Arrest | 21.9 | |||
| Unwitnessed | 6800 (61.9%) | 3079 (45.3%) | 3721 (54.7%) | |
| Witnessed | 4191 (38.1%) | 2367 (56.5%) | 1824 (43.5%) | |
| Presumed Cardiac Arrest Etiology | 11.4 | |||
| Drowning/Submersion | 1100 (10.0%) | 634 (57.6%) | 466 (42.4%) | |
| Presumed Cardiac Etiology | 4757 (43.3%) | 2330 (49.0%) | 2427 (51.0%) | |
| Respiratory/Asphyxia | 4103 (37.3%) | 2009 (49.0%) | 2094 (51.0%) | |
| Other | 1031 (9.4%) | 473 (45.9%) | 558 (54.1%) | |
| Initially Detected Rhythm | 12.7 | |||
| Non-shockable | 9979 (90.8%) | 4844 (48.5%) | 5135 (51.5%) | |
| Shockable | 1003 (9.1%) | 597 (59.5%) | 406 (40.5%) | |
| Missing | 9 (0.1%) | 5 | 4 | |
Percentage absolute standardized difference greater than 10% indicates a significant imbalance between groups.
Abbreviations: CPR, cardiopulmonary resuscitation; EMS, emergency medical services.
The proportion of patients receiving bystander CPR remained stable at approximately 50% throughout the study period, and the median time to bystander CPR initiation was slightly longer over time (Table S2).
Time to Bystander CPR and Survival
For persons in whom bystander CPR was performed, the median time to CPR initiation was 3.0 minutes (IQR 1.0–9.0), with 2,109 (38.7%) patients received CPR within 1 minute, 2,943 (54%) within 3 minutes, and 3,620 (66.5%) within 5 minutes (Figure 2). For persons without bystander CPR, the median time to CPR initiation by a first responder or EMS was 8.0 minutes (IQR 4.3–13.0). Over the course of the study, the proportions of patients who survived to hospital discharge and those with favorable neurological survival remained relatively stable (Table S2). Overall, 1,677 (15.3%) patients survived to hospital discharge. Survival to discharge was highest when bystander CPR was initiated within the first 3 minutes of the 9-1-1 call (26–27%) and progressively declined to 6.0% when initiated at ≥10 minutes (Table 2; Figure S1). As compared to patients without bystander CPR, there was a graded inverse relationship between time to bystander CPR and survival to discharge, with a survival association when bystander CPR was initiated up to 5 minutes after the 9-1-1 call: for bystander CPR at 0–1 minute (adjusted odds ratio [aOR], 1.91 [95% CI: 1.65–2.20]); 2–3 minutes (aOR, 1.98 [1.63–2.40]); 4–5 minutes (aOR, 1.37 [1.09–1.72]); 6–7 minutes (aOR, 0.76 [0.51–1.13]); 8–9 minutes (aOR, 0.67 [0.41–1.08]); ≥10 minutes (aOR, 0.59 [0.46–0.77]). These results were unchanged when time to bystander CPR was modeled in 1-minute intervals (Table S3). Among patients with information on modality of CPR provided (4,812/5,446 [88.4%]), further adjustment for the type of bystander CPR yielded a similar relationship, although the effect sizes were modestly attenuated (Table S4). Results were also consistent after additional adjustment for initial rhythm; however, the associations were attenuated and no longer statistically significant for survival to discharge in the 4–5 minute stratum (Table S5).
Figure 2. Time to Bystander CPR.
Among those who received bystander cardiopulmonary resuscitation (CPR), the distribution of time to bystander CPR is shown. Median time to bystander CPR was 3.0 minutes (interquartile range of 1.0 to 9.0 minutes).
Abbreviations: CPR, cardiopulmonary resuscitation; OHCA, out-of-hospital cardiac arrest
Table 2: Association between Time to Bystander CPR and Survival to Discharge and Favorable Neurological Survival.
Bystander CPR initiated within the first 5 minutes of arrest was associated with improved survival for each outcome.
| Time to Bystander CPR (min.) | Unadjusted Survival (%) | Adjusted OR (95% CI) |
|---|---|---|
| SURVIVAL TO DISCHARGE | ||
| No bystander CPR* | 640/5545 (11.5) | Reference |
| 0–1 | 552/2109 (26.2) | 1.91 (1.65–2.20) |
| 2–3 | 227/834 (27.2) | 1.98 (1.63–2.40) |
| 4–5 | 128/677 (18.9) | 1.37 (1.09–1.72) |
| 6–7 | 32/305 (10.5) | 0.76 (0.51–1.13) |
| 8–9 | 21/245 (8.6) | 0.67 (0.41–1.08) |
| 10+ | 77/1276 (6.0) | 0.59 (0.46–0.77) |
| FAVORABLE NEUROLOGICAL SURVIVAL | ||
| No bystander CPR* | 522/5545 (9.4) | Reference |
| 0–1 | 488/2109 (23.1) | 2.00 (1.72–2.32) |
| 2–3 | 200/834 (24.0) | 2.06 (1.68–2.52) |
| 4–5 | 112/677 (16.5) | 1.46 (1.14–1.86) |
| 6–7 | 27/305 (8.9) | 0.79 (0.51–1.22) |
| 8–9 | 16/245 (6.5) | 0.62 (0.36–1.07) |
| 10+ | 55/1276 (4.3) | 0.54 (0.40–0.72) |
No bystander CPR indicates CPR was initiated by first responders or EMS personnel after their arrival.
Abbreviations: CI, confidence interval; CPR, cardiopulmonary resuscitation; EMS, emergency medical services; OR, odds ratio.
We found a significant interaction between time to bystander CPR initiation and etiology of arrest (P=0.002). In fully adjusted models, improved survival (OR >1.0) was seen when bystander CPR was initiated within 3 minutes for respiratory arrests, 5 minutes for arrests due to drowning, and 7 minutes for arrests with presumed cardiac etiology (Table 3). Notably, drowning cases showed the strongest survival association (aOR >3) when CPR was initiated within 5 minutes, potentially attributable to more rapid arrest recognition in supervised environments (such as swimming pools) and the greater availability of trained responders (such as lifeguards) who can provide high-quality resuscitation. There was a borderline interaction between time to bystander CPR initiation and age group (P=0.053). Overall, the pattern suggests a shorter window of benefit associated with bystander CPR in younger children, with a survival association observed when CPR was initiated within 3 minutes in infants (<1 year), within 5 minutes in young children (1 to <8 years), and within 7 minutes in older children (8 to <18 years). The relationship between time of bystander CPR initiation and survival to discharge did not differ by the witnessed status of the OHCA (P=0.32).
Table 3: Time to Bystander CPR and Survival to Discharge by Age Group, Cardiac Arrest Etiology, and Witnessed Status.
The association between time to bystander CPR and survival varied by patient characteristics. The window of benefit was shorter for younger patients and those with respiratory arrest.
| Time to Bystander CPR (minutes) | P for Interaction | ||||||
|---|---|---|---|---|---|---|---|
| 0–1 | 2–3 | 4–5 | 6–7 | 8–9 | 10+ | ||
| AGE GROUP | 0.053 | ||||||
| Infants (< 1 yr) | 1.92 (1.53–2.43) | 1.66 (1.18–2.35) | 0.91 (0.58–1.43) | 0.51 (0.23–1.12) | 0.46 (0.16–1.28) | 0.48 (0.31–0.76) | |
| Young children (1 to < 8 yrs) | 1.80 (1.38–2.35) | 2.46 (1.74–3.47) | 2.24 (1.54–3.24) | 0.63 (0.29–1.40) | 0.68 (0.30–1.55) | 0.55 (0.33–0.92) | |
| Older children (8 – 17 yrs) | 1.99 (1.58–2.51) | 1.94 (1.42–2.64) | 1.14 (0.76–1.69) | 1.16 (0.63–2.12) | 0.86 (0.41–1.80) | 0.75 (0.51–1.13) | |
| CARDIAC ARREST ETIOLOGY | 0.002 | ||||||
| Respiratory | 1.63 (1.30–2.05) | 1.40 (1.0–1-95) | 0.87 (0.57–1.32) | 0.46 (0.22–1.00) | 0.73 (0.32–165) | 0.37 (0.23–0.57) | |
| Drowning | 3.64 (2.38–5.55) | 4.85 (2.94–7.99) | 3.42 (2.08–5.61) | 0.88 (0.33–2.32) | 1.42 (0.55–3.67) | 1.25 (0.69–2.26) | |
| Cardiac | 1.95 (1.57–2.43) | 2.07 (1.54–2.79) | 1.28 (0.87–1.89) | 1.18 (0.64–2.18) | 0.51 (0.21–1.24) | 0.74 (0.49–1.13) | |
| Other | 1.46 (0.94–2.29) | 1.54 (0.80–2.97) | 1.51 (0.73–3.12) | 0.75 (0.24–2.34) | 0.39 (0.09–1.81) | 0.58 (0.26–1.27) | |
| WITNESS STATUS | 0.32 | ||||||
| Witnessed | 1.87 (1.58–2.21) | 1.81 (1.44–2.29) | 1.11 (0.83–1.50) | 0.88 (0.53–1.46) | 0.62 (0.32–1.19) | 0.59 (0.39–0.88) | |
| Unwitnessed | 1.93 (1.49–2.50) | 2.34 (1.68–3.25) | 1.84 (1.31–2.59) | 0.59 (0.30–1.19) | 0.74 (0.37–1.49) | 0.61 (0.44–0.86) | |
Abbreviations: CPR, cardiopulmonary resuscitation.
A total of 1,420 (12.9%) patients had favorable neurological survival. A similar pattern was observed between time to bystander CPR and favorable neurological survival. As compared to patients without bystander CPR, children who received bystander CPR within 3 minutes had approximately two-fold higher odds of favorable neurological survival, with similar effect sizes observed for CPR initiated at 0–1 minutes (aOR, 2.00 [1.72–2.32]) and 2–3 minutes (aOR, 2.06 [1.68–2.52]), and there was no significant association when bystander CPR was initiated after 5 minutes of the 9-1-1 call (Table 2; Figure S2).
We compared patients included in the analytic cohort with those excluded due to missing data on time to bystander CPR. Patients with missing data were more likely to be younger, have an unwitnessed arrest, and present with an initial non-shockable rhythm (Table S6). In sensitivity analyses using inverse probability weighting to account for exclusion of patients with missing data on time to CPR, results were consistent with the primary analyses, which showed higher survival and favorable neurological outcomes when bystander CPR was initiated within 5 minutes of cardiac arrest (Table S7).
DISCUSSION
Although bystander CPR is associated with higher survival in children with OHCA,2,6 the association between time to bystander CPR initiation and survival is less clearly defined. Leveraging data from a large U.S. registry, we found that nearly half of patients received bystander CPR, with a median time to CPR initiation of 3 minutes. Compared with children without bystander CPR, there was a graded inverse relationship between time to bystander CPR and outcomes, with nearly 2-fold higher odds of survival to hospital discharge and favorable neurologic survival when CPR was initiated within the first 3 minutes and no association with survival outcomes when bystander CPR was initiated after 5 minutes. There was also some difference in this relationship between time to bystander CPR and survival by patients’ age group and cardiac arrest etiology. Collectively, these findings underscore the critical, time-sensitive relationship between time to bystander CPR initiation in children and survival after OHCA and suggest that the window for this potentially life-saving intervention may be smaller than it is for adults with OHCA.
Prior studies on the association between time to bystander CPR and OHCA survival in children are lacking. Earlier studies examining time to bystander CPR were small in sample size, involved adults, and lacked information on cardiac arrest characteristics.13,14 A recent study using CARES data found a graded inverse relationship between time to bystander CPR and survival in adults with OHCA.9 As compared with persons without bystander CPR, those with bystander CPR initiated within 1 minute had 78% higher odds of survival to hospital discharge (adjusted OR: 1.78, 95% CI: 1.73–1.84) and the association between time to bystander CPR and survival in adults with OHCA persisted even when bystander CPR was initiated at 8 to 9 minutes after the 9-1-1 call (adjusted OR: 1.13, 95% CI: 1.03–1.25).A Swedish study of young persons (0–30 years) also found an inverse, graded relationship between time to any CPR (by a bystander or 9-1-1 responder) and survival, but half of patients in that study were young adults aged 21–30 and the authors did not specifically examine time to bystander CPR vs. those without bystander CPR.15 To our knowledge, our study is the first to examine the association between time of bystander CPR initiation in children and survival, as well as to examine for potential differences in this relationship by age group, witnessed status of arrest, and etiology of arrest. In contrast to the relationship found for adults with OHCA,9 we found that the association between bystander CPR and survival decreases rapidly during the first 5 minutes, with no association with survival when bystander CPR is initiated after 5 minutes of the 9-1-1 call.
There are several reasons why the association between time to bystander CPR initiation and survival in children has a smaller time period for benefit, as compared with adults. First, respiratory and drowning events are more common causes of cardiac arrest in children,8 and these comprised 47.3% (5203/10,991) of pediatric OHCAs in our study. In contrast, over 80% of OHCAs in adults have a cardiac etiology.16 Because rescue breathing may be an important intervention for these pediatric arrest etiologies, the associated benefit of bystander CPR in children may be more time-sensitive since contemporary CPR training for laypersons primarily emphasizes hands-only CPR. In fact, we found that only 36% of children who received bystander CPR had standard CPR delivered (i.e., chest compressions with rescue breathing). This is supported by the fact that there was a significant interaction between time of CPR initiation and arrest etiology in our study, with respiratory arrests having the smallest window of benefit for bystander CPR (≤3 minutes), as compared to OHCAs due to drowning (≤5 minutes) and cardiac etiologies (≤7 minutes), with the last group mirroring the time window for OHCAs in adults, in which 87% of OHCAs were of cardiac etiology.9 Second, children are less likely than adults to have an initially detected cardiac arrest rhythm that is shockable.8 Persons with a non-shockable rhythm are less likely to survive an OHCA,17 and this may influence the association between time to bystander CPR initiation and survival in children. Third, bystanders may be more hesitant to initiate or encounter greater difficulty in performing effective CPR on infants and young children than adults. This is supported by the fact that our interaction analysis between time to bystander CPR and age group was borderline significant (P=0.053), with a smaller window of benefit for infants than older children.
Throughout the U.S., there have been important efforts to disseminate CPR knowledge and training. The recent 2025 American Heart Association guidelines highlighted the most critical initial response measures for OHCA: early recognition, emergency response activation, and CPR initiation, which form the first three links in the Chain of Survival.1 Early recognition of cardiac arrest is the critical first step in this chain and directly influences the timeliness of bystander CPR. Contemporary pediatric resuscitation guidelines emphasize that survival from OHCA is higher in regions with greater rates of witnessed arrests and lay responder CPR, underscoring the importance of rapid identification and response. Accordingly, efforts to improve outcomes should not only focus on CPR quality and timeliness of bystander response but also on earlier recognition of arrest, prompt activation of emergency medical services, and the role of dispatchers in facilitating early recognition and intervention.
Over the past few decades, there has been a strong emphasis on CPR training in schools.18,19 In 2004, the American Heart Association recommended that all teachers be trained in CPR and first aid and incorporate CPR instruction into student emergency preparedness programs.20 Since then, legislative initiatives have increasingly sought to make CPR training mandatory in school curricula.21 However, our study highlights additional considerations. Since half of pediatric OHCAs occur in children 1 year or younger, there may be a need to focus CPR training on parents and other household members who come for routine pediatric clinic visits. Beyond teaching proper CPR techniques, our findings highlight the critical importance of early CPR initiation in children with OHCA. The shorter time window for potential survival benefit when initiating bystander CPR in children with OHCA has implications for future layperson CPR training programs as well as for dispatchers who engage bystanders to initiate CPR. Emphasizing the critical importance of initiating dispatcher-assisted CPR as soon as cardiac arrest is suspected could significantly improve survival outcomes.22 Lastly, because pediatric OHCA is often driven by respiratory and drowning etiologies, public education and CPR training should emphasize the importance of standard CPR (compressions with rescue breaths), particularly for parents and caregivers and in young children with OHCA who are more likely to have a respiratory etiology for their cardiac arrest. However, compression-only CPR remains far better than no CPR and should be initiated immediately when cardiac arrest is suspected.23
Our study should be considered with certain limitations. First, time to bystander CPR is estimated from the reported time of the 9-1-1 call and the time of CPR initiation, as reported by the bystander. Inaccuracies in either component could lead to misclassification of time to bystander CPR, including potential clustering of observations in which CPR initiation is presumed to coincide with the 9-1-1 call. Such misclassification is expected to be nondifferential and would bias our findings toward the null, although differential misclassification is also possible. For example, the accuracy of time reporting may vary by event characteristics such as whether the arrest was witnessed, the setting, or the presence of trained responders, which could systematically influence observed associations. We did examine whether time to bystander CPR and survival outcomes differed for witnessed arrests (where time reporting may be more reliable) and found no significant interaction vs. unwitnessed arrests. Second, we lacked data on the quality of bystander CPR, the number of bystanders involved, and whether CPR was initiated independently or in response to dispatcher instructions, all of which could impact outcomes. Third, although dispatcher-assisted CPR is captured in the CARES registry, this variable documents whether a dispatcher provided any CPR instruction during the 9-1-1 call, irrespective of whether CPR was initiated by a bystander or whether a bystander had already begun CPR, limiting its utility for analysis. Fourth, although our models adjusted for demographic and arrest characteristics, unmeasured confounding may persist. In particular, bystander-level factors such as education, health literacy, prior CPR training, and socioeconomic status are not captured in CARES and may influence both the timeliness of CPR initiation and patient outcomes. Finally, although the CARES registry captures data from the majority of the U.S. population, our findings may not be fully generalizable to all children in the U.S., particularly those living in rural areas.
Conclusion
In this nationally representative study of over 10,000 children with OHCA, we found that earlier initiation of bystander CPR was associated with better survival outcomes, with the highest survival observed when CPR was started within the first 5 minutes. These findings underscore the critical, time-sensitive relationship between time to bystander CPR initiation in children and survival after OHCA and suggest that the window for this potentially life-saving intervention may be smaller than it is for adults with OHCA.
Supplementary Material
WHAT IS KNOWN
Early bystander CPR is associated with improved survival in adults with out-of-hospital cardiac arrest (OHCA), but the association between timing of bystander CPR and survival in pediatric cases has not been evaluated.
Prior studies in children have focused on whether bystander CPR occurred, with limited data on the timing of CPR initiation and survival or neurological outcomes.
WHAT THE STUDY ADDS
A graded, time-sensitive relationship between CPR timing and survival outcomes was observed: each minute of delay significantly reduced survival, and bystander CPR initiated within 3 minutes was associated with nearly two-fold higher odds of survival with less clear benefit after 5 minutes.
The window for benefit for time to bystander CPR was narrower for infants under one year of age and those with respiratory cause for cardiac arrest.
The findings highlight the urgent need to expand CPR education to improve rapid recognition of cardiac arrest and immediate CPR initiation in pediatric OHCA by parents, teachers, caregivers, and other lay rescuers.
Sources of Funding:
Dr. Abdel Jawad is supported by the National Heart, Lung, and Blood Institute under Award Number T32H110837. Dr. Chan is supported by the National Heart, Lung and Blood Institutes of Health under Award Number R01HL160734 and R01HL178638, as well as from the American Heart Association for editorial work. The contents of this project are solely the responsibility of the authors and do not necessarily represent official views of the National Center for Advancing Translational Sciences; the National Heart, Lung, and Blood Institute; the National Institutes of Health; or the Department of Health and Human Services.
Nonstandard Abbreviations and Acronyms
- CARES
Cardiac Arrest Registry to Enhance Surviva
- CPR
cardiopulmonary resuscitation
- EMS
emergency medical service
- OHCA
out-of-hospital cardiac arrest
Footnotes
Conflict of Interest Disclosures:
None.
REFERENCES
- 1.Joyner BL Jr., Dewan M, Bavare A, de Caen A, DiMaria K, Donofrio-Odmann J, Fosse G, Haskell S, Mahgoub M, Meckler G, et al. Part 6: Pediatric Basic Life Support: 2025 American Heart Association and American Academy of Pediatrics Guidelines for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care. Circulation. 2025;152:S424–s447. doi: 10.1161/cir.0000000000001370 [DOI] [PubMed] [Google Scholar]
- 2.Akahane M, Tanabe S, Ogawa T, Koike S, Horiguchi H, Yasunaga H, Imamura T. Characteristics and outcomes of pediatric out-of-hospital cardiac arrest by scholastic age category. Pediatr Crit Care Med. 2013;14:130–136. doi: 10.1097/PCC.0b013e31827129b3 [DOI] [PubMed] [Google Scholar]
- 3.Fink EL, Prince DK, Kaltman JR, Atkins DL, Austin M, Warden C, Hutchison J, Daya M, Goldberg S, Herren H, et al. Unchanged pediatric out-of-hospital cardiac arrest incidence and survival rates with regional variation in North America. Resuscitation. 2016;107:121–128. doi: 10.1016/j.resuscitation.2016.07.244 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Herlitz J, Ekström L, Wennerblom B, Axelsson A, Bång A, Holmberg S. Effect of bystander initiated cardiopulmonary resuscitation on ventricular fibrillation and survival after witnessed cardiac arrest outside hospital. Br Heart J. 1994;72:408–412. doi: 10.1136/hrt.72.5.408 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Sasson C, Rogers MA, Dahl J, Kellermann AL. Predictors of survival from out-of-hospital cardiac arrest: a systematic review and meta-analysis. Circ Cardiovasc Qual Outcomes. 2010;3:63–81. doi: 10.1161/circoutcomes.109.889576 [DOI] [PubMed] [Google Scholar]
- 6.Forrest A, Butt WW, Namachivayam SP. Outcomes of children admitted to intensive care after out-of-hospital cardiac arrest in Victoria, Australia. Crit Care Resusc. 2017;19:150–158. [PubMed] [Google Scholar]
- 7.Martin SS, Aday AW, Almarzooq ZI, Anderson CAM, Arora P, Avery CL, Baker-Smith CM, Barone Gibbs B, Beaton AZ, Boehme AK, et al. 2024 Heart Disease and Stroke Statistics: A Report of US and Global Data From the American Heart Association. Circulation. 2024;149:e347–e913. doi: 10.1161/cir.0000000000001209 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Herlitz J, Svensson L, Engdahl J, Gelberg J, Silfverstolpe J, Wisten A, Angquist KA, Holmberg S. Characteristics of cardiac arrest and resuscitation by age group: an analysis from the Swedish Cardiac Arrest Registry. Am J Emerg Med. 2007;25:1025–1031. doi: 10.1016/j.ajem.2007.03.008 [DOI] [PubMed] [Google Scholar]
- 9.O'Keefe EL, Abdel Jawad M, Kennedy KF, Nguyen D, Ikemura N, Chan PS. Time to bystander CPR and survival for witnessed out-of-hospital cardiac arrest. Resuscitation. 2025;209:110566. doi: 10.1016/j.resuscitation.2025.110566 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.McNally B, Stokes A, Crouch A, Kellermann AL. CARES: Cardiac Arrest Registry to Enhance Survival. Ann Emerg Med. 2009;54:674–683.e672. doi: 10.1016/j.annemergmed.2009.03.018 [DOI] [PubMed] [Google Scholar]
- 11.McNally B, Robb R, Mehta M, Vellano K, Valderrama AL, Yoon PW, Sasson C, Crouch A, Perez AB, Merritt R, et al. Out-of-hospital cardiac arrest surveillance --- Cardiac Arrest Registry to Enhance Survival (CARES), United States, October 1, 2005--December 31, 2010. MMWR Surveill Summ. 2011;60:1–19. [PubMed] [Google Scholar]
- 12.Jacobs I, Nadkarni V, Bahr J, Berg RA, Billi JE, Bossaert L, Cassan P, Coovadia A, D'Este K, Finn J, et al. Cardiac arrest and cardiopulmonary resuscitation outcome reports: update and simplification of the Utstein templates for resuscitation registries: a statement for healthcare professionals from a task force of the International Liaison Committee on Resuscitation (American Heart Association, European Resuscitation Council, Australian Resuscitation Council, New Zealand Resuscitation Council, Heart and Stroke Foundation of Canada, InterAmerican Heart Foundation, Resuscitation Councils of Southern Africa). Circulation. 2004;110:3385–3397. doi: 10.1161/01.Cir.0000147236.85306.15 [DOI] [PubMed] [Google Scholar]
- 13.Larsen MP, Eisenberg MS, Cummins RO, Hallstrom AP. Predicting survival from out-of-hospital cardiac arrest: a graphic model. Ann Emerg Med. 1993;22:1652–1658. doi: 10.1016/s0196-0644(05)81302-2 [DOI] [PubMed] [Google Scholar]
- 14.Valenzuela TD, Roe DJ, Cretin S, Spaite DW, Larsen MP. Estimating effectiveness of cardiac arrest interventions: a logistic regression survival model. Circulation. 1997;96:3308–3313. doi: 10.1161/01.cir.96.10.3308 [DOI] [PubMed] [Google Scholar]
- 15.Fovaeus H, Holmen J, Mandalenakis Z, Herlitz J, Rawshani A, Castellheim AG. Out-of-hospital cardiac arrest: Survival in children and young adults over 30 years, a nationwide registry-based cohort study. Resuscitation. 2024;195:110103. doi: 10.1016/j.resuscitation.2023.110103 [DOI] [PubMed] [Google Scholar]
- 16.Chan PS, Girotra S, Blewer A, Kennedy KF, McNally BF, Benoit JL, Starks MA. Race and Sex Differences in the Association of Bystander CPR for Cardiac Arrest. Circulation. 2024;150:677–686. doi: 10.1161/circulationaha.124.068732 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Herlitz J, Svensson L, Engdahl J, Silfverstolpe J. Characteristics and outcome in out-of-hospital cardiac arrest when patients are found in a non-shockable rhythm. Resuscitation. 2008;76:31–36. doi: 10.1016/j.resuscitation.2007.06.027 [DOI] [PubMed] [Google Scholar]
- 18.Scapigliati A, Zace D, Matsuyama T, Pisapia L, Saviani M, Semeraro F, Ristagno G, Laurenti P, Bray JE, Greif R, et al. Community Initiatives to Promote Basic Life Support Implementation-A Scoping Review. J Clin Med. 2021;10. doi: 10.3390/jcm10245719 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Plant N, Taylor K. How best to teach CPR to schoolchildren: a systematic review. Resuscitation. 2013;84:415–421. doi: 10.1016/j.resuscitation.2012.12.008 [DOI] [PubMed] [Google Scholar]
- 20.Hazinski MF, Markenson D, Neish S, Gerardi M, Hootman J, Nichol G, Taras H, Hickey R, O'Connor R, Potts J, et al. Response to cardiac arrest and selected life-threatening medical emergencies: the medical emergency response plan for schools--a statement for healthcare providers, policymakers, school administrators, and community leaders. Ann Emerg Med. 2004;43:83–99. doi: 10.1016/j.annemergmed.2003.11.001 [DOI] [PubMed] [Google Scholar]
- 21.Cave DM, Aufderheide TP, Beeson J, Ellison A, Gregory A, Hazinski MF, Hiratzka LF, Lurie KG, Morrison LJ, Mosesso VN Jr., et al. Importance and implementation of training in cardiopulmonary resuscitation and automated external defibrillation in schools: a science advisory from the American Heart Association. Circulation. 2011;123:691–706. doi: 10.1161/CIR.0b013e31820b5328 [DOI] [PubMed] [Google Scholar]
- 22.Nikolaou N, Dainty KN, Couper K, Morley P, Tijssen J, Vaillancourt C. A systematic review and meta-analysis of the effect of dispatcher-assisted CPR on outcomes from sudden cardiac arrest in adults and children. Resuscitation. 2019;138:82–105. doi: 10.1016/j.resuscitation.2019.02.035 [DOI] [PubMed] [Google Scholar]
- 23.Kitamura T, Iwami T, Kawamura T, Nagao K, Tanaka H, Nadkarni VM, Berg RA, Hiraide A. Conventional and chest-compression-only cardiopulmonary resuscitation by bystanders for children who have out-of-hospital cardiac arrests: a prospective, nationwide, population-based cohort study. Lancet. 2010;375:1347–1354. doi: 10.1016/s0140-6736(10)60064-5 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.