Abstract
Background:
Prompt initiation of bystander cardiopulmonary resuscitation (CPR) is critical to survival for out-of-hospital cardiac arrest (OHCA). However, the association between delays in bystander CPR and OHCA survival is poorly understood.
Methods:
In this observational study using a nationally representative U.S. registry, we identified patients who received bystander CPR from a layperson for a witnessed OHCA from 2013 to 2021. Hierarchical logistic regression was used to estimate the association between time to CPR (<1 minute vs 2–3, 4–5, 6–7, 8–9, and ≥10-minute intervals) and survival to hospital discharge and favorable neurologic survival (survival to discharge with Cerebral Performance Category of 1 or 2 [i.e., without severe neurological disability]).
Results:
Of 78,048 patients with a witnessed OHCA treated with bystander CPR, the mean age was 63.5 ± 15.7 years and 25, 197 (32.3%) were women. The median (IQR) time to bystander CPR was 2 (1, 5) minutes, with 10% of patients having a ≥10-minute delay before initiation of CPR. Overall, 15,000 (19.2%) patients survived to hospital discharge and 13,159 (16.9%) had favorable neurologic survival. There was a graded inverse relationship between time to bystander CPR and survival to hospital discharge (p for trend <0.001). Compared with patients who received CPR within 1 minute, those with a time to CPR of 2–3 minutes were 9% less likely to survive to discharge (adjusted OR 0.91 [95% CI 0.87–0.95]) and those with a time to CPR 4–5 minutes were 27% less likely to survive (adjusted OR 0.73 [95% CI 0.68–0.77]). A similar graded inverse relationship was found between time to bystander CPR and favorable neurologic survival (p for trend <0.001).
Conclusions:
Among patients with witnessed OHCA, there was a dose-response relationship between delays in bystander initiation of CPR and lower survival rates.
Keywords: cardiac arrest, cardiopulmonary resuscitation, bystander, survival
INTRODUCTION
Bystander cardiopulmonary resuscitation (CPR) is a crucial link in the chain of survival for out-of-hospital cardiac arrest (OHCA).1 Bystander CPR maintains blood flow to vital organs, increases the probability of successful defibrillation for shockable rhythms, and is associated with a higher likelihood of survival from OHCA.2–6 Efforts to optimize bystander CPR are paramount for improving OHCA survival, which remains low at ~10%.7
While national efforts have largely focused on increasing overall rates of bystander CPR, less is known about when bystander CPR is begun after OHCA and the association between delays in bystander CPR initiation and survival. Previous reports on time to CPR have either been small, single-center studies conducted when CPR training was not commonplace8–11 or larger studies that were not restricted to CPR initiated by bystanders.12, 13 Since bystander-initiated CPR is one of the most effective interventions to improve OHCA survival, understanding the association between even small to modest delays in bystander initiation of CPR and survival would inform ongoing public health and training efforts to maximize the benefit of this intervention.
To address this gap in knowledge, we leveraged the Cardiac Arrest Registry to Enhance Survival (CARES), the largest contemporary OHCA surveillance registry in the U.S. We focused our cohort on witnessed OHCAs to ensure a reliable estimate of the time of cardiac arrest onset and examined the association between delays in CPR initiation and survival outcomes.
METHODS
The data that support the findings of this study are available from the corresponding author upon request and approval by CARES.
Data source
CARES is a prospective, multicenter, observational registry of OHCA in the U.S. Established by Emory University and the Centers for Disease Control and Prevention, CARES currently represents a catchment area of 175 million people (53% of the U.S. population). The program includes 30 state-based registries, the District of Columbia, and sites in 16 additional states, with over 2,300 emergency medical service (EMS) agencies participating nationwide. The rationale and methods for CARES have been previously described.14, 15 Briefly, all patients with a confirmed non-traumatic OHCA, defined as apnea and pulselessness in whom resuscitation is attempted, are identified through review of 9-1-1 dispatch centers, EMS agencies, and receiving hospitals. Data are obtained on each patient with OHCA using standardized international Utstein definitions16 and are reviewed for completeness and internal consistency by CARES analysts.15. This study was performed using de-identified CARES data and was granted a waiver of informed consent by the Saint Luke’s Hospital Institutional Review Board.
Study cohort
Between January 1, 2013, and December 31, 2021, we identified 736,066 patients with OHCA within CARES (Figure S1 In the Supplementary Appendix). To restrict the study population to adults with OHCA occurring outside of a healthcare facility, we excluded patients <18 years old (n=18,759) and OHCAs at a nursing home or healthcare facility (n=95,757). As the focus of the study was on time to bystander CPR, we excluded arrests that were witnessed by 9-1-1 responders (n=90,337) and those without bystander CPR (n=120,821). We further excluded patients with unwitnessed OHCA (n=291,685) since the time of cardiac arrest onset would be unknown. Additionally, we excluded patients with missing data on time of cardiac arrest or CPR initiation, as these are needed to calculate time to CPR (n=29,897), a negative time to CPR (n=205), or a time to CPR greater than 30-minutes (n=3,230). Finally, we excluded patients whose cardiac arrest was due to drowning, electrocution, or exsanguination (n=6,219) and those with missing data on witnessed status of cardiac arrest (n=32) and survival (n=1,076). Our final cohort comprised 78,048 patients with witnessed OHCA who received bystander CPR.
Independent variable and outcomes
The independent variable was time to first bystander CPR, in which bystander was defined as CPR administered by any layperson (family member, stranger, or medical provider) who was not part of the emergency 9-1-1 response. Time to CPR was calculated as the difference between the time of CPR initiation and the time of the 9-1-1 call at the dispatch center. Time of CPR initiation was obtained from bystander self-report by responding EMS personnel, and time of 9-1-1 call at the dispatch center was used as a surrogate for the estimated time of cardiac arrest since all OHCAs were witnessed. The primary outcome was survival to hospital discharge and the secondary outcome was favorable neurologic survival (survival with a discharge Cerebral Performance Category score [range: 1–5] of 1 or 2, where 1 denotes no to mild neurological disability and 2 denotes moderate disability).
Statistical analysis
Baseline characteristics and survival outcomes of patients in the study cohort were compared to the 29,897 patients who were excluded due to missing data to calculate time to CPR to assess potential bias in their exclusion. Given the large sample size, standardized differences were used, with a threshold of ≥10% denoting significant differences.17 Standardized differences compare differences in a baseline characteristic between two groups in units of the pooled standard deviation and are not influenced by sample size. We then examined the distribution of time to bystander CPR using median, interquartile range, 90th percentile, and 95th percentile. For descriptive purposes, we categorized patients into quartiles of time to bystander CPR and compared baseline characteristics across quartiles using one-way analysis of variance for continuous variables or χ2 analysis for categorical variables.
To examine the association between time to bystander CPR and survival to discharge, we constructed a two-level multivariable hierarchical logistic regression model, with EMS agency modeled as a random effect to account for clustering of patient outcomes within site. To avoid assumptions of linearity and make our results prognostically interpretable, time to bystander CPR was categorized into discrete 2-minute intervals (0–1, 2–3, 4–5, 6–7, 8–9 and ≥10 minutes), with 0–1 minutes as the reference category. Besides time to bystander CPR, the model adjusted for age, sex, race, arrest location (public versus home), presumed arrest etiology (cardiac versus non-cardiac), whether a bystander applied an automated external defibrillator (AED), time to EMS arrival (calculated as the difference between the time of EMS arrival at the patient’s side and the time the 9-1-1 call was received at the dispatch center) and urbanicity of the census tract where the arrest occurred (urban area, ≥50,000 residents; urban cluster, ≥2,500 to <50,000 residents; rural, <2,500 residents)18 as fixed effects. The decision was made a priori to not adjust for initially detected cardiac arrest rhythm (shockable versus non-shockable) because this variable may be a mediator, rather than a confounder, of the association between time to first CPR and survival (i.e., a longer time to CPR would increase the likelihood that the initially detected rhythm was pulseless electrical activity or asystole). If initial cardiac rhythm were a mediator of the association between time to CPR and survival, adjusting for initial cardiac rhythm could potentially bias the association between time to CPR and survival towards the null. The Cochrane-Armitage test for trend was used to examine whether there was a graded relationship between time to bystander CPR and survival to discharge. To ensure the robustness of our analyses, we conducted a sensitivity analysis to also adjust for initial cardiac arrest rhythm in this model.
To determine if the association between time to bystander CPR and survival to discharge differed by location of arrest (as there may be higher CPR quality and/or more potential rescuers in public locations), we repeated the analyses above separately for OHCAs occurring at home and in public. All above analyses were repeated for the outcome of favorable neurological survival.
A two-tailed P-value of <0.05 was considered statistically significant. All analyses were performed with SAS software, version 9.4 (SAS Institute, Cary, N.C.).
RESULTS
Overall, there were no differences in demographic or cardiac arrest characteristics between patients in the study cohort and those excluded due to missing time data to calculate time to CPR (Table S1). Patients with missing data on time to bystander CPR data had similar rates of survival to discharge (18.6% versus 19.2%, standardized difference 1.5%) but modestly lower rates of favorable neurologic survival (15.3% vs 16.9%, standardized difference 12.6%).
The median time to layperson bystander CPR was 2 minutes (interquartile range: 1 to 5 minutes) (Figure 1). However, 10% of patients had a time to bystander CPR of ≥10 minutes. For OHCAs at home, median time to bystander CPR was 2 minutes (interquartile range: 1 to 5 minutes), whereas it was 1 minute (interquartile range: 0 to 4 minutes) for OHCAs occurring in public locations (Figure S2).
Figure 1. Time to initiation of bystander CPR for witnessed OHCAs.
Time to CPR is shown for overall cohort. Abbreviations: CPR, cardiopulmonary resuscitation; OHCA, out-of-hospital cardiac arrest.
Baseline and arrest characteristics of the study cohort, stratified by quartile of time to bystander CPR, are presented in Table 1. Mean age was 63.5 ± 15.7 years and 32.3% were women. Of those with known race and ethnicity, 72.4% were White, 17.2% were Black, and 7.1% were Hispanic. Approximately 1 in 4 OHCAs occurred in public locations, 88.1% were of presumed cardiac etiology, and 37.4% had an initially detected shockable rhythm. Patients in the quartile with the longest time to bystander CPR (≥5 minutes) were more likely to be female, Black or Hispanic, have their OHCA at home, and an initially detected rhythm that was non-shockable.
Table 1. Patient characteristics by time to bystander CPR.
Data are provided for the overall cohort and by quartiles of time to CPR.
| Entire Cohort | Quartile 1 (0 min) |
Quartile 2 (1 min) |
Quartile 3 (2 to 4 min) |
Quartile 4 (≥5 min) |
P | |
|---|---|---|---|---|---|---|
| N = 78048 | N = 17275 | N= 16592 | N= 21789 | N = 22392 | ||
| Time to CPR, min (IQR) | 2.0 (1.0, 5.0) | 0.0 (0.0, 0.0) | 1.0 (1.0, 1.0) | 2.0 (2.0, 3.0) | 8.0 (5.0, 11.0) | <0.001 |
| Demographics | ||||||
| Age, years | 63.5 ± 15.7 | 63.3 ± 15.7 | 63.6 ± 15.5 | 63.4 ± 15.6 | 63.7 ± 16.0 | 0.027 |
| Female sex | 25197 (32.3%) | 5449 (31.5%) | 5244 (31.6%) | 6885 (31.6%) | 7619 (34.0%) | <0.001 |
| Race/ethnicity | ||||||
| White | 45781 (72.4%) | 10295 (74.7%) | 9943 (74.0%) | 12647 (72.6%) | 12896 (69.4%) | <0.001 |
| Black | 10868 (17.2%) | 2317 (16.8%) | 2170 (16.2%) | 2893 (16.6%) | 3488 (18.8%) | <0.001 |
| Asian/Pacific Islander | 1534 (2.4%) | 257 (1.9%) | 373 (2.8%) | 429 (2.5%) | 475 (2.6%) | <0.001 |
| Native American | 597 (0.9%) | 119 (0.9%) | 144 (1.1%) | 160 (0.9%) | 174 (0.9%) | 0.33 |
| Hispanic | 4516 (7.1%) | 814 (5.9%) | 814 (6.1%) | 1316 (7.6%) | 1572 (8.5%) | <0.001 |
| Unknown/Missing | 14817 | 3485 | 3161 | 4363 | 3808 | |
| Arrest characteristics | ||||||
| Arrest location | <0.001 | |||||
| Home | 58888 (75.5%) | 12296 (71.2%) | 11599 (69.9%) | 16568 (76.0%) | 18425 (82.3%) | |
| Public | 19159 (24.5%) | 4979 (28.8%) | 4993 (30.1%) | 5220 (24.0%) | 3967 (17.7%) | |
| Presumed arrest etiology | <0.001 | |||||
| Cardiac | 68764 (88.1%) | 15064 (87.2%) | 14666 (88.4%) | 19353 (88.8%) | 19681 (87.9%) | |
| Respiratory | 6107 (7.8%) | 1496 (8.7%) | 1303 (7.9%) | 1590 (7.3%) | 1718 (7.7%) | |
| Drug overdose | 2086 (2.7%) | 447 (2.6%) | 392 (2.4%) | 557 (2.6%) | 690 (3.1%) | |
| Other | 1091 (1.4%) | 268 (1.6%) | 231 (1.4%) | 289 (1.3%) | 303 (1.4%) | |
| Initial cardiac arrest rhythm | <0.001 | |||||
| Shockable | 29169 (37.4%) | 6394 (37.0%) | 6808 (41.0%) | 8644 (39.7%) | 7323 (32.7%) | |
| Non-shockable | 48869 (62.6%) | 10878 (63.0%) | 9782 (59.0%) | 13143 (60.3%) | 15066 (67.3%) | |
| Bystander AED application | 4576 (5.9%) | 1460 (8.5%) | 1357 (8.2%) | 1114 (5.1%) | 645 (2.9%) | <0.001 |
| Time to EMS arrival | ||||||
| Median (IQR) | 9.0 (7.0, 12.0) | 9.0 (7.0, 13.0) | 9.0 (7.0, 12.0) | 9.0 (7.0, 12.0) | 10.0 (7.0, 13.0) | <0.001 |
| Missing | 9802 | 2605 | 1960 | 2514 | 2723 | |
| Urbanicity | <0.001 | |||||
| Urbanized Area | 45168 (75.6%) | 9647 (71.5%) | 9651 (76.5%) | 12767 (76.7%) | 13103 (77.0%) | |
| Urban Cluster | 4114 (6.9%) | 1056 (7.8%) | 920 (7.3%) | 1133 (6.8%) | 1005 (5.9%) | |
| Rural Area | 9858 (16.5%) | 2589 (19.2%) | 1938 (15.4%) | 2604 (15.6%) | 2727 (16.0%) | |
| Undefined* | 623 (1.0%) | 196 (1.5%) | 105 (0.8%) | 136 (0.8%) | 186 (1.1%) | |
| Missing | 18285 | 3787 | 3978 | 5149 | 5371 |
Indicates arrest that was unable to be linked to census tract-level data
Abbreviations: CPR, cardiopulmonary resuscitation; AED, automated external defibrillator; SD, standard deviation; IQR, interquartile range.
Time to CPR and survival outcomes.
A total of 15,000 (19.2%) patients survived to hospital discharge and 13,159 (16.9%) had favorable neurologic survival. Rates of both survival outcomes by time to bystander CPR are presented in Figure 2. When bystander CPR was initiated within 1 minute of cardiac arrest, 22.4% survived to hospital discharge. However, survival decreased in a graded fashion to 10.5% when time to bystander CPR was 10 minutes or greater. Similarly, rates of favorable neurologic survival were 19.9% when bystander CPR was initiated within 1 minute and decreased to 8.8% when time to CPR was 10 minutes or greater.
Figure 2. Survival outcomes by time to bystander CPR.
Rates of survival to discharge (Panel A) and favorable neurological survival (Panel B) are shown by time to bystander CPR. Abbreviations: CPR, cardiopulmonary resuscitation.
After adjusting for patient and cardiac arrest characteristics, there was a graded inverse relationship (p for trend <0.001) between time to bystander CPR and survival to hospital discharge (Table 2). Compared to patients in whom bystander CPR was initiated within the first minute of cardiac arrest, those with a time to CPR of 2 to 3 minutes were 9% less likely to survive to discharge (adjusted OR: 0.91 [95% CI 0.85–0.95], p=0.001) whereas those with time to CPR of 4 to 5 minutes were 27% less likely to survive (adjusted OR: 0.73 [95% CI 0.68–0.77], p<0.001) and those with a time to CPR of 10 minutes of greater were 49% less likely to survive (adjusted OR: 0.51 [95% CI 0.47–0.55], p<0.001). This relationship remained after including initial cardiac arrest rhythm in the model (Table S2). When examined separately for OHCAs occurring at home and in public locations, we found a similar graded inverse relationship between time to bystander CPR and survival to discharge (see Table 2).
Table 2.
Association between time to bystander CPR and survival to hospital discharge.
| Time to First CPR (mins) | Unadjusted Rate | Adjusted OR* (95% CI) | P-value | P for Trend |
|---|---|---|---|---|
| Entire study cohort | ||||
| 0–1 | 22.4% | Reference | Reference | <0.001 |
| 2–3 | 20.5% | 0.91 (0.87–0.95) | <0.001 | |
| 4–5 | 16.4% | 0.73 (0.68–0.77) | <0.001 | |
| 6–7 | 16.2% | 0.73 (0.67–0.80) | <0.001 | |
| 8–9 | 15.0% | 0.68 (0.61–0.76) | <0.001 | |
| 10+ | 10.5% | 0.51 (0.47–0.55) | <0.001 | |
| Home location | ||||
| 0–1 | 15.4% | Reference | Reference | <0.001 |
| 2–3 | 15.4% | 0.94 (0.88–1.00) | 0.047 | |
| 4–5 | 13.3% | 0.80 (0.74–0.86) | <0.001 | |
| 6–7 | 13.3% | 0.79 (0.71–0.87) | <0.001 | |
| 8–9 | 12.4% | 0.73 (0.64–0.83) | <0.001 | |
| 10+ | 8.8% | 0.54 (0.49–0.60) | <0.001 | |
| Public location | ||||
| 0–1 | 39.1% | Reference | Reference | <0.001 |
| 2–3 | 36.1% | 0.90 (0.83, 0.97) | 0.006 | |
| 4–5 | 28.4% | 0.64 (0.58, 0.71) | <0.001 | |
| 6–7 | 28.8% | 0.66 (0.57, 0.78) | <0.001 | |
| 8–9 | 26.8% | 0.61 (0.50, 0.75) | <0.001 | |
| 10+ | 20.3% | 0.45 (0.39, 0.53) | <0.001 |
Models adjusted for age, sex, race, arrest location (public versus home), cardiac arrest etiology (cardiac versus non-cardiac), and urbanicity.
Abbreviations: CPR, cardiopulmonary resuscitation; OR, odds ratio; CI, confidence interval.
There was also a graded inverse relationship (p for trend <0.001) between time to bystander CPR and favorable neurologic survival (Table 3). Compared with patients in whom bystander CPR was initiated within the first minute, those with a time to CPR of 2 to 3 minutes were 10% less likely to have favorable neurologic survival (adjusted OR: 0.90 [95% CI 0.85–0.95], p<0.001), those with a time to CPR of 4 to 5 minutes were 28% less likely to have favorable neurologic survival (adjusted OR: 0.72 [95% CI 0.68–0.77], p<0.001), and those with a time to CPR of 10 minutes or greater were 51% less likely to achieve this outcome (adjusted OR: 0.49 [95% CI 0.45–0.54], p<0.001). As with the outcome of survival to discharge, additional adjustment for initial cardiac arrest rhythm did not meaningfully change this inverse relationship (Table S3) and similar findings were obtained when analyses were conducted separately for OHCAs at home and in public locations (see Table 3).
Table 3.
Association between time to bystander CPR and survival to favorable neurologic survival.
| Time to First CPR (mins) | Unadjusted Rate | Adjusted OR* (95% CI) | P-value | P for Trend |
|---|---|---|---|---|
| Entire study cohort | ||||
| 0–1 | 19.9% | Reference | Reference | <0.001 |
| 2–3 | 18.0% | 0.90 (0.85–0.95) | <0.001 | |
| 4–5 | 14.2% | 0.72 (0.68–0.77) | <0.001 | |
| 6–7 | 13.8% | 0.71 (0.65–0.78) | <0.001 | |
| 8–9 | 12.4% | 0.65 (0.58–0.73) | <0.001 | |
| 10+ | 8.8% | 0.49 (0.45–0.54) | <0.001 | |
| Home location | ||||
| 0–1 | 13.0% | Reference | Reference | <0.001 |
| 2–3 | 13.0% | 0.93 (0.87, 1.00) | 0.049 | |
| 4–5 | 11.2% | 0.80 (0.74, 0.87) | <0.001 | |
| 6–7 | 11.0% | 0.78 (0.69, 0.87) | <0.001 | |
| 8–9 | 9.8% | 0.69 (0.60–0.80) | <0.001 | |
| 10+ | 7.2% | 0.54 (0.49–0.60) | <0.001 | |
| Public location | ||||
| 0–1 | 36.5% | Reference | Reference | <0.001 |
| 2–3 | 33.2% | 0.88 (0.81–0.95) | 0.002 | |
| 4–5 | 25.9% | 0.63 (0.56–0.70) | <0.001 | |
| 6–7 | 25.6% | 0.63 (0.53–0.74) | <0.001 | |
| 8–9 | 24.6% | 0.61 (0.49–0.75) | <0.001 | |
| 10+ | 17.8% | 0.43 (0.37–0.50) | <0.001 |
Models adjusted for age, sex, race, arrest location (public versus residential), cardiac arrest etiology (cardiac versus non-cardiac), and urbanicity.
Abbreviations: CPR, cardiopulmonary resuscitation; OR, odds ratio; CI, confidence interval.
DISCUSSION
While the importance of bystander CPR in improving survival for OHCA is well recognized, the consequences of delays in its initiation are not well understood. Using a nationally representative OHCA registry in the U.S., we found that, although the median time to bystander CPR for a witnessed OHCA was 2 minutes, 1 in 4 patients did not have bystander CPR initiated until at least 5 minutes after onset of their cardiac arrest. There was an inverse dose-response relationship between any delay in initiating bystander CPR after the first minute and survival outcomes. A delay in initiating CPR of 2 to 3 minutes was associated with an 9% lower likelihood of overall survival to discharge and a 10% lower likelihood of favorable neurologic survival whereas a 4 to 5-minute delay was associated with a 27% lower likelihood of survival to discharge and 28% lower likelihood of favorable neurologic survival. This inverse graded relationship was consistent for witnessed OHCAs occurring at home and in public locations, where the number of potential bystanders may differ. Collectively, our findings support efforts to reduce delays in initiating bystander CPR to maximize the effectiveness of this potentially life-saving intervention.
Studies describing the association of time to bystander CPR and survival outcomes are limited. Earlier studies reported a 2.3% absolute reduction9 and 10% lower odds of overall survival11 with each 1-minute delay in initiating bystander CPR. These reports, however, were limited by small sample sizes and lack of information on cardiac arrest characteristics, examined only OHCAs with an initial rhythm of ventricular fibrillation, and were conducted before the year 2000 during a time when CPR training was not common and may not reflect current bystanders.8–11 In the past decade, several larger studies found a significant association between time to first CPR and survival but these studies did not restrict the OHCA cohort to those with only bystander CPR (i.e., included CPR initiated by first responders or EMS personnel),12, 13 sometimes included unwitnessed arrests (when time of cardiac arrest may not be known),13 and/or dichotomized time to CPR (e.g., ≤2 vs. >2 minutes).19 Moreover, these studies were conducted outside the U.S., where there are differences in EMS organization, CPR training, and geography. We extend this prior literature by restricting our cohort to only patients with witnessed OHCA who were treated with layperson bystander CPR and quantified the inverse graded association between time to bystander CPR and survival outcomes. We did this for witnessed OHCAs overall and separately for OHCAs occurring at home and in public locations.
Our findings inform current efforts to increase rates of bystander CPR in the U.S. While the finding of longer times to initiation of bystander CPR and lower survival is not surprising, few studies have quantified the impact of different delays to bystander initiation of CPR on OHCA survival. As 25% of patients with a witnessed OHCA did not have bystander CPR initiated within 5 minutes and 10% did not have bystander CPR initiated within 10 minutes, this suggests that strategies to reduce delays in time to first bystander CPR, coupled with current efforts to increase overall rates of bystander CPR, will be critical to ongoing efforts to improve OHCA survival.
The reasons for delays in initiating bystander CPR are not captured in CARES. In 2010, the American Heart Association introduced significant changes to adult Basic Life Support guidelines to reduce delays in starting bystander CPR. These included de-emphasizing pulse checks to confirm the presence of an OHCA before initiating CPR in a non-responsive, non-breathing adult.1 Although our study cohort included patients after the 2010 guidelines, it is possible laypersons may not have received updated CPR training. Delays in time to CPR could also be because the first bystanders for an OHCA victim did not have CPR training, resulting in a time lag until a trained bystander arrived on scene and initiated CPR. This scenario would be more likely for OHCAs in a public location, but we found similar delays to bystander CPR for OHCAs at home. Additionally, dispatcher-assisted CPR increases the likelihood of bystander CPR and reduces the time to first CPR.20 However, there may be significant variability in how rapidly dispatchers recognize a cardiac arrest and provide telephone instructions to facilitate bystander CPR. For laypersons who initiate bystander CPR with a dispatcher’s assistance, shortening the time from activation of 9-1-1 to delivery of CPR instructions is a potentially significant opportunity to reduce delays in time to bystander CPR. Lastly, Basic Life Support training for CPR should re-emphasize the time-sensitive nature of OHCA where every minute without CPR reduces the likelihood of survival.
Our study has some limitations. First, our models were adjusted for measured covariates in CARES. Unmeasured factors, such as the quality of CPR, may confound the association between time to bystander CPR and survival outcomes. Second, time to CPR was derived from the time of the 9-1-1 call (as a surrogate for the estimated time of cardiac arrest) and bystander reporting of initiation of CPR. Any misclassification of either time component would have yielded inaccurate times to bystander CPR, although any misclassification would likely have been nondifferential and biased our estimates for the association between time to bystander CPR and survival toward the null. Third, we excluded approximately 25% of eligible patients with missing times to bystander CPR. However, excluded patients had similar characteristics as those in the study cohort. Fourth, we did not have information on the reasons why delays in initiating bystander CPR occurred. Fifth, whether dispatch-assisted CPR is performed is not routinely collected in CARES. Similarly, CARES does not collect data on dispatcher times to recognition of OHCA and to providing CPR instructions to bystanders. Future studies with a focus on these dispatcher times may provide insights as to whether there are systems-related opportunities to reduce time to bystander CPR. Lastly, although CARES is the largest U.S. cardiac arrest registry and represents a catchment area of approximately 53% of the population, our findings may not be generalizable to populations not represented by CARES.
In conclusion, for patients with witnessed OHCA who received bystander CPR, delays in initiating CPR was not uncommon, and there was an inverse graded relationship between time to bystander CPR and lower survival. Besides increasing overall rates of bystander CPR, strategies to reduce delays in initiating bystander CPR should be prioritized to maximize the benefit of this critical intervention in OHCA.
Supplementary Material
WHAT IS KNOWN
Bystander cardiopulmonary resuscitation (CPR) is critical for survival in out-of-hospital cardiac arrest (OHCA).
Less is known about when bystander CPR is begun in relation to the OHCA and the impact of delays in bystander CPR and survival outcomes.
WHAT THE STUDY ADDS
Among witnessed arrests in a nationally representative OHCA registry in the U.S., delays in initiating bystander CPR were not uncommon.
There was an inverse graded relationship between any delay in initiating bystander CPR after the first minute of OHCA and survival outcomes.
These findings suggest that besides increasing overall rates of bystander CPR, strategies to reduce delays in initiating bystander CPR should be prioritized.
Sources of funding.
Drs. Nguyen and Uzendu are currently supported by the National Heart, Blood and Lung Institutes of Health under Award Number T32HL110837. Dr. Chan receives research funding from the National Heart, Lung, and Blood Institute (NHLBI) R01HL160734. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
Disclosures.
Dr. Chan receives funding from the American Heart Association and is a consultant for Optum Rx.
Dr. McNally is the Executive Director of the CARES, which receives funding from the American Heart Association and American Red Cross.
Dr. Spertus reports providing consultative services on patient-reported outcomes and evidence valuation to Alnylam, AstraZeneca, Bayer, Merck, Janssen, Bristol Meyers Squibb, Edwards, Kineksia, 4DT Medical, Terumo, Cytokinetics, Imbria, and United Healthcare. He holds research grants from Bristol Meyers Squibb, Abbott Vascular and Janssen. He owns the copyright to the Seattle Angina Questionnaire, Kansas City Cardiomyopathy Questionnaire, and Peripheral Artery Questionnaire and serves on the Board of Directors for Blue Cross Blue Shield of Kansas City.
Drs. Nguyen, Gupta, and Uzendu and Mr. Kennedy do not have any financial disclosures.
ABBREVIATIONS
- CPR
cardiopulmonary resuscitation
- OHCA
out-of-hospital cardiac arrest
- CARES
Cardiac Arrest Registry to Enhance Survival
Footnotes
REFERENCES
- 1.Panchal AR, Bartos JA, Cabañas JG, Donnino MW, Drennan IR, Hirsch KG, Kudenchuk PJ, Kurz MC, Lavonas EJ, Morley PT, et al. Part 3: Adult Basic and Advanced Life Support: 2020 American Heart Association Guidelines for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care. Circulation. 2020;142:S366–S468. doi: 10.1161/CIR.0000000000000916 [DOI] [PubMed] [Google Scholar]
- 2.Stiell IG, Wells GA, Field B, Spaite DW, Nesbitt LP, De Maio VJ, Nichol G, Cousineau D, Blackburn J, Munkley D, et al. Advanced Cardiac Life Support in Out-of-Hospital Cardiac Arrest. https://doi.org/101056/NEJMoa040325. 2004;351:647–656. doi: 10.1056/NEJMOA040325 [DOI] [PubMed] [Google Scholar]
- 3.Hasselqvist-Ax I, Riva G, Herlitz J, Rosenqvist M, Hollenberg J, Nordberg P, Ringh M, Jonsson M, Axelsson C, Lindqvist J, et al. Early Cardiopulmonary Resuscitation in Out-of-Hospital Cardiac Arrest. New England Journal of Medicine. 2015;372:2307–2315. doi: 10.1056/NEJMOA1405796/SUPPL_FILE/NEJMOA1405796_DISCLOSURES.PDF [DOI] [PubMed] [Google Scholar]
- 4.Sasson C, Rogers MAM, 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]
- 5.Patil KD, Halperin HR, Becker LB. Cardiac Arrest. Circ Res. 2015;116:2041–2049. doi: 10.1161/CIRCRESAHA.116.304495 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Ok Ahn K, McNally B, Al-Araji R, Cisneros C, Chan PS. Sex differences in the association between bystander CPR and survival for Out-of-Hospital cardiac arrest. Resuscitation. 2023;182. doi: 10.1016/J.RESUSCITATION.2022.09.016 [DOI] [PubMed] [Google Scholar]
- 7.Chan PS, McNally B, Tang F, Kellermann A. Recent trends in survival from out-of-hospital cardiac arrest in the United States. Circulation. 2014;130:1876–1882. doi: 10.1161/CIRCULATIONAHA.114.009711/-/DC1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Cummins RO, Eisenberg MS, Hallstrom AP, Litwin PE. Survival of out-of-hospital cardiac arrest with early initiation of cardiopulmonary resuscitation. The American Journal of Emergency Medicine. 1985;3:114–119. doi: 10.1016/0735-6757(85)90032-4 [DOI] [PubMed] [Google Scholar]
- 9.Larsen MP, Eisenberg MS, Cummins RO, Hallstrom AP. Predicting survival from out-of-hospital cardiac arrest: A graphic model. Annals of Emergency Medicine. 1993;22:1652–1658. doi: 10.1016/S0196-0644(05)81302-2 [DOI] [PubMed] [Google Scholar]
- 10.Weaver WD, Cobb LA, Hallstrom AP, Fahrenbruch C, Copass MK, Ray R. Factors influencing survival after out-of-hospital cardiac arrest. Journal of the American College of Cardiology. 1986;7:752–757. doi: 10.1016/S0735-1097(86)80332-1 [DOI] [PubMed] [Google Scholar]
- 11.Valenzuela TD, Roe DJ, Cretin S, Spaite DW, Larsen MP. Estimating Effectiveness of Cardiac Arrest Interventions. Circulation. 1997;96:3308–3313. doi: 10.1161/01.CIR.96.10.3308 [DOI] [PubMed] [Google Scholar]
- 12.Hara M, Hayashi K, Hikoso S, Sakata Y, Kitamura T. Different impacts of time from collapse to first cardiopulmonary resuscitation on outcomes after witnessed out-of-hospital cardiac arrest in adults. Circulation: Cardiovascular Quality and Outcomes. 2015;8:277–284. doi: 10.1161/CIRCOUTCOMES.115.001864 [DOI] [PubMed] [Google Scholar]
- 13.Fridman M, Barnes V, Whyman A, Currell A, Bernard S, Walker T, Smith KL. A model of survival following pre-hospital cardiac arrest based on the Victorian Ambulance Cardiac Arrest Register. Resuscitation. 2007;75:311–322. doi: 10.1016/J.RESUSCITATION.2007.05.005 [DOI] [PubMed] [Google Scholar]
- 14.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. Morbidity and Mortality Weekly Report: Surveillance Summaries. 2011;60:1–19. [PubMed] [Google Scholar]
- 15.McNally B, Stokes A, Crouch A, Kellermann AL. CARES: Cardiac Arrest Registry to Enhance Survival. Annals of Emergency Medicine. 2009;54:674–683.e2. doi: 10.1016/j.annemergmed.2009.03.018 [DOI] [PubMed] [Google Scholar]
- 16.Jacobs I, Nadkarni V, null null, null null, Bahr J, Berg RA, Billi JE, Bossaert L, Cassan P, Coovadia A, et al. Cardiac Arrest and Cardiopulmonary Resuscitation Outcome Reports. Circulation. 2004;110:3385–3397. doi: 10.1161/01.CIR.0000147236.85306.15 [DOI] [PubMed] [Google Scholar]
- 17.Austin PC. Using the Standardized Difference to Compare the Prevalence of a Binary Variable Between Two Groups in Observational Research. Communications in Statistics - Simulation and Computation. 2009;38:1228–1234. doi: 10.1080/03610910902859574 [DOI] [Google Scholar]
- 18.United States Census Bureau. 2010 Census Urban and Rural Classification and Urban Area Criteria. Published 2010. https://www.census.gov/programs-surveys/geography/guidance/geo-areas/urban-rural/2010-urban-rural.html [Google Scholar]
- 19.Holmberg M, Holmberg S, Herlitz J. Factors modifying the effect of bystander cardiopulmonary resuscitation on survival in out-of-hospital cardiac arrest patients in Sweden. European Heart Journal. 2001;22:511–519. doi: 10.1053/euhj.2000.2421 [DOI] [PubMed] [Google Scholar]
- 20.Painter I, Chavez DE, Ike BR, Yip MP, Tu SP, Bradley SM, Rea TD, Meischke H. Changes to DA-CPR instructions: can we reduce time to first compression and improve quality of bystander CPR? Resuscitation. 2014;85:1169–1173. doi: 10.1016/J.RESUSCITATION.2014.05.015 [DOI] [PubMed] [Google Scholar]
- 22.Tsao CW, Aday AW, Almarzooq ZI, Alonso A, Beaton AZ, Bittencourt MS, Boehme AK, Buxton AE, Carson AP, Commodore-Mensah Y, et al. Heart Disease and Stroke Statistics—2022 Update: A Report From the American Heart Association. Circulation. 2022;145(8):e153–e639. doi: 10.1161/CIR.0000000000001052 [DOI] [PubMed] [Google Scholar]
- 21.2022 CARES Annual Report. Accessed August 7, 2023. https://mycares.net/sitepages/uploads/2023/2022_flipbook/index.html?page=1
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.