Abstract
Background
Following the implementation of the HeartRescue project, with interventions in the community, emergency medical services, and hospitals to improve care and outcomes for out‐of‐hospital cardiac arrests (OHCA) in North Carolina, improved bystander and first responder treatments as well as survival were observed. This study aimed to determine whether these improvements were consistent across Black versus White individuals.
Methods and Results
Using the Cardiac Arrest Registry to Enhance Survival (CARES), we identified OHCA from 16 counties in North Carolina (population 3 million) from 2010 to 2014. Temporal changes in interventions and outcomes were assessed using multilevel multivariable logistic regression, adjusted for patient and socioeconomic neighborhood‐level factors. Of 7091 patients with OHCA, 36.5% were Black and 63.5% were White. Black patients were younger, more females, had more unwitnessed arrests and non‐shockable rhythm (Black: 81.0%; White: 75.4%). From 2010 to 2014, the adjusted probabilities of bystander cardiopulmonary resuscitation (CPR) went from 38.5% to 51.2% in White, P<0.001; and 36.9% to 45.6% in Black, P=0.002, and first‐responder defibrillation went from 13.2% to 17.2% in White, P=0.002; and 14.7% to 17.3% in Black, P=0.16. From 2010 to 2014, survival to discharge only increased in White (8.0% to 11.4%, P=0.004; Black 8.9% to 9.5%, P=0.60), though, in shockable patients the probability of survival to discharge went from 24.8% to 34.6% in White, P=0.02; and 21.7% to 29.0% in Black, P=0. 10.
Conclusions
After the HeartRescue program, bystander CPR and first‐responder defibrillation increased in both patient groups; however, survival only increased significantly for White patients.
Keywords: intervention, OHCA, race, survival
Subject Categories: Cardiopulmonary Arrest, Cardiopulmonary Resuscitation and Emergency Cardiac Care, Epidemiology, Race and Ethnicity, Mortality/Survival
Nonstandard Abbreviations and Acronyms
- CARES
Cardiac Arrest Registry to Enhance Survival
- CPC
Cerebral Performance Category
- IOM
Institute of Medicine
- OHCA
out‐of‐hospital cardiac arrest
- PP
predicted probabilities
- RACE‐CARS
Regional Approach to Cardiovascular Emergencies Cardiac Arrest Resuscitation System
- VIF
variance inflation factors
Clinical Perspective
What Is New?
Following the implementation of the HeartRescue project with interventions in the community, emergency medical services, and hospitals to improve care and outcomes for out‐of‐hospital cardiac arrests in North Carolina from 2010 to 2014, bystander cardiopulmonary resuscitation and first‐responder defibrillation increased in both Black and White patients, though overall survival only increased significantly in White patients.
Looking at patients with an initial shockable rhythm, improved survival was observed for both Black and White patients, indicating that the observed lower frequency of initial shockable rhythm among Black patients could be an important contributor.
What Are the Clinical Implications?
The interventions appear to be effective at improving bystander cardiopulmonary resuscitation and early defibrillation irrespective of race, although survival was significantly improved only in White patients. Further research in improving care and outcomes in Black patients is needed.
Out‐of‐hospital cardiac arrest (OHCA) remains a significant health problem with a poor prognosis, affecting ≈700 000 people in North America and Europe annually 1 , 2 of which around 10% survive. 2 Over the years, several initiatives in cardiac arrest management have been implemented worldwide with subsequent improved survival. 3 , 4 , 5 However, overall outcomes remain poor and vary across different patient and cardiac arrest‐related characteristics as highlighted in the recent Institute of Medicine (IOM) report, “Strategies to Improve Cardiac Arrest Survival: A Time to Act.” The IOM report specifically called for studies to improve understanding of differences across race and socioeconomic aspects that will help identify future targets for improving care and outcomes for underprivileged groups. 6 In this context, racial differences are of great importance since previous studies have shown that Black patients are much more likely to suffer a cardiac arrest at a younger age, less likely to receive bystander cardiopulmonary resuscitation (CPR) and defibrillation, and ultimately less likely to survive. 7 , 8 , 9 , 10 Socioeconomic status has earlier been suggested as an explaining factor for racial difference, 11 , 12 but studies on this field are lacking and other factors including biological differences may also be important.
In 2010, North Carolina initiated the North Carolina Regional Approach to Cardiovascular Emergencies Cardiac Arrest Resuscitation System (RACE‐CARS) program as part of the HeartRescue project, 13 and by this implemented multifaceted interventions with population‐based training in CPR and use of automated external defibrillators (AED) as well as dispatch center training in recognizing cardiac arrests. Subsequently, an increase in both bystander and first‐responder interventions (CPR and defibrillation) and survival from OHCA in North Carolina was observed from 2010. 14 However, it remains unknown whether the effects of these interventions in North Carolina from 2010 to 2014 differed between Black and White race. The aim of this study was therefore to assess Black and White differences in bystander CPR, defibrillation, and survival following the implementation of the RACE‐CARS program/HeartRescue project.
Methods
The authors declare that all supporting data are available within the article and its online supplementary files.
Data Source and Setting
This study is based on the CARES (Cardiac Arrest Registry to Enhance Survival) registry, which is a voluntary, prospective clinical registry including all non‐traumatic patients with OHCA where resuscitation has been attempted by a 911 responder in the United States. The registry was established by the Centers for Disease Control and Prevention and Emory University for public health surveillance and continuous quality improvement. 15 , 16 Data are collected from emergency medical service (EMS) agencies and receiving hospitals and afterwards reviewed for completeness and accuracy by a CARES analyst according to the Utstein template. 15 As part of the HeartRescue Project in North Carolina participating EMS agencies received training, quality control, and data feedback.
The included EMS agencies in this study all had a two‐tiered response system with first responders armed with AEDs 17 and the EMS (paramedics). First responders are defined by CARES as “personnel who respond to the medical emergency in an official capacity as part of an organized medical response team, but are not the designated transporter of the patient to the hospital,” 17 and in North Carolina consists of firefighters, police officers and other life‐saving and rescue squads trained to perform basic life support until EMS arrives. Bystanders are defined as other people on the scene and not dispatched by the dispatch centers. 18
The location of the OHCAs was based on the physical address of the OHCA and was geocoded by ArcGIS 10.2 software (ESRI, Redlands, CA). The geocoding assigned a latitude and longitude coordinate to each address. We achieved a 97% geocoding rate, where non‐geocoded records included PO boxes and other non‐physical locations. This process verified the county in which each OHCA occurred. We used census tracts as proxies for neighborhoods, as previously done, 19 since they represent socio‐economically homogeneous groups of 4000 to 7000 people. 20 The 2010 United States Census Summary Files were used to link each geocoded address with neighborhood‐level variables as median household income, percentage with high school diploma, urban (areas of >2500 people) and rural census information etc. 21
The HeartRescue Project in North Carolina
In 2010, North Carolina initiated a statewide multifaceted quality‐improvement program (RACE‐CARS) as a part of the HeartRescue Project. 13 The project protocol is freely available, and the project has previously been described in detail elsewhere. 14 , 22 Overall, the project included interventions for hospital personnel and administration (with establishment of in‐hospital treatment protocols), EMS dispatchers (with training in recognizing of cardiac arrest, provide assisted‐CPR, and implement protocols for transporting of the patients), first responders (with instruction in team‐based high‐quality CPR and AED use), and community members (with CPR and AED training).
Study Population
We identified all adult patients of Black and White race with OHCA of presumed cardiac etiology from 16 counties in North Carolina (population ≈3 million) from the CARES registry with complete registry enrollment during the entire study period (2010–2014) as done previously. 14 Following Utstein guidelines, we excluded EMS‐witnessed cases and cases with “do not resuscitate” orders 18 as well as cases with non‐matched geo‐coding, missing or other race than Black or White. Figure 1 shows the study selection process. Patient race was obtained from the CARES registry where it is reported by the patient, family or healthcare provider, as defined by CARES registry guidelines. 17
Figure 1. Predicted probabilities of bystander and first responder CPR and defibrillation.
Adjusted predicted probabilities with 95% CIs from 2010 to 2014 for bystander CPR (A), bystander defibrillation (B), first responder CPR (C) and first responder defibrillation (D) in Black and White patients. The analyses are adjusted for age, sex, location of arrest, witnessed status, initial heart rhythm and neighborhood factors (income, education and urban/rural status) and includes an interaction between race and calendar year. The interaction is showed as a P value in the Figure. A P value <0.05 is considered statistically significant. CPR indicates cardiopulmonary resuscitation; and OHCA, out‐of‐hospital cardiac arrest.
Outcomes
The outcomes were bystander and first responder initiation of CPR and defibrillation, survival to discharge and survival with favorable neurologic outcome (Cerebral Performance Category [CPC] 1–2), for Black compared with White patients.
Statistical Analysis
Descriptive statistics of overall characteristics and outcomes according to patient race were shown as frequencies and percentages for categorical variables and as medians with 25%–75% percentiles for continuous variables. To simplify interpretation and to provide identifiable cutoffs/thresholds that are more useful for identifying areas for intervention we divided the neighborhood variables: household income and education in subgroups that for income was based on tertiles. Differences between the Black and White patients were tested with Chi‐Square tests for categorical variables and the Kruskall‐Wallis tests for continuous variables. P values <0.05 were considered statistically significant. Only complete case analyses were performed.
Multilevel logistic regression analyses with mixed models were used to examine differences in bystander and first responder interventions as well as survival to discharge and survival with favorable neurological outcome for Black patients versus White patients from 2010 to 2014. We used multilevel logistic regression models to account for patients nested within census tracts. Interactions between patient race and time (year) were included to account for potential temporal differences between the patient groups. The regression models for the association between patient race, bystander and first responder CPR and defibrillation were performed in 3 steps (1) unadjusted, (2) adjusted for patient factors (age and sex) and neighborhood factors (neighborhood income, educational status, urban/rural setting), and (3) adjusted for patient factors, neighborhood factors, and location of arrest, witnessed status, and rhythm. The results from the unadjusted models (1) and the patient and neighborhood adjusted models (2) are presented in the supplemental material. The regression models for the association between patient race and survival status were performed in 4 steps: (1) unadjusted, (2) adjusted for patient factors (age, sex, location of arrest, witnessed status, and rhythm), (3) adjusted for patient and neighborhood factors (neighborhood income, neighborhood educational status, urban/rural setting), and (4) adjusted for patient factors, neighborhood factors and interventions (bystander and first responder CPR and defibrillation). The changes in interventions and outcomes were also examined in a more homogenous population of patients with initial shockable heart rhythm (n=1595 patients). Predicted probabilities (PP and 95% CI) were calculated from the multilevel logistic regression models and reported as percentages (PP×100) to facilitate interpretation of the key findings; and P values were reported for the trend within each patient group (Black and White patients) and for the tests of interactions between patient race and time (temporal change).
Tests of multicollinearity among covariates (namely neighborhood variables) were assessed based on variance inflation factors (VIF), tolerance levels, and condition values; and the results suggested no evidence of collinearity in the models (eg, VIFs≤2.87).
All analyses where performed using SAS version 9.4 (SAS institute, Cary, NC) and Stata version 15.0 (StataCorp, College Station, TX).
Ethics
This study was approved by the Duke University Medical Center Institutional Review Board for analyses and publication of the findings. A waiver of the requirement for written informed consent and Health Insurance Portability and Accountability Act authorization was granted on the basis of (1) using existing central CARES registry data and under existing waiver of HIPAA consent, and (2) using aggregated and limited data.
Results
The patient selection process is shown in Figure S1. Of 7091 patients, 36.5% were Black and 63.5% were White. The overall incidence of OHCA per 100 000 inhabitants per year was 54.7, with corresponding 71.3 patients with OHCA of Black race and 48.3 of White race per 100 000 inhabitants per year. Baseline characteristics are shown in Table. Black patients were younger, more often female, unwitnessed, and had a non‐shockable heart rhythm (Black: 81.0%; White: 75.4%), also in witnessed arrests with bystander CPR (Black: 64.9%; White: 59.8%). The median EMS response time was 8 minutes in both Black and White patients.
Table 1.
Overall Characteristics
|
Black (n=2591) |
White (n=4500) |
P Value |
In total (n=7091) |
Missing data | |
|---|---|---|---|---|---|
| Average background population per year (2010–2014)* | 727 000 | 1 866 000 | … | 2 593 000 | … |
| Incidence per 100 000 inhabitants per year | 71.3 | 48.2 | … | 54.7 | … |
| Patient‐related factors: | |||||
| Median age (Q1–Q3) |
63 (53–75) |
68 (56–79) |
<0.001 |
66 (55–78) |
12 (0.2) |
| Female sex, n (%) |
1171 (45.2) |
1547 (34.4) |
<0.001 |
2718 (38.3) |
1 |
| Cardiac arrest‐related factors: | |||||
| Arrests in private homes, n (%) |
2046 (79.0) |
3578 (79.5) |
0.59 |
5624 (79.3) |
… |
| Witnessed arrests, n (%) |
1080 (41.7) |
2108 (46.9) |
<0.001 |
3188 (45.0) |
1 |
| Who initiated CPR, n (%) | |||||
| Bystander |
1083 (42.7) |
2113 (48.5) |
<0.001 |
3196 (46.4) |
196 (2.8) |
| First responder |
1084 (42.7) |
1695 (38.9) |
<0.001 |
2779 (40.3) |
196 (2.8) |
| EMS |
369 (14.6) |
551 (12.6) |
<0.001 |
920 (13.3) |
196 (2.8) |
| AED application prior to EMS arrival, n (%) | |||||
| AED application by bystander |
121 (5.5) |
161 (3.9) |
0.009 |
282 (4.5) |
784 (11.1) |
| AED application by first responder |
985 (44.5) |
1784 (43.6) |
0.009 |
2769 (43.9) |
784 (11.1) |
| AED use, n (%) | |||||
| AED use by bystander |
62 (2.4) |
98 (2.2) |
0.55 |
160 (2.3) |
3 |
| AED use by first responder |
352 (13.6) |
747 (16.6) |
<0.001 |
1099 (15.5) |
3 |
| AED use in shockable patients, n (%) | |||||
| AED use by bystander |
40 (8.2) |
72 (6.5) |
0.24 |
112 (7.0) |
… |
| AED use by first responder |
236 (48.2) |
524 (47.4) |
0.78 |
760 (47.6) |
… |
| Median EMS response time (Q1–Q3) † |
8 (6–9) |
8 (6–11) |
<0.001 |
8 (6–10) |
391 (5.5) |
| Initial shockable heart rhythm, n (%) |
490 (18.9) |
1105 (24.6) |
<0.001 |
1595 (22.5) |
3 (0.1) |
| Shockable rhythm in witnessed arrests with bystander CPR, n (%) |
170 (35.1) |
454 (40.2) |
0.05 |
624 (38.6) |
2 (0.1) |
| Area‐related factors: | |||||
| Household income, USD | |||||
| Low income (<40 000) |
1247 (48.1) |
1254 (27.9) |
<0.001 |
2501 (35.3) |
… |
| Medium income (40–54 999) |
731 (28.2) |
1505 (33.4) |
<0.001 |
2236 (31.5) |
… |
| High income (≥55 000) |
613 (23.7) |
1741 (38.7) |
<0.001 |
2354 (33.2) |
… |
| Percentage of high school diploma or higher, n (%) | |||||
| Areas of <80% with high school diploma or higher |
1128 (43.5) |
1228 (27.3) |
<0.001 |
2356 (33.2) |
… |
| Areas of 80–90% with high school diploma or higher |
842 (32.5) |
1522 (33.8) |
<0.001 |
2364 (33.3) |
… |
| Areas of >90% with high school diploma or higher |
621 (24.0) |
1750 (38.9) |
<0.001 |
2371 (33.5) |
… |
| Urban area, n (%) |
2375 (91.7) |
3283 (73.0) |
<0.001 |
5658 (79.8) |
… |
| In‐hospital care: ‡ | |||||
| Therapeutic hypothermia in hospital, (%) |
378 (63.6) |
656 (60.9) |
0.26 |
1034 (61.8) |
115 (6.4) |
| Performed coronary angiography, (%) |
118 (27.1) |
310 (35.7) |
0.01 |
428 (32.8) |
483 (27.0) |
| Cardiac stent placed in patients with performed coronary angiography, n (%) |
39 (33.6) |
145 (47.1) |
0.01 |
184 (43.4) |
4 (0.9) |
| Outcomes: | |||||
| ROSC, n (%) |
691 (26.8) |
1317 (29.3) |
0.02 |
2008 (28.4) |
18 (0.3) |
| In shockable arrests, n (%) |
251 (51.3) |
543 (49.1) |
0.42 |
794 (49.8) |
1 (0.1) |
| Survival to discharge, n (%) |
235 (9.1) |
432 (9.7) |
0.45 |
667 (9.5) |
32 (0.5) |
| In shockable arrests, n (%) |
147 (30.1) |
314 (28.6) |
0.54 |
461 (29.1) |
10 (0.6) |
| Survival with favorable neurological outcome (CPC 1–2) |
193 (7.5) |
387 (8.7) |
0.09 |
580 (8.2) |
32 (0.5) |
Q1–Q3=interquartile range. AED indicates automated external defibrillator; CPR, cardiopulmonary resuscitation; EMS, emergency medical services; and ROSC, return of spontaneous circulation.
Average background population from 2010 to 2014, rounded to nearest 1000.
The EMS response time was calculated based on time of receipt of 911‐call at the dispatch center of arrest to time of EMS/ambulance arrival on scene.
Numbers for the in‐hospital factors are based on patients admitted to the hospital (n=1787 patients). Missing data on 17 patients.
Neighborhood Characteristics
Black patients with OHCA were more likely to arrest in neighborhoods that were urban (Black: 91.7%; White: 73.0%), had a lower percentage of residents with high school diplomas (Black: 43.5%; White: 27.3%), and with lower median household income (Black: 48.1%; White: 27.9%) compared with White patients with OHCA (Table).
Cardiopulmonary Resuscitation
Black patients had lower bystander CPR (Black: 42.7%; White: 48.5%), but higher first responder CPR (Black: 42.7%; White: 38.9%) (Table), compared with White patients. From 2010 to 2014, the fully adjusted probability of bystander CPR increased significantly in both patient groups (Black: 36.9% [95% CI 33.2%–40.6%] in 2010 to 45.6% [95% CI 42.2%–49.0%] in 2014, P=0.002; White: 38.5% [95% CI 35.6%–41.4%] in 2010 to 51.2% [95% CI 48.5%–53.9%] in 2014, P<0.001) (Figure 1A), whereas first responder CPR went from 45.7% [95% CI 41.8%–49.6%] in 2010 to 41.9% [95% CI 38.5%–45.3%] in 2014 in Black patients, (P=0.21) and from 43.0% [95% CI 40.4%–45.6%] in 2010 to 38.5% [95% CI 35.9%–41.1%] in 2014 in White patients (P=0.03) (Figure 1C). Though, no significant difference between the 2 groups was observed (P=0.83). Overall the same trends were observed in the crude analysis and when adjusting only for age, sex, and neighborhood factors (Figure S2 and S3), as well as in only shockable patients (Figure 2).
Figure 2. Predicted probabilities of bystander and first responder CPR and defibrillation in patients with shockable heart rhythm.
Adjusted predicted probabilities with 95% CIs from 2010 to 2014 for bystander CPR (A), bystander defibrillation (B), first responder CPR (C) and first responder defibrillation (D) in Black and White patients in only shockable patients. The analyses are adjusted for age, sex, location of arrest, witnessed status and neighborhood factors (income, education, and urban/rural status) and includes an interaction between race and calendar year. The interaction is showed as a P value in the Figure. A P value <0.05 is considered statistically significant. CPR indicates cardiopulmonary resuscitation; and OHCA, out‐of‐hospital cardiac arrest.
Defibrillation
No difference was found overall in AED application and bystander defibrillation between the 2 groups (Black: 2.4%; White: 2.2%), whereas Black patients were less likely to be defibrillated by a first responder compared to White patients (Black: 13.5%; White: 16.6%) (Table). Over the study period, no significant temporal change was observed in bystander defibrillation, but first responder defibrillation went from 14.7% [95% CI 12.3%–17.1%] in 2010 to 17.3% [95% CI 15.0%–19.6%] in 2014 in Black patients, (P=0.16) and from 13.2% [95% CI 11.6%–14.8%] in 2010 to 17.2% [95% CI 15.6%–18.8%] in 2014 in White patients (P<0.001). The trend was only statistically significant for White patients, though no substantial differences in temporal trends between the 2 groups in either bystander (P=0.73) or first responder defibrillation (P=0.50) were observed (Figure 1B and 1D). The same trend was observed in the crude analysis and when adjusting for only age, sex, and neighborhood factors (Figure S2 and S3). A small temporal non‐significant increase was observed in bystander defibrillation for both groups, but only for White patients in first responder defibrillation in shockable patients (Figure 2B and 2D).
Survival
The estimated crude and adjusted probability of return of spontaneous circulation increased significantly for both patient groups from 2010 to 2014 (Figure S4). Figure 3 shows the estimated probabilities of survival to discharge from 2010 to 2014. Both in crude and adjusted models, the probabilities of survival increased significantly for White patients, whereas survival remained overall similar for Black over time (fully adjusted analysis: Black: 8.9% [95% CI 7.0%–10.8%] in 2010 to 9.5% [95% CI 7.8%–11.2%] in 2014, P=0.60 versus White: 8.0% [95% CI 6.6%–9.4%] in 2010 to 11.4% [95% CI 9.9%–12.9%] in 2014, P=0.004; P for difference between the 2 groups=0.14). The same was observed for survival with favorable neurological outcome (Figure S5). In a restricted analysis including only patients with an initial shockable rhythm, the estimated adjusted probability for survival to discharge went from 21.7% [95% CI 15.3%–28.1%] in 2010 to 29.0% [95% CI 22.9%–35.1%] in 2014 in Black patients (P=0.10) and from 24.8% [95% CI 20.1%–29.5%] in 2010 to 34.6% [95% CI 29.9%–39.3%] in 2014 in White patients (P=0.02) (Figure 4). There was no significant difference between the groups (P=0.78).
Figure 3. Predicted probabilities of survival to discharge.
Predicted probabilities for survival to discharge with 95% CIs from 2010 to 2014 comparing Black and White patients. The Figure shows (A) a crude analysis, (B) adjusted for patient factors (age, sex, location of arrest, witnessed status, and initial rhythm), (C) adjusted for patient factors and neighborhood factors (income, education, urban/rural status) and (D) adjusted for patient factors, neighborhood factors and interventions (bystander and first responder CPR and defibrillation). All analyses include an interaction between race and calendar year. The interaction is showed as a P value in the Figure. A P value <0.05 is considered statistically significant. CPR indicates cardiopulmonary resuscitation; and OHCA, out‐of‐hospital cardiac arrest.
Figure 4. Predicted probabilities of survival to discharge in patients with shockable heart rhythm.
Predicted probabilities for survival to discharge in only shockable patients with 95% CIs from 2010 to 2014 comparing Black and White patients. The Figure shows (A) a crude analysis, (B) adjusted for patient factors (age, sex, location of arrest, and witnessed status), (C) adjusted for patient factors and neighborhood factors (income, education, urban/rural status) and (D) adjusted for patient factors, neighborhood factors, and interventions (bystander and first responder CPR and defibrillation). All analyses include an interaction between race and calendar year. The interaction is showed as a P value in the Figure. A P value <0.05 is considered statistically significant. CPR indicates cardiopulmonary resuscitation; and OHCA, out‐of‐hospital cardiac arrest.
Discussion
This study aimed to examine potential racial differences and changes in rates of bystander and/or first responder interventions due to the implementation of the RACE‐CARS program/HeartRescue project in North Carolina from 2010 to 2014, and to analyze changes in survival, accounting especially for neighborhood characteristics. Our study had 3 main findings: (1) from 2010 to 2014 bystander CPR and first responder defibrillation increased significantly in White patients, whereas only bystander CPR increased significantly in Black patients where an overall lower rate also was observed compared to White patients; (2) even though no significant difference was observed between the 2 groups, survival to discharge and survival with favorable neurological outcome increased significantly for White patients from 2010 to 2014, whereas no change was observed for Black patients in the time period. This was observed both in crude and adjusted analyses of non‐modifiable factors (age, sex, location of arrest, witnessed status, and initial rhythm), neighborhood factors (income status, educational status and urban/rural), and modifiable factors (bystander and first responder CPR and defibrillation prior to EMS arrival). Importantly (3), when limiting the analysis to patients with initial shockable rhythm increased survival was observed for both Black and White patients. Overall, these findings suggest that the observed difference in survival between races could be related to both differences in non‐modifiable pre‐arrest factors as well as in more modifiable factors. Nonetheless, additional interventions may be needed to help increase survival for Black patients. Notably, the observed increase in survival only among shockable patients has also been reported among other patient groups (men versus women) and is an important issue to address since our current strategies may thus only benefit a minority of cardiac arrest patients. 22 , 23
This study adds novel findings to understand the gap in outcomes between Black and White patients who suffer OHCA. Importantly, our findings could indicate that many different factors including general health risk factors and typically measured socioeconomic parameters and other than pre‐hospital interventions may contribute to gaps in care and outcomes between Black and White patients. Differences in bystander interventions (CPR and defibrillation) according to race have previously been observed with overall lower rates of bystander CPR and defibrillation in Black patients and also in neighborhoods with a higher proportion of Black inhabitants. 7 , 8 , 9 , 24 Importantly, our study found significant increases over time in both bystander CPR and first responder defibrillation in both Black and White patients indicating the initiatives made as part of the RACE‐CARS program/HeartRescue Project in North Carolina 13 , 14 overall had a positive effect, irrespective of race. These results persisted even when adjusting for both important patient and neighborhood socioeconomic factors.
Even though we observed increased probability of return of spontaneous circulation in both patient groups from 2010 to 2014, as a likely result of the improved pre‐hospital care, survival to discharge, as well as survival with favorable neurological outcome, only increased significantly in White patients, in both crude and adjusted analyses. Factors other than pre‐hospital factors may partly explain why survival to discharge increased significantly in White patients and not in Black patients. For example, it could be the differential use of emergency coronary interventions, which is more common in White patients, that resulted in greater likelihood of White patients surviving, as well as other elements of post‐resuscitation hospital care. Previous research has also suggested that racial disparities in survival could be a result of differences in both non‐modifiable factors as age, sex, location of arrest, witnessed status and initial heart rhythm and in more modifiable factors as bystander and first responder CPR and defibrillation. 9 In relation to race socioeconomic effects are also often mentioned, and socioeconomic differences have earlier been found associated with both intervention and survival, 11 , 25 and have earlier been suggested to explain some of the racial differences in survival. Supporting this, we found that Black patients were more likely to have OHCA in neighborhoods with predominantly poorer or less educated residents. However, adjusting for all these factors (the non‐modifiable factors, the modifiable factors, and the neighborhood socioeconomic factors) in our multivariable model did not change our findings, and survival was still observed to be lower and with limited change over time in Black patients compared to an increase for White patients. Thus, our results suggest that differences in socioeconomic factors do not fully explain racial differences in survival.
To investigate the observed differences further, we examined only patients with initial shockable rhythm given Black patients had lower rates of shockable rhythms—a finding in previous studies as well. 9 , 26 Importantly, when limiting the analysis to patients with a shockable rhythm, we observed an increase in survival in both Black and White patients. The increase in survival remained lower for Black patients compared to White patients and was not statistically significant, but the lack of significance may be due to a small sample size.
However, the lower rate of shockable rhythm seemed to be an important contributor to the observed overall lower survival for Black patients, since survival seemed to increase in both patient groups over time when including only shockable patients. Several factors may influence the likelihood of having an initial shockable heart rhythm and thereby improved survival. Factors as younger age, performed CPR and defibrillation, 27 including the availability of CPR‐trained bystanders and nearby AEDs, the willingness or potential barriers to intervene and contact authorities 8 , 28 are all examples of factors that can affect the window for having a shockable heart rhythm by affecting time from collapse to call for help, to CPR and to contact 911/EMS. Another aspect is also the fact that more Black patients have their cardiac arrest in urban areas, where EMS is more likely to be first on scene compared with first responders which overall could be associated with a delay in interventions. Importantly, bystander interventions prior to EMS arrival increased for both groups and perhaps more targeted CPR training and AED availability including dispatch‐assistance could potentially help the Black patients further. 29 Other factors that may also affect the observed difference in heart rhythm could be underlying physiological differences including potential severity of cardiac pathology, and genetic differences 30 that could affect the etiology of the cardiac arrest. Studies have shown that Black patients have a higher rate of sudden cardiac death, 9 , 26 , 31 electrocardiogram abnormalities, 32 , 33 left ventricular hypertrophy and cardiomyophathy. 34 , 35 This is supported by our finding of a higher incidence of cardiac arrests in Black patients per 100 000 inhabitants and that the lower rate of shockable rhythm in Black patients was observed even when restricting the analysis to witnessed arrests and arrests with bystander CPR. More research into potential reasons for lower rates of shockable heart rhythm among Black patients is warranted. Lastly and importantly, even though this study found a positive impact of the RACE‐CARS program/HeartRescue interventions in North Carolina irrespective of race, future programs and research should be targeted minority communities to develop tailored interventions so more minorities can be reached for cardiac arrest education and more effectively can respond to cardiac arrest. It also includes studies on how to improve survival for all patients with an initial non‐shockable heart rhythm. This is needed to improve overall survival, since only a minority of patients has an initial shockable heart rhythm for all demographic groups, and current strategies and treatments only seem to benefit those with an initial shockable heart rhythm.
Limitations
This study had several limitations of which the first is its observational design, where the relationships must be seen as associations and not causal effects. We pursued high‐quality data with prospective and uniform data collection through the CARES registry following Utstein guidelines 18 and we only included data from counties with 100% enrollment, to reduce the risk of bias due to changes in reporting over time, as previously done. 14 There may be differences between the included and excluded counties, yet characteristics in the included counties are similar to other cardiac arrest populations, indicating some generalizability. 7 , 8 , 9 , 10 Another limitation is the reporting of patient race in the CARES registry that could lead to misclassification, since it is seldom reported by the patient, who is unable to self‐report race when incapacitated, but is primarily based on provider assessment, or rarely by family if they are available and if they are asked to provide this information. We do not have further details on who reported race for each patient, which may add to imprecision. Additionally, this study was limited by the lack of important information on several factors that may influence outcome as for example patient socioeconomic factors, time for recognition of arrest and time from recognition to 911 call, however, these are practically impossible to obtain, as is the case in other cardiac arrest studies. Further, time to interventions (CPR and defibrillation), the quality of performed CPR, assistance from dispatchers including identification of arrest, as well as additional information on the receiving hospitals and the in‐hospital care of the patients, etc. were also not available. We only included cardiac arrests of presumed cardiac etiology since the CARES registry in the current time period only required these cases to be captured (prior to 2013). From the beginning of 2013 the CARES registry began collecting cases of all non‐traumatic etiologies. Including cases with other etiologies than presumed cardiac etiology or a further sub‐specification of the cardiac etiology might have helped enlighten the interesting difference in for example initial shockable rhythm between Black and White patients.
Conclusions
After the RACE‐CARS/HeartRescue quality improvement program, bystander CPR and first responder defibrillation increased for both Black and White patients, whereas survival overall only increased in White patients, also when adjusting for important patient‐, cardiac arrest‐related, and neighborhood socioeconomic factors. This indicates that the improvements work irrespective of race, but more work remains to be done to explain why survival to discharge only increased significantly in White patients. The lower initial shockable rhythm among Black patients could be an important contributor due to the observed improvement in survival in only shockable patients. Examining more factors influencing the underlying rhythm could be an important focus in future studies.
Sources of Funding
Dr Moeller has received grants from Karen Elise Jensen Fonden, Denmark, and the Laerdal Foundation, Norway. Dr Malta Hansen reports to have received grants from TrygFonden, Denmark, Helsefonden, Denmark, and the Laerdal foundation Norway. Dr Kragholm reports to have received grants from The Laerdal Foundation, Norway. Dr McNally received grant funding from the American Red Cross, Washington DC, and American Heart Association, Dallas, Texas. Drs Becker, Jollis and Granger report having received grants from the Medtronic Foundation, Minneapolis, Minnesota.
Disclosures
None.
Supporting information
Figure S1–S5
Supplementary Material for this article is available at https://www.ahajournals.org/doi/suppl/10.1161/JAHA.120.019082
For Sources of Funding and Disclosures, see page 11.
References
- 1. Atwood C, Eisenberg MS, Herlitz J, Rea TD. Incidence of EMS‐treated out‐of‐hospital cardiac arrest in Europe. Resuscitation. 2005;67:75–80. doi: 10.1016/j.resuscitation.2005.03.021 [DOI] [PubMed] [Google Scholar]
- 2. Benjamin EJ, Virani SS, Callaway CW, Chamberlain AM, Chang AR, Cheng S, Chiuve SE, Cushman M, Delling FN, Deo R, et al. Heart disease and stroke statistics‐2018 update: a report from the American Heart Association. Circulation. 2018;137:e67–e492. DOI: 10.1161/CIR.0000000000000558 [DOI] [PubMed] [Google Scholar]
- 3. Hollenberg J, Herlitz J, Lindqvist J, Riva G, Bohm K, Rosenqvist M, Svensson L. Improved survival after out‐of‐hospital cardiac arrest is associated with an increase in proportion of emergency crew–witnessed cases and bystander cardiopulmonary resuscitation. Circulation. 2008;118:389–396. DOI: 10.1161/CIRCULATIONAHA.107.734137 [DOI] [PubMed] [Google Scholar]
- 4. Wissenberg M, Lippert FK, Folke F, Weeke P, Hansen CM, Christensen EF, Jans H, Hansen PA, Lang‐Jensen T, Olesen JB, et al. Association of national initiatives to improve cardiac arrest management with rates of bystander intervention and patient survival after out‐of‐hospital cardiac arrest. JAMA. 2013;310:1377–1384. DOI: 10.1001/jama.2013.278483 [DOI] [PubMed] [Google Scholar]
- 5. Iwami T, Nichol G, Hiraide A, Hayashi Y, Nishiuchi T, Kajino K, Morita H, Yukioka H, Ikeuchi H, Sugimoto H, et al. Continuous improvements in "chain of survival" increased survival after out‐of‐hospital cardiac arrests: a large‐scale population‐based study. Circulation. 2009;119:728–734. DOI: 10.1161/CIRCULATIONAHA.108.802058 [DOI] [PubMed] [Google Scholar]
- 6. Graham R, McCoy MA, Schultz AM editors. Strategies to Improve Cardiac Arrest Survival: A Time to Act. The National Academies Collection: Reports funded by National Institutes of Health. Washington (DC), 2015. [PubMed]
- 7. Sasson C, Magid DJ, Chan P, Root ED, McNally BF, Kellermann AL, Haukoos JS, Group CS . Association of neighborhood characteristics with bystander‐initiated CPR. N Engl J Med. 2012;367:1607–1615. DOI: 10.1056/NEJMoa1110700 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Starks MA, Schmicker RH, Peterson ED, May S, Buick JE, Kudenchuk PJ, Drennan IR, Herren H, Jasti J, Sayre M, et al. Association of neighborhood demographics with out‐of‐hospital cardiac arrest treatment and outcomes: where you live may matter. JAMA cardiology. 2017;2:1110–1118. DOI: 10.1001/jamacardio.2017.2671 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Becker LB, Han BH, Meyer PM, Wright FA, Rhodes KV, Smith DW, Barrett J. Racial differences in the incidence of cardiac arrest and subsequent survival. The CPR Chicago project. N Engl J Med. 1993;329:600–606. DOI: 10.1056/NEJM199308263290902 [DOI] [PubMed] [Google Scholar]
- 10. Galea S, Blaney S, Nandi A, Silverman R, Vlahov D, Foltin G, Kusick M, Tunik M, Richmond N. Explaining racial disparities in incidence of and survival from out‐of‐hospital cardiac arrest. Am J Epidemiol. 2007;166:534–543. DOI: 10.1093/aje/kwm102 [DOI] [PubMed] [Google Scholar]
- 11. Vaillancourt C, Lui A, De Maio VJ, Wells GA, Stiell IG. Socioeconomic status influences bystander CPR and survival rates for out‐of‐hospital cardiac arrest victims. Resuscitation. 2008;79:417–423. DOI: 10.1016/j.resuscitation.2008.07.012 [DOI] [PubMed] [Google Scholar]
- 12. Zhao DI, Post WS, Blasco‐Colmenares E, Cheng A, Zhang Y, Deo R, Pastor‐Barriuso R, Michos ED, Sotoodehnia N, Guallar E. Racial differences in sudden cardiac death. Circulation. 2019;139:1688–1697. DOI: 10.1161/CIRCULATIONAHA.118.036553 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. van Diepen S, Abella BS, Bobrow BJ, Nichol G, Jollis JG, Mellor J, Racht EM, Yannopoulos D, Granger CB, Sayre MR. Multistate implementation of guideline‐based cardiac resuscitation systems of care: description of the HeartRescue project. Am Heart J. 2013;166:647–653.e2. DOI: 10.1016/j.ahj.2013.05.022 [DOI] [PubMed] [Google Scholar]
- 14. Malta Hansen C, Kragholm K, Pearson DA, Tyson C, Monk L, Myers B, Nelson D, Dupre ME, Fosbøl EL, Jollis JG, et al. Association of bystander and first‐responder intervention with survival after out‐of‐hospital cardiac arrest in North Carolina, 2010–2013. JAMA. 2015;314:255–264. DOI: 10.1001/jama.2015.7938 [DOI] [PubMed] [Google Scholar]
- 15. McNally B, Stokes A, Crouch A, Kellermann AL, Group CS . CARES: cardiac arrest registry to enhance survival. Ann Emerg Med. 2009;54:674–683.e2. DOI: 10.1016/j.annemergmed.2009.03.018 [DOI] [PubMed] [Google Scholar]
- 16. 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]
- 17. Cardiac Arrest Registry to Enhance Survival (CARES) . 2021. Data Dictionary. https://mycares.net/sitepages/uploads/2020/Data%20Dictionary%20(2021).pdf. Published 2021. Accessed 2021.
- 18. Perkins GD, Jacobs IG, Nadkarni VM, Berg RA, Bhanji F, Biarent D, Bossaert LL, Brett SJ, Chamberlain D, de Caen AR, et al. Cardiac arrest and cardiopulmonary resuscitation outcome reports: update of the utstein resuscitation registry templates for out‐of‐hospital cardiac arrest: a statement for healthcare professionals from a task force of the international liaison committee on resuscitation (American Heart Association, European resuscitation council, Australian and New Zealand council on resuscitation, heart and stroke foundation of Canada, InterAmerican Heart Foundation, resuscitation council of Southern Africa, resuscitation council of Asia); and the American Heart Association emergency cardiovascular care committee and the council on cardiopulmonary, critical care, perioperative and resuscitation. Resuscitation. 2015;96:328–340. DOI: 10.1016/j.resuscitation.2014.11.002 [DOI] [PubMed] [Google Scholar]
- 19. Sasson C, Magid DJ, Chan P, Root ED, McNally BF, Kellermann AL, Haukoos JS. Association of neighborhood characteristics with bystander‐initiated CPR. N Engl J Med. 2012;367:1607–1615 DOI: 10.1056/NEJMoa1110700 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Krieger N. Overcoming the absence of socioeconomic data in medical records: validation and application of a census‐based methodology. Am J Public Health. 1992;82:703–710. DOI: 10.2105/AJPH.82.5.703 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. United States Census Bureau . Census Bureau Releases New Local‐Level Demographic Information from 2010 Census for Alaska, Colorado, Connecticut, Nebraska and North Carolina. https://www.census.gov/newsroom/releases/archives/2010_census/cb11‐cn159.html. Published 2011. Accessed 2019.
- 22. Malta Hansen C, Kragholm K, Dupre ME, Pearson DA, Tyson C, Monk L, Rea TD, Starks MA, Nelson D, Jollis JG, et al. Association of bystander and first‐responder efforts and outcomes according to sex: results from the North Carolina heartrescue statewide quality improvement initiative. J Am Heart Assoc. 2018;7:e009873. DOI: 10.1161/JAHA.118.009873 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Rajan S, Folke F, Hansen SM, Hansen CM, Kragholm K, Gerds TA, Lippert FK, Karlsson L, Møller S, Køber L, et al. Incidence and survival outcome according to heart rhythm during resuscitation attempt in out‐of‐hospital cardiac arrest patients with presumed cardiac etiology. Resuscitation. 2017;114:157–163. DOI: 10.1016/j.resuscitation.2016.12.021 [DOI] [PubMed] [Google Scholar]
- 24. Benson PC, Eckstein M, McClung CD, Henderson SO. Racial/ethnic differences in bystander CPR in Los Angeles, California. Ethn Dis. 2009;19:401–406. [PubMed] [Google Scholar]
- 25. Andersen LW, Holmberg MJ, Granfeldt A, Løfgren B, Vellano K, McNally BF, Siegerink B, Kurth T, Donnino MW, Group CS . Neighborhood characteristics, bystander automated external defibrillator use, and patient outcomes in public out‐of‐hospital cardiac arrest. Resuscitation. 2018;126:72–79. DOI: 10.1016/j.resuscitation.2018.02.021 [DOI] [PubMed] [Google Scholar]
- 26. Teodorescu C, Reinier K, Dervan C, Uy‐Evanado A, Samara M, Mariani R, Gunson K, Jui J, Chugh SS. Factors associated with pulseless electric activity versus ventricular fibrillation: the Oregon sudden unexpected death study. Circulation. 2010;122:2116–2122. DOI: 10.1161/CIRCULATIONAHA.110.966333 [DOI] [PubMed] [Google Scholar]
- 27. Sekimoto M, Noguchi Y, Rahman M, Hira K, Fukui M, Enzan K, Inaba H, Fukui T. Estimating the effect of bystander‐initiated cardiopulmonary resuscitation in Japan. Resuscitation. 2001;50:153–160. DOI: 10.1016/S0300-9572(01)00330-6 [DOI] [PubMed] [Google Scholar]
- 28. Sasson C, Haukoos JS, Bond C, Rabe M, Colbert SH, King R, Sayre M, Heisler M. Barriers and facilitators to learning and performing cardiopulmonary resuscitation in neighborhoods with low bystander cardiopulmonary resuscitation prevalence and high rates of cardiac arrest in Columbus, OH. Circ Cardiovasc Qual Outcomes. 2013;6:550–558. DOI: 10.1161/CIRCOUTCOMES.111.000097 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Root ED, Gonzales L, Persse DE, Hinchey PR, McNally B, Sasson C. A tale of two cities: the role of neighborhood socioeconomic status in spatial clustering of bystander CPR in Austin and Houston. Resuscitation. 2013;84:752–759. DOI: 10.1016/j.resuscitation.2013.01.007 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Deo R, Albert CM. Epidemiology and genetics of sudden cardiac death. Circulation. 2012;125:620–637. DOI: 10.1161/CIRCULATIONAHA.111.023838 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Reinier K, Nichols GA, Huertas‐Vazquez A, Uy‐Evanado A, Teodorescu C, Stecker EC, Gunson K, Jui J, Chugh SS. Distinctive clinical profile of blacks versus whites presenting with sudden cardiac arrest. Circulation. 2015;132:380–387. DOI: 10.1161/CIRCULATIONAHA.115.015673 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Prineas RJ, Le A, Soliman EZ, Zhang ZM, Howard VJ, Ostchega Y, Howard G, Reasons for G, Racial Differences in Stroke I . US national prevalence of electrocardiographic abnormalities in black and white middle‐aged (45‐ to 64‐Years) and older (≥65‐Years) adults (from the Reasons for Geographic and Racial Differences in Stroke Study). Am J Cardiol. 2012;109:1223–1228. DOI: 10.1016/j.amjcard.2011.11.061 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Walsh JA III, Prineas R, Daviglus ML, Ning H, Liu K, Lewis CE, Sidney S, Schreiner PJ, Iribarren C, Lloyd‐Jones DM. Prevalence of electrocardiographic abnormalities in a middle‐aged, biracial population: coronary artery risk development in young adults study. J Electrocardiol. 2010;43:385.e1–385.e9. DOI: 10.1016/j.jelectrocard.2010.02.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Drazner MH, Dries DL, Peshock RM, Cooper RS, Klassen C, Kazi F, Willett D, Victor RG. Left ventricular hypertrophy is more prevalent in blacks than whites in the general population: the Dallas Heart Study. Hypertension. 2005;46:124–129. DOI: 10.1161/01.HYP.0000169972.96201.8e [DOI] [PubMed] [Google Scholar]
- 35. Sorensen LL, Pinheiro A, Dimaano VL, Pozios I, Nowbar A, Liu H, Luo H‐C, Lin X, Olsen NT, Hansen TF, et al. Comparison of clinical features in blacks versus whites with hypertrophic cardiomyopathy. Am J Cardiol. 2016;117:1815–1820. DOI: 10.1016/j.amjcard.2016.03.017 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Figure S1–S5