Abstract
Background
Variability in the provision of bystander cardiopulmonary resuscitation (BCPR) may influence neurological outcomes in patients with out-of-hospital cardiac arrest (OHCA) according to regional socioeconomic factors. We aimed to investigate the association between BCPR and neurological outcomes of OHCA and whether these associations differ across regions with different levels of deprivation.
Methods
Using data collected from the Korean National OHCA Registry between 2013 and 2023, adults (≥18 years of age) with OHCA of presumed cardiac origin were identified. The main exposure variable was BCPR and the regional deprivation index. The primary outcome was a favourable neurological outcome, defined as a Cerebral Performance Category score of 1 or 2. Multivariate logistic regression was used to evaluate the associations between BCPR and favourable neurological outcomes and to assess the interaction between BCPR and regional deprivation index.
Results
Of 184,727 individuals who experienced OHCA, 112,728 (61.0%) received BCPR. There were differences in the rate of BCPR according to regional deprivation group: high, 57.9%; moderate, 59.9%; and low, 62.6%. In total, 10,669 (5.8%) patients experienced favourable neurological outcomes. Multivariable models revealed different associations between BCPR and favourable neurological outcomes according to regional deprivation level: high, adjusted odds ratio (OR) 1.56 (95% confidence interval [CI] 1.23–1.99); moderate, 1.31 (95% CI 1.19–1.44); and low, 1.34 (95% CI 1.23–1.45). The magnitude of the association showed a significant interaction (p < 0.01).
Conclusions
Although high-deprivation regions had lower BCPR rates, the association between BCPR and favourable neurological outcomes was stronger in high-deprivation regions than in moderate- and low-deprivation regions. Multifaceted interventions are needed to increase BCPR rates in highly deprived regions.
Keywords: Out-of-hospital cardiac arrest, Bystander cardiopulmonary resuscitation, Deprivation index
Introduction
Bystander cardiopulmonary resuscitation (BCPR) is a crucial link in the survival chain of individuals who experience out-of-hospital cardiac arrest (OHCA).1, 2 BPCR provides immediate, continuous chest compressions that maintain minimal cerebral perfusion to patients with cardiac arrest. This crucial intervention is strongly associated with a higher likelihood of favourable neurological outcomes.3 Previous study reported that BCPR increased the odds for favourable neurological outcome by 1.93 fold (95% Confidence interval 1.56–2.39).4
However, the provision of BCPR varies according to the socioeconomic status (SES) of the area in which the OHCA occurs.5 Socio-economic deprivation in the region where cardiac arrest occurs is a representative socioeconomic indicator associated with BCPR provision. Area-based deprivation refers to a structural lack of essential physical resources, adequate housing, and financial security within a definite geographic boundary.6 This concept focuses on the observable lack of goods and amenities relative to societal standards, rather than on individual income poverty.7, 8 It is a contextual characteristic of an area that independently contributes to spatial and health inequalities.9 A recent study from England, which reported a BCPR rate of 58.2%, found that neighbourhood deprivation was significantly greater in regions with low-BCPR rates.10 Similar associations between BCPR and neighbourhood deprivation have been reported in the United States and South Korea.11, 12
There are several mechanisms for disparities in BCPR across regional deprivation levels. Factors associated with deprivation (e.g. health literacy, CPR training, education, social isolation, and population demographics) contribute to differences in bystander interventions. Previous study reported that SES influences the likelihood of previous CPR training.13 A survey of 9022 participants from across the United States found that those with lower income and poorer educational attainment were less likely to be trained in CPR.14
It remains unclear whether the effect of BCPR on the clinical outcomes in OHCA differs according to the regional deprivation for OHCA. Accordingly, we investigated whether the association between BCPR and favourable neurological outcomes following OHCA of presumed cardiac origin differed across regions with different levels of deprivation.
Methods
Study design and setting
This retrospective cohort study used data obtained from the Korean OHCA Registry (KOHCAR), a population-based surveillance registry that captures all cases of emergency medical services (EMS)-assessed OHCA.15 The standardized international Utstein definitions were used to delineate clinical variables and outcomes to ensure uniformity.16 The design of the KOHCAR registry and data elements have been described previously.15
The Republic of Korea has a public EMS system, that includes a dispatcher-assisted CPR program.17 Mandatory CPR training programs for first responders and for students and teachers at primary, middle, and high school was implemented in 2008 under the EMS act and in 2012 under the School Health Act.18
The specific program protocols were approved by the Institutional Review Board of Seoul National University Hospital (Seoul, Korea). Due to the retrospective design of the study and the use of anonymised data, the requirement for informed consent was waived. The study protocol was developed, executed, and verified in strict adherence to the STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) checklist for observational studies to ensure comprehensive and transparent reporting (supplement).
Subject of the study
The present study included data from individuals who experienced presumed cardiac origin OHCA registered in the KOHCAR between 1 January 2013 and 31 December 2023. We included patients with OHCA who received resuscitation from EMS personnel. Individuals younger than 18 years were excluded to restrict the analysis to adults, and OHCAs witnessed by EMS providers were excluded. OHCAs occurring in nursing homes or healthcare facilities were also excluded because these sites have healthcare professionals and their processes-of-care and outcomes differ from those of most OHCAs. Finally, patients with missing data for location, BCPR, or survival outcomes were also excluded.
Independent variable and study outcomes
The primary exposure variable was BCPR. BCPR is performed by an individual who responds to cardiac arrest but is not part of an organised emergency response system. Physicians, nurses, and paramedics may be described as performing BCPR if they are not part of the emergency response system involved in victim resuscitation. BCPR may involve compression only or compression with ventilation.13
The regional deprivation of the OHCA location was examined as an effect modifier to determine whether the association between BCPR and survival differed according to regional deprivation. We used the publicly available deprivation index from the Busan Public Health Policy Institute in the Republic of Korea.19, 20 Regional deprivation was constructed based on the resources, education, and housing status of a geographical area using 10 variables from the 2015 and 2020 Korean population census data. Each variable was consisted of poor living conditions, less than high school education, the head of the household's occupation belongs to the manual labour class, 65 years and older population, single-person households, home-owning households, apartment households, female-headed households, divorce/widowhood, and households without cars. The deprivation index in the Republic of Korea is reported every 5 years at the district level (n = 250). We applied the 2015 index to OHCA cases from 2013 to 2014 and the 2020 index to OHCA cases from 2015 to 2019. For OHCA cases from 2020 to 2023, we also applied the 2020 index because the 2025 index had not yet been released. Regional deprivation was classified into three groups based on previously used cut-off points, defined a priori as follows: low (≤50th percentile); moderate (50–90th percentile); and high (>90th percentile).21, 22 Regional deprivation was categorised such that higher values indicate greater deprivation.
The primary outcome was favourable neurological outcome at hospital discharge, which was defined as survival to discharge with a Cerebral Performance Category Scale (CPC) score of 1 or 2, which denotes survival without severe neurological disability.23 Poor neurological outcome was defined as survival to discharge with a CPC score of 3–5, which denotes survival with severe neurological disability and death. Survival to hospital discharge, regardless of neurological status, was also evaluated as a secondary outcome.
Statistical analysis
The primary objective was to evaluate whether the association between BCPR and survival outcomes differed among regional deprivation groups. Baseline differences among patients with OHCA with and without BCPR are expressed as median with interquartile range (IQR) for continuous variables and count (with percentage distribution) for categorical variables. The independent association between BCPR and favourable neurological outcomes in the regional deprivation group was first assessed by constructing a multivariable model using logistic regression analysis. In addition to BCPR and regional level deprivation groups, these models also included patient age (categorised as <55, 55–64, 65–75, 75–84, ≥85 years), sex, medical insurance status, witnessed arrest status, location of OHCA (public, private and others), EMS response time interval (5 min), initial presenting rhythm (shockable vs. non-shockable), season (spring, summer, autumn, winter), and day of the week (weekday vs. weekend). The model calendar year of the OHCA was also included to control for temporal trends. To examine whether the association between BCPR and neurological outcome differed according to regional deprivation group, an interaction term was included between the BCPR and regional deprivation groups in the model. We hypothesised that the strength of BCPR in linking the chain of survival in patients with cardiac arrest may vary by region. Similar models were constructed for the secondary outcome of survival to discharge. For each analysis, the null hypothesis was evaluated at a two-sided significance level of 0.05, and calculated 95% CIs using robust standard errors. All statistical analyses were performed using SAS version 9.4 (SAS Institute, Cary, NC, USA).
Results
Between 2013 and 2023, data from 321,737 individuals who experienced OHCA were enrolled in the KOHCAR registry. For the purposes of the present study, however, the following were excluded: paediatric cases (n = 6179); OHCAs of non-cardiac origin (n = 75,375); OHCAs witnessed by an EMS provider (n = 26,931); and OHCA in a nursing home of a healthcare facility (n = 23,840). There was no missing information regarding BCPR or the address of the district where the OHCA occurred (Fig. 1). The final cohort comprised data from 184,727 patients.
Fig. 1.
Patients selection flow.
Overall, 112,728 (61.0%) individuals who experienced OHCA received BCPR. A comparison of characteristics of those with and without BCPR is presented in Table 1. Those who received BCPR were more likely than those who did not receive BCPR to have a witnessed OHCA (49.1% vs. 36.4%), an OHCA in a public location (22.5% vs. 18.4%), and an OHCA with an initial shockable arrest rhythm (12.7% vs. 7.3%). In total, 10,669 patients (5.8%) experienced favourable neurological outcome. The rate of favourable neurological outcome was 7.4% (8324/112,728) in those with BCPR and 3.4% (2345/71,999) among those who did not.
Table 1.
Patients characteristics based on bystander CPR.
| Total |
Bystander CPR |
p-value | |||||
|---|---|---|---|---|---|---|---|
|
No (n = 71,999) |
Yes (n = 112,728) |
||||||
| N | N | (%) | N | (%) | |||
| Sex | 0.33 | ||||||
| Female | 62,747 | 23,404 | 32.5 | 39,343 | 34.9 | ||
| Male | 121,980 | 48,595 | 67.5 | 73,385 | 65.1 | ||
| Age group (years old) | |||||||
| <55 | 65,104 | 23,212 | 32.2 | 41,892 | 37.2 | ||
| 56–65 | 39,202 | 15,940 | 22.1 | 23,262 | 20.6 | ||
| 66–75 | 54,593 | 22,773 | 31.6 | 31,820 | 28.2 | ||
| >75 | 25,828 | 10,074 | 14.0 | 15,754 | 14.0 | ||
| Health insurance | 0.06 | ||||||
| NHS | 168,765 | 64,602 | 89.7 | 104,163 | 92.4 | ||
| Medical aid | 15,962 | 7397 | 10.3 | 8565 | 7.6 | ||
| Place of arrest | |||||||
| Public | 38,595 | 13,273 | 18.4 | 25,322 | 22.5 | ||
| Private | 144,602 | 58,149 | 80.8 | 86,453 | 76.7 | ||
| Other | 1530 | 577 | 0.8 | 953 | 0.8 | ||
| Witnessed status | 0.003 | ||||||
| No | 103,144 | 45,819 | 63.6 | 57,325 | 50.9 | ||
| Yes | 81,583 | 26,180 | 36.4 | 55,403 | 49.1 | ||
| Bystander type | |||||||
| Family | 46,270 | 17,455 | 24.2 | 28,815 | 25.6 | ||
| Lay person | 131,601 | 51,419 | 71.4 | 80,182 | 71.1 | ||
| Unknown | 6856 | 3125 | 4.3 | 3731 | 3.3 | ||
| Year of arrest | |||||||
| 2013 | 13,546 | 7404 | 10.3 | 6142 | 5.4 | ||
| 2014 | 15,154 | 7020 | 9.8 | 8134 | 7.2 | ||
| 2015 | 15,774 | 6546 | 9.1 | 9228 | 8.2 | ||
| 2016 | 15,868 | 6097 | 8.5 | 9771 | 8.7 | ||
| 2017 | 15,693 | 5831 | 8.1 | 9862 | 8.7 | ||
| 2018 | 16,492 | 6222 | 8.6 | 10,270 | 9.1 | ||
| 2019 | 16,972 | 6258 | 8.7 | 10,714 | 9.5 | ||
| 2020 | 18,126 | 6561 | 9.1 | 11,565 | 10.3 | ||
| 2021 | 18,892 | 6619 | 9.2 | 12,273 | 10.9 | ||
| 2022 | 19,505 | 6941 | 9.6 | 12,564 | 11.1 | ||
| 2023 | 18,705 | 6500 | 9.0 | 12,205 | 10.8 | ||
| Season | |||||||
| Spring | 47,059 | 18,265 | 25.4 | 28,794 | 25.5 | ||
| Summer | 40,253 | 15,640 | 21.7 | 24,613 | 21.8 | ||
| Fall | 45,551 | 17,555 | 24.4 | 27,996 | 24.8 | ||
| Winter | 51,864 | 20,539 | 28.5 | 31,325 | 27.8 | ||
| Day of week | 0.26 | ||||||
| Weekday | 130,365 | 51,657 | 71.7 | 78,708 | 69.8 | ||
| Weekend | 54,362 | 20,342 | 28.3 | 34,020 | 30.2 | ||
| Time of arrest | |||||||
| Non-office hours | 69,430 | 26,951 | 37.4 | 42,479 | 37.7 | ||
| Office hours (8 am–5 pm) | 115,297 | 45,048 | 62.6 | 70,249 | 62.3 | ||
| Respond time interval (minutes) | |||||||
| Missing | 285 | 105 | 0.1 | 180 | 0.2 | ||
| <5 | 44,461 | 18,157 | 25.2 | 26,304 | 23.3 | ||
| 6–10 | 96,722 | 36,852 | 51.2 | 59,870 | 53.1 | ||
| 11–15 | 29,623 | 11,427 | 15.9 | 18,196 | 16.1 | ||
| 16–20 | 8634 | 3408 | 4.7 | 5226 | 4.6 | ||
| >20 | 5002 | 2050 | 2.8 | 2952 | 2.6 | ||
| Initial ECG rhythm | 0.01 | ||||||
| Non-shockable | 165,193 | 66,770 | 92.7 | 98,423 | 87.3 | ||
| Shockable | 19,534 | 5229 | 7.3 | 14,305 | 12.7 | ||
| Survival to discharge | 0.01 | ||||||
| No | 169,158 | 67,721 | 94.1 | 101,437 | 90.0 | ||
| Yes | 15,569 | 4278 | 5.9 | 11,291 | 10.0 | ||
| Neurologic outcome | 0.01 | ||||||
| Poor (CPC 3–5) | 174,058 | 69,654 | 96.7 | 104,404 | 92.6 | ||
| Good (CPC 1 & 2) | 10,669 | 2345 | 3.3 | 8324 | 7.4 | ||
There were no significant differences in sex, individual SES, OHCA location, time of OHCA, or witnessed status among the regional deprivation groups (Table 2). The rates of BCPR differed significantly across the regional deprivation groups: high, 57.9% (10,879/18,793); moderate, 59.9% (44,542/74,396); and low, 62.6% (57,307/91,538) (p-for trend = 0.04) (Table 2). The rates of favourable neurological outcomes according to deprivation group were as follows: high, 3.5% (657/18,793); moderate, 5.4% (4052/74,396); and low, 6.5% (5960/91,538) (p-for trend = 0.01).
Table 2.
Patients characteristics based on regional deprivation.
| Total |
Regional deprivation* |
p-for trend | |||||||
|---|---|---|---|---|---|---|---|---|---|
|
High (n = 18793) |
Moderate (n = 74396) |
Low (n = 91538) |
|||||||
| N | N | (%) | N | (%) | N | (%) | |||
| Sex | 0.37 | ||||||||
| Female | 62,747 | 6404 | 34.1 | 25,226 | 33.9 | 31,117 | 34.0 | ||
| Male | 121,980 | 12,389 | 65.9 | 49,170 | 66.1 | 60,421 | 66.0 | ||
| Age group (years old) | |||||||||
| <55 | 65,104 | 5414 | 28.8 | 24,996 | 33.6 | 34,694 | 37.9 | ||
| 56–65 | 39,202 | 4037 | 21.5 | 16,270 | 21.9 | 18,895 | 20.6 | ||
| 66–75 | 54,593 | 6320 | 33.6 | 22,600 | 30.4 | 25,673 | 28.0 | ||
| >75 | 25,828 | 3022 | 16.1 | 10,530 | 14.2 | 12,276 | 13.4 | ||
| Health insurance | 0.48 | ||||||||
| NHS | 168,765 | 17,049 | 90.7 | 67,532 | 90.8 | 84,184 | 92.0 | ||
| Medical aid | 15,962 | 1744 | 9.3 | 6864 | 9.2 | 7354 | 8.0 | ||
| Place of arrest | |||||||||
| Public | 38,595 | 3685 | 19.6 | 15,728 | 21.1 | 19,182 | 21.0 | ||
| Private | 144,602 | 14,908 | 79.3 | 58,015 | 78.0 | 71,679 | 78.3 | ||
| Other | 1530 | 200 | 1.1 | 653 | 0.9 | 677 | 0.7 | ||
| Witnessed status | 0.09 | ||||||||
| No | 103,144 | 10,902 | 58.0 | 41,903 | 56.3 | 50,339 | 55.0 | ||
| Yes | 81,583 | 7891 | 42.0 | 32,493 | 43.7 | 41,199 | 45.0 | ||
| Bystander CPR | 0.04 | ||||||||
| No | 71,999 | 7914 | 42.1 | 29,854 | 40.1 | 34,231 | 37.4 | ||
| Yes | 112,728 | 10,879 | 57.9 | 44,542 | 59.9 | 57,307 | 62.6 | ||
| Bystander type | |||||||||
| Family | 46,270 | 4548 | 24.2 | 18,840 | 25.3 | 22,882 | 25.0 | ||
| Lay person | 131,601 | 13,376 | 71.2 | 52,796 | 71.0 | 65,429 | 71.5 | ||
| Unknown | 6856 | 869 | 4.6 | 2760 | 3.7 | 3227 | 3.5 | ||
| Year of arrest | |||||||||
| 2013 | 13,546 | 1398 | 7.4 | 5524 | 7.4 | 6624 | 7.2 | ||
| 2014 | 15,154 | 1594 | 8.5 | 6173 | 8.3 | 7387 | 8.1 | ||
| 2015 | 15,774 | 1597 | 8.5 | 6286 | 8.4 | 7891 | 8.6 | ||
| 2016 | 15,868 | 1614 | 8.6 | 6437 | 8.7 | 7817 | 8.5 | ||
| 2017 | 15,693 | 1666 | 8.9 | 6324 | 8.5 | 7703 | 8.4 | ||
| 2018 | 16,492 | 1620 | 8.6 | 6726 | 9.0 | 8146 | 8.9 | ||
| 2019 | 16,972 | 1731 | 9.2 | 6837 | 9.2 | 8404 | 9.2 | ||
| 2020 | 18,126 | 1853 | 9.9 | 7257 | 9.8 | 9016 | 9.8 | ||
| 2021 | 18,892 | 1961 | 10.4 | 7563 | 10.2 | 9368 | 10.2 | ||
| 2022 | 19,505 | 2015 | 10.7 | 7813 | 10.5 | 9677 | 10.6 | ||
| 2023 | 18,705 | 1744 | 9.3 | 7456 | 10.0 | 9505 | 10.4 | ||
| Season | |||||||||
| Spring | 47,059 | 4813 | 25.6 | 18,820 | 25.3 | 23,426 | 25.6 | ||
| Summer | 40,253 | 4074 | 21.7 | 16,091 | 21.6 | 20,088 | 21.9 | ||
| Fall | 45,551 | 4652 | 24.8 | 18,428 | 24.8 | 22,471 | 24.5 | ||
| Winter | 51,864 | 5254 | 28.0 | 21,057 | 28.3 | 25,553 | 27.9 | ||
| Day of week | 0.69 | ||||||||
| Weekday | 130,365 | 13,099 | 69.7 | 52,608 | 70.7 | 64,658 | 70.6 | ||
| Weekend | 54,362 | 5694 | 30.3 | 21,788 | 29.3 | 26,880 | 29.4 | ||
| Time of arrest | 0.065 | ||||||||
| Non-office hours | 69,430 | 6602 | 35.1 | 27,854 | 37.4 | 34,974 | 38.2 | ||
| Office hours (8 am–5 pm) | 115,297 | 12,191 | 64.9 | 46,542 | 62.6 | 56,564 | 61.8 | ||
| Respond time interval (minutes) | |||||||||
| Missing | 285 | 71 | 0.4 | 129 | 0.2 | 85 | 0.1 | ||
| <5 | 44,461 | 3538 | 18.8 | 18,541 | 24.9 | 22,382 | 24.5 | ||
| 6–10 | 96,722 | 7954 | 42.3 | 38,939 | 52.3 | 49,829 | 54.4 | ||
| 11–15 | 29,623 | 4297 | 22.9 | 11,335 | 15.2 | 13,991 | 15.3 | ||
| 16–20 | 8634 | 1790 | 9.5 | 3331 | 4.5 | 3513 | 3.8 | ||
| >20 | 5002 | 1143 | 6.1 | 2121 | 2.9 | 1738 | 1.9 | ||
| Initial ECG rhythm | 0.03 | ||||||||
| Non-shockable | 165,193 | 17,172 | 91.4 | 66,684 | 89.6 | 81,337 | 88.9 | ||
| Shockable | 19,534 | 1621 | 8.6 | 7712 | 10.4 | 10,201 | 11.1 | ||
| Survival to discharge | 0.01 | ||||||||
| No | 169,158 | 17,843 | 94.9 | 68,380 | 91.9 | 82,935 | 90.6 | ||
| Yes | 15,569 | 950 | 5.1 | 6016 | 8.1 | 8603 | 9.4 | ||
| Neurologic recovery | 0.01 | ||||||||
| Poor (CPC 3–5) | 174,058 | 18,136 | 96.5 | 70,344 | 94.6 | 85,578 | 93.5 | ||
| Good (CPC 1 & 2) | 10,669 | 657 | 3.5 | 4052 | 5.4 | 5960 | 6.5 | ||
Regional deprivation was classified into three groups as follows: low (≤50th percentile); moderate (50–90th percentile); and high (>90th percentile).
After multivariable adjustment, BCPR was associated with higher odds for favourable neurological outcome (adjusted odds ratio [OR] 1.32 [95% CI 1.25–1.41]; p < 0.001). When the high-deprivation group was used as the reference, the adjusted OR for favourable neurological outcome was 1.52 [95% CI 1.36–1.69] (p < 0.001) in the moderate-deprivation group and 1.77 [95% CI 1.59–1.97] (p < 0.001) in the low-deprivation group (Table 3). Results of the interaction analyses between BCPR and deprivation group for each outcome are summarised in Table 4. The interaction was significant for the comparison between the high- and moderate-deprivation groups (p < 0.01), and between the high- and low-deprivation groups (p = 0.03). The magnitude of the association between BCPR and favourable neurological outcome was greater in the high-deprivation group (adjusted OR 1.56 [95% CI 1.23–1.99]) than that in the moderate (adjusted OR 1.31 [95% CI 1.19–1.44]) and low (adjusted OR 1.34 [95% CI 1.23–1.45]) deprivation groups.
Table 3.
Multivariable analysis for clinical outcomes in patients with OHCA.
| n/N |
Survival to discharge (n = 15,569/165,195) |
n/N |
Favourable neurologic outcome* (n = 10,669/174,058) |
|||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Adjusted OR |
95% CI |
Adjusted OR |
95% CI |
|||||||
| LL | UL | LL | UL | |||||||
| Bystander CPR | ||||||||||
| No | 4278/71999 | Ref | 2345/71999 | Ref | ||||||
| Yes | 11291/112728 | 1.03 | 0.98 | 1.08 | 8324/112728 | 1.32 | 1.25 | 1.41 | ||
| Regional deprivation group** | ||||||||||
| High | 950/18793 | Ref | 657/18793 | Ref | ||||||
| Moderate | 6016/74396 | 1.61 | 1.48 | 1.76 | 4052/74396 | 1.52 | 1.36 | 1.69 | ||
| Low | 8603/91538 | 1.86 | 1.70 | 2.03 | 5960/91538 | 1.77 | 1.59 | 1.97 | ||
CPR; cardiopulmonary resuscitation, OR; odds ratio, CI; confidence interval.
Favourable neurological outcome was defined as survival to discharge with a Cerebral Performance Category Scale (CPC) score of 1 or 2. Poor neurological outcome defined as survival to discharge with a CPC score of from 3 to 5 which denotes survival with severe neurological disability and death.
The models included adjusted variables; sex, age group, health insurance status, place of arrest, witnessed status, bystander type, year of arrest.
Regional deprivation was classified into three groups as follows: low (≤50th percentile); moderate (50–90th percentile); and high (>90th percentile).
Table 4.
Association between bystander CPR and outcomes considering the effect of regional deprivation; interaction analysis.
|
Regional deprivation* |
|||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
|
High (n = 18793) |
Moderate (n = 74396) |
Low (n = 91538) |
|||||||||
| Adjusted OR |
95% CI |
Adjusted OR |
95% Cl |
Adjusted OR |
95% CI |
||||||
| LL | UL | LL | UL | LL | UL | ||||||
| Survival to discharge (n = 15,569/165,195) | |||||||||||
| Bystander CPR | |||||||||||
| No (n = 71999) | Ref | Ref | Ref | ||||||||
| Yes (n = 112728) | 1.30 | 1.08 | 1.56 | 1.02 | 0.95 | 1.10 | 1.02 | 0.95 | 1.08 | ||
| Interaction P-value | 0.006 | 0.026 | |||||||||
| Favourable neurologic recovery**(n = 10,669/174,058) | |||||||||||
| Bystander CPR | |||||||||||
| No (n = 71999) | Ref | Ref | Ref | ||||||||
| Yes (n = 112728) | 1.56 | 1.23 | 1.99 | 1.31 | 1.19 | 1.44 | 1.34 | 1.23 | 1.45 | ||
| Interaction P-value | 0.011 | 0.035 | |||||||||
CPR; cardiopulmonary resuscitation, OR; odd ratio, CI; confidence interval.
Regional deprivation was classified into three groups as follows: low (≤50th percentile); moderate (50–90th percentile); and high (>90th percentile).
The models included adjusted variables; sex, age group, health insurance status, place of arrest, witnessed status, bystander type, year of arrest.
Favourable neurological outcome was defined as survival to discharge with a Cerebral Performance Category Scale (CPC) score of 1 or 2. Poor neurological outcome defined as survival to discharge with a CPC score of from 3 to 5 which denotes survival with severe neurological disability and death.
Discussion
Using data from a large national registry, we found that the higher the regional deprivation, the lower the BCPR rate, and the higher the regional deprivation level, the less favourable the clinical outcome. However, our qualitative interaction findings suggest that the association between BCPR and clinical outcomes was greater in the high-deprivation region. This was the case for both favourable neurological outcome and overall survival after adjustment for potential confounders.
Our study extends the literature addressing BCPR across regional deprivation groups in several ways. This finding is consistent with those of several studies that demonstrated a link between deprivation and survival after cardiac arrest.21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34 The consistency of findings across studies lends credence to the overall validity of the association between increased deprivation and decreased survival.
To the best of our knowledge, this study is the first to directly examine the differential associations between BCPR and clinical outcomes across regional deprivation groups. The fact that the association between the high deprivation group and overall survival-to-discharge in this study was larger among patients who received BCPR suggests that the high-deprivation group benefitted more than the moderate- or low-deprivation groups when they received BCPR. The absence of significant differences in patient sex, ethnicity, OHCA location, or witnessed status among the deprivation groups, may support the interpretation that differences in BCPR rates contribute to the variation in outcomes across the deprivation groups. These results are consistent with those of previous studies.24 A previous meta-analysis investigating predictors of OHCA survival reported that the pooled OR for survival among patients receiving BCPR with the lowest baseline survival rates was higher than those in studies with the highest baseline rates (5.01 [95% CI 2.57–9.78] vs. 1.23 [95% CI 0.71–2.11]).25 In our study, the adjusted OR among patients receiving BCPR in the high-deprivation region with the lowest survival rate was greater than that in the moderate- and low-deprivation groups with higher survival rates.
The reasons for the stronger association between BCPR and survival outcomes in the high-deprivation group are unclear. One of possible explanation is that the effects of BCPR could differ according to EMS response intervals. The EMS response time was relatively longer in the high-deprivation region. The proportion of cases with an EMS response time interval of 15 min or less was 94.2% in low-deprivation areas and 92.4% in moderate-deprivation areas, compared with only 84.0% in high-deprivation areas (Table 2). In the case of BCPR, the negative effects of EMS response time delay could be somewhat attenuated. A previous study reported that if BCPR is administered, the response time interval range showing significant neurological outcomes and survival improvement appears to be relatively lengthened.26 The reasons for delayed EMS response times in areas with a high regional deprivation index are likely multifactorial. In addition to primary geographical factors, delays in EMS activation, unclear addresses in underserved neighbourhoods, delays caused by elevator use in high-rise buildings, and ambulance resource allocation should also be considered.
Ironically, despite the fact that BCPR is highly effective in areas with a high deprivation index, the BCPR rate was lowest in areas with a high deprivation index. Increasing the BCPR rate in regions with a high deprivation index will contribute to improved clinical outcomes for individuals who experience OHCA. Sasson et al. investigated the barriers to CPR training in neighbourhoods with low rates of BCPR. These included the financial costs of training, geographical accessibility of training, childcare implications of attending training, and prioritising attending CPR training among the daily challenges of life, including work and family committments.27 A previous study of barriers to bystander CPR in deprived communities found that environmental barriers included concerns about the safety of the physical environment in which people lived, and fear of reprisal from gangs or the police.28 Although some low-cost mandatory CPR training courses have been successful in reaching disadvantaged communities, tailored strategies are still required to address the previously identified barriers in highly deprived communities.29, 30
The present study had several limitations. The primary limitation of this study lies in the use of the regional deprivation score as a proxy for individual-level SES and health risks. Assigning aggregated neighbourhood-level measures to individuals inherently risks ecological fallacy.31 Administrative units often exhibit substantial intra-regional heterogeneity, which may risk considerable SES disparities among residents living within the same area.32 To minimise residual confounding and provide a more comprehensive assessment of individual-risk, our models were adjusted for individual-level SES indicators (such as, health insurance status33) alongside regional deprivation indices.34 The second of which was its observational design, thus precluding us from drawing causal inferences. Third, our data did not collect information on bystander characteristics, such as previous CPR training, which would have enabled us to perform more robust adjustments. Finally, our findings may not be generalizable to other communities and EMS settings. It is possible that the association between BCPR and survival due to deprivation is even more divergent in regions with a wider gradient of deprivation.
Conclusion
This study found disparities in BCPR rates and clinical outcomes across deprivation groups among individuals who experienced OHCA. The association between BCPR and favourable neurological outcome was larger in the high-deprivation group than that in the moderate- or low-deprivation groups. Multifaceted interventions are needed to increase BCPR rates in highly deprived regions.
CRediT authorship contribution statement
Kyoung-Jun Song: Writing – review & editing, Writing – original draft. Sang Do Shin: Supervision, Resources, Project administration, Funding acquisition. Ki Jeong Hong: Supervision, Funding acquisition, Data curation. Jeong Ho Park: Methodology, Investigation, Conceptualization. Yeongho Choi: Validation, Supervision, Data curation. Seulki Choi: Validation, Supervision, Data curation. Ki Ok Ahn: Writing – review & editing, Methodology, Conceptualization.
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Footnotes
Supplementary material to this article can be found online at https://doi.org/10.1016/j.resplu.2026.101484.
Appendix A. Supplementary material
The following are the Supplementary material to this article:
STROBE Statement—Checklist of items that should be included in reports of cross-sectional studies.
References
- 1.Cummins R.O., Ornato J.P., Thies W.H., Pepe P.E. Improving survival from sudden cardiac arrest: the “chain of survival” concept. A statement for health professionals from the Advanced Cardiac Life Support Subcommittee and the Emergency Cardiac Care Committee, American Heart Association. Circulation. 1991;83:1832–1847. doi: 10.1161/01.cir.83.5.1832. [DOI] [PubMed] [Google Scholar]
- 2.Yan S., Gan Y., Jiang N., Wang R., Chen Y., Luo Z., et al. The global survival rate among adult out-of-hospital cardiac arrest patients who received cardiopulmonary resuscitation: a systematic review and meta-analysis. Crit Care. 2020;24:61. doi: 10.1186/s13054-020-2773-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Shimai R., Ouchi S., Miyazaki T., Hirabayashi K., Abe H., Yabe K., et al. Impact of bystander cardiopulmonary resuscitation on neurological outcomes in patients undergoing veno-arterial extracorporeal membrane oxygenation. Int J Emerg Med. 2023;16:8. doi: 10.1186/s12245-023-00485-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Lee J., Lee W., Lee Y.J., Sim H., Lee W.K. Effectiveness of bystander cardiopulmonary resuscitation in improving the survival and neurological recovery of patients with out-of-hospital cardiac arrest: a nationwide patient cohort study. PLoS One. 2020;15 doi: 10.1371/journal.pone.0243757. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.van Nieuwenhuizen B.P., Oving I., Kunst A.E., Daams J., Blom M.T., Tan H.L. Socio-economic differences in incidence, bystander cardiopulmonary resuscitation and survival from out-of-hospital cardiac arrest: a systematic review. Resuscitation. 2019;141:44–62. doi: 10.1016/j.resuscitation.2019.05.018. [DOI] [PubMed] [Google Scholar]
- 6.Carstairs V., Morris R. Deprivation: explaining differences in mortality between Scotland and England and Wales. BMJ. 1989;299:886–889. doi: 10.1136/bmj.299.6704.886. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Townsend P. Allen Lane and Penguin Books; 1979. Poverty in the United Kingdom. [Google Scholar]
- 8.Townsend P. Deprivation. J Soc Policy. 1987;16:125–146. [Google Scholar]
- 9.Macintyre S., Ellaway A., Cummins S. Place effects on health: how can we conceptualise, operationalise and measure them? Soc Sci Med. 2002;55:125–139. doi: 10.1016/s0277-9536(01)00214-3. [DOI] [PubMed] [Google Scholar]
- 10.Brown T.P., Booth S., Hawkes C.A., Soar J., Mark J., Mapstone J., et al. Characteristics of neighbourhoods with high incidence of out-of-hospital cardiac arrest and low bystander cardiopulmonary resuscitation rates in England. Eur Heart J Qual Care Clin Outcomes. 2019;5:51–62. doi: 10.1093/ehjqcco/qcy026. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Huebinger R., Jarvis J., Schulz K., Persse D., Chan H.K., Miramontes D., et al. Community disparities in out of hospital cardiac arrest care and outcomes in Texas. Resuscitation. 2021;163:101–107. doi: 10.1016/j.resuscitation.2021.03.021. [DOI] [PubMed] [Google Scholar]
- 12.Lee S.Y., Song K.J., Shin S.D., Ro Y.S., Hong K.J., Kim Y.T., et al. A disparity in outcomes of out-of-hospital cardiac arrest by community socioeconomic status: a ten-year observational study. Resuscitation. 2018;126:130–136. doi: 10.1016/j.resuscitation.2018.02.025. [DOI] [PubMed] [Google Scholar]
- 13.Abdulhay N.M., Totolos K., McGovern S., Hewitt N., Bhardwaj A., Buckler D.G., et al. Socioeconomic disparities in layperson CPR training within a large U.S. city. Resuscitation. 2019;141:13–18. doi: 10.1016/j.resuscitation.2019.05.038. [DOI] [PubMed] [Google Scholar]
- 14.Blewer A.L., Ibrahim S.A., Leary M., Dutwin D., McNally B., Anderson M.L., et al. Cardiopulmonary resuscitation training disparities in the United States. J Am Heart Assoc. 2017;6 doi: 10.1161/JAHA.117.006124. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Park J.H., Choi Y., Ro Y.S., Song K.J., Shin S.D. Establishing the Korean out-of-hospital cardiac arrest registry (KOHCAR) Resusc Plus. 2023;17 doi: 10.1016/j.resplu.2023.100529. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Bray J.E., Grasner J.T., Nolan J.P., Iwami T., Ong M.E.H., Finn J., et al. Cardiac arrest and cardiopulmonary resuscitation outcome reports: 2024 update of the Utstein out-of-hospital cardiac arrest registry template. Circulation. 2024;150:e203–e223. doi: 10.1161/CIR.0000000000001243. [DOI] [PubMed] [Google Scholar]
- 17.Lee S.C.L., Mao D.R., Ng Y.Y., Leong B.S., Supasaovapak J., Gaerlan F.J., et al. Emergency medical dispatch services across Pan-Asian countries: a web-based survey. BMC Emerg Med. 2020;20:1. doi: 10.1186/s12873-019-0299-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Kim Y.T., Shin S.D., Hong S.O., Ahn K.O., Ro Y.S., Song K.J., et al. Effect of national implementation of utstein recommendation from the global resuscitation alliance on ten steps to improve outcomes from out-of-hospital cardiac arrest: a ten-year observational study in Korea. BMJ Open. 2017;7 doi: 10.1136/bmjopen-2017-016925. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Hong J., Yi S., Yoon T. The impact of the COVID-19 pandemic on life expectancy by the level of area deprivation in South Korea. Front Public Health. 2023;11 doi: 10.3389/fpubh.2023.1215914. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Kim E., Lee H., Lloyd-Jones D., Ko Y.G., Kim B.G., Kim H.C. Area deprivation and premature cardiovascular mortality: a nationwide population-based study in South Korea. BMJ Public Health. 2024;2 doi: 10.1136/bmjph-2023-000877. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Huded C.P., Dalton J.E., Kumar A., Krieger N.I., Kassis N., Phelan M., et al. Relationship of neighborhood deprivation and outcomes of a comprehensive ST-segment-elevation myocardial infarction protocol. J Am Heart Assoc. 2021;10 doi: 10.1161/JAHA.121.024540. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Cheek L., Schmicker R.H., Crowe R., Goren E., West A., McMullan J., et al. Rurality and area deprivation and outcomes after out-of-hospital cardiac arrest. JAMA Netw Open. 2025;8 doi: 10.1001/jamanetworkopen.2025.3435. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Edgren E., Hedstrand U., Kelsey S., Sutton-Tyrrell K., Safar P. Assessment of neurological prognosis in comatose survivors of cardiac arrest. BRCT I Study Group. Lancet. 1994;343:1055–1059. doi: 10.1016/s0140-6736(94)90179-1. [DOI] [PubMed] [Google Scholar]
- 24.Owen P., Hannah J., King P., Deakin C., Plumb J., Jackson A.I.R. Is there an association between 30-day mortality from out-of-hospital cardiac arrest (OHCA) and deprivation levels within Hampshire? A retrospective cohort study. Resusc Plus. 2025;22 doi: 10.1016/j.resplu.2025.100898. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Sasson C., Rogers M.A., Dahl J., Kellermann A.L. 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]
- 26.Moon S., Ryoo H.W., Ahn J.Y., Lee D.E., Shin S.D., Park J.H. Association of response time interval with neurological outcomes after out-of-hospital cardiac arrest according to bystander CPR. Am J Emerg Med. 2020;38:1760–1766. doi: 10.1016/j.ajem.2020.05.108. [DOI] [PubMed] [Google Scholar]
- 27.Sasson C., Haukoos J.S., Bond C., Rabe M., Colbert S.H., King R., et al. 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]
- 28.Dobbie F., Uny I., Eadie D., Duncan E., Stead M., Bauld L., et al. Barriers to bystander CPR in deprived communities: Findings from a qualitative study. PLoS One. 2020;15 doi: 10.1371/journal.pone.0233675. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Jensen T.W., Folke F., Andersen M.P., Blomberg S.N., Jørgensen A.J., Holgersen M.G., et al. Socio-demographic characteristics of basic life support course participants in Denmark. Resuscitation. 2021;170:167–177. doi: 10.1016/j.resuscitation.2021.11.006. [DOI] [PubMed] [Google Scholar]
- 30.Health education forum primary and secondary school curriculum; 2008. [accessed on 9 March 2026]. Available online: http://www.gsy.or.kr/main/ko/index.html.
- 31.Subramanian S.V., et al. Encyclopedia of epidemiology. 2009. Ecological fallacy; pp. 305–308. [Google Scholar]
- 32.Krieger N., Chen J.T., Waterman P.D., Soobader M.J., Subramanian S.V., Carson R. Geocoding and monitoring of US socioeconomic inequalities in mortality and cancer incidence: does the choice of area-based measure and geographic level matter?: the Public Health Disparities Geocoding Project. Am J Epidemiol. 2002;156:471–482. doi: 10.1093/aje/kwf068. [DOI] [PubMed] [Google Scholar]
- 33.Kim T.H., Ro Y.S., Shin S.D., Song K.J., Hong K.J., Park J.H., et al. Association of health insurance with post-resuscitation care and neurological outcomes after return of spontaneous circulation in out-of-hospital cardiac arrest patients in Korea. Resuscitation. 2019;135:176–182. doi: 10.1016/j.resuscitation.2018.12.023. [DOI] [PubMed] [Google Scholar]
- 34.Krieger N., Chen J.T., Waterman P.D., Soobader M.J., Subramanian S.V., Carson R. Choosing area based socioeconomic measures to monitor social inequalities in low birth weight and childhood lead poisoning: the Public Health Disparities Geocoding Project (US) J Epidemiol Community Health. 2003;57:186–199. doi: 10.1136/jech.57.3.186. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
STROBE Statement—Checklist of items that should be included in reports of cross-sectional studies.