Bystander cardiopulmonary resuscitation and long-term outcomes in out-of-hospital cardiac arrest according to location of arrest

Bystander cardiopulmonary resuscitation and long-term outcomes in out-of-hospital cardiac arrest... Abstract Aims Bystander cardiopulmonary resuscitation (CPR) has increased in several countries following nationwide initiatives to facilitate bystander resuscitative efforts in out-of-hospital cardiac arrest (OHCA). We examined the importance of public or residential location of arrest on temporal changes in bystander CPR and outcomes. Methods and results From the nationwide Danish Cardiac Arrest Registry, all OHCAs from 2001 to 2014 of presumed cardiac cause and between 18 and 100 years of age were identified. Arrests witnessed by emergency medical services personnel were excluded. Of 25 505 OHCAs, 26.4% (n = 6738) and 73.6% (n = 18 767) were in public and residential locations, respectively. Bystander CPR increased during 2001–2014 in both locations: from 36.4% [95% confidence interval (CI) 30.6–42.6%] to 83.1% (95% CI 80.0–85.8%) in public (P < 0.001) and from 16.0% (95% CI 13.2–19.3%) to 61.0% (95% CI 58.7–63.2%) in residential locations (P < 0.001). Concurrently, 30-day survival increased in public from 6.4% (95% CI 4.0–10.0%) to 25.2% (95% CI 22.1–28.7%) (P < 0.001), and in residential from 2.9% (95% CI 1.8–4.5%) to 10.0% (95% CI 8.7–11.4%) (P < 0.001). Among 2281 30-day survivors, 1-year risk of anoxic brain damage/nursing home admission during 2001–2014 decreased from 18.8% (95% CI 6.6–43.0%) to 6.8% (95% CI 3.9–11.8%) in public (P < 0.001), whereas the corresponding change was insignificant in residential locations from 11.8% (95% CI 3.3–34.3) to 17.6% (95% CI 12.7–23.9%) (P = 0.52). Conclusion During 2001–2014, bystander CPR and 30-day survival more than doubled in both public and residential OHCA locations. A significant decrease in anoxic brain damage/nursing home admission was observed among 30-day survivors in public, but not among survivors from residential OHCAs. View largeDownload slide View largeDownload slide OHCA, Bystander CPR, Long-term outcomes, Location of arrest See page 319 for the editorial comment on this article (doi: 10.1093/eurheartj/ehy911) One-sentence summary Nationwide initiatives to improve out-of-hospital cardiac arrest pre-hospital care seem equally beneficial on improving bystander cardiopulmonary resuscitation and survival rates in public and residential location of arrest. Introduction Early bystander cardiopulmonary resuscitation (CPR) is pivotal for survival and long-term outcomes in survivors from out-of-hospital cardiac arrest (OHCA).1–3 Consequently, numerous countries have made national efforts to improve pre-hospital resuscitative efforts, including CPR training, which has led to increased CPR rates and a concomitant increase in survival.4–6 However, survival from OHCA varies greatly with location of arrest, and arrests in residential locations have much lower survival rates compared with arrests in public locations.7–9 It has previously been observed that OHCAs in residential locations have lower CPR rates compared with arrests elsewhere10,11; and it was recently discovered that bystander defibrillation rates from onsite automated external defibrillators (AEDs) increased in public locations from 2001 to 2012, while bystander defibrillation remained unchanged and limited in residential locations.12 In Denmark, bystander CPR and survival rates more than doubled between 2001 and 2010 following several national initiatives to improve cardiac arrest management.4 Yet it remains unknown how these improvements might differ according to location of arrest. The aims of this study were: (i) to assess time trends in bystander CPR and long-term outcomes in public and residential location of arrest; (ii) to identify pre-arrest factors associated with not receiving bystander CPR separately for each location. Methods Study setting This study took place in Denmark with a population of approximately 5.6 million inhabitants and demographics varying from urban to rural areas. The Danish Cardiac Arrest Registry All OHCAs in Denmark where resuscitative efforts have been initiated by bystander or emergency medical services (EMS) personnel have been reported in the Danish Cardiac Arrest Registry (http://www.isrctn.com/ISRCTN14261134) since June 2001. Out-of-hospital cardiac arrest cases with obvious late signs of death without a resuscitation attempt are not included in the registry. Case ascertainment is close to complete, as EMS is dispatched to all emergency calls in Denmark and it is mandatory for the dispatched EMS personnel to complete a short report form with detailed information describing the OHCA case according to the Utstein criteria. Location of arrest was divided in public (public spaces, public areas, e.g. hallways and roads, places in the nature, shops, workplaces, and institutions) and residential (private homes) location of arrest. Out-of-hospital cardiac arrests of presumed cardiac cause were identified using diagnosis codes from death certificates and discharge diagnoses.4 The Danish nationwide registries Danish healthcare data are extensively recorded in multiple national registries. Cross-referencing of the registries is possible, as Danish citizens are issued a unique and permanent civil registration number at birth or immigration. Information on sex, age, and vital status are recorded in the Danish Civil Registration System. All hospitalizations are recorded in the National Patient Registry and coded according to the 10th edition of the International Classification of Diseases (ICD10). The Danish Register of Medicinal Product Statistics contains recordings of all redeemed prescriptions in Denmark classified according to the Anatomical Therapeutic Classification system. Income and civil status are recorded on an annual level by Statistics Denmark as well as information on nursing home admissions. Deaths and causes of death are recorded in the Danish Cause of Death Register. Study population, comorbidity and pharmacotherapy The study population consisted of all OHCAs 18–100 years of age in Denmark from June 2001 to 2014, excluding patients with arrest of presumed non-cardiac cause and arrests witnessed by EMS personnel. Information on comorbidity was obtained from discharge diagnoses up to 5 years prior to OHCA, with exception of diabetes, where the redemption of a prescription for antidiabetic medication up to 180 days prior to cardiac arrest defined diabetes. Anoxic brain damage was established from the diagnosis code G93.1 at any time during the 1-year follow-up period, as previously described.3 Daytime, evening, and night-time was defined as 8 AM to 3.59 PM, 4 PM to 11.59 PM, and 12 PM to 7.59 AM, respectively. Income was based on household earnings the year before the cardiac arrest and indexed according to 2010. Income group was categorized in low, medium and high based on the 1st and 3rd income quartile. Study outcome The primary study outcome was bystander CPR according to location of arrest. Secondary outcomes were 30-day survival and 1-year survival, and in 30-day survivors also all-cause mortality and the cumulative risk of a composite endpoint of anoxic brain damage or admission to nursing home during a 1-year follow-up period according to location of arrest and bystander CPR. Statistics Baseline characteristics were summarized according to bystander CPR and by public or residential location of arrest. Continuous variables were compared using Kruskal–Wallis test and categorical values were compared using χ2 test. Linear calendar time trends were examined using univariate logistic regression for bystander CPR and survival, univariate Fine-Gray regression for anoxic brain damage or nursing home admission, and univariate Cox regression for all-cause mortality using calendar year of cardiac arrest as a numeric variable. The association between bystander CPR and pre-arrest conditions were analysed by multivariable logistic regression separately for each location of arrest. Risks of anoxic brain damage or nursing home admission and all-cause mortality in 30-day survivors were analysed using multivariable cause-specific Cox regression analyses adjusted for age, sex, witnessed status, time from arrest to EMS arrival, Charlson Comorbidity Index, and calendar year (2001–2005, 2006–2011, 2012–2014). When analysing the risk of the combined endpoint anoxic brain damage or nursing home admission, death was a competing risk. In the multivariable logistic regression analyses and the multivariable cause-specific Cox regression analyses, missing data were handled by multiple imputations on 200 datasets by the smcfcs package separately for public and residential arrests.3 For complete case analyses, see Supplementary material online. Analyses were carried out using SAS software version 9.4 (SAS institute Inc., NC, USA) and R version 3.3.3 (R Development Core Team).13 Statistical significance was defined as a two-sided P-value < 0.05. Ethics This project was approved by The Danish Data Protection Agency (Ref.no. 2007-58-0015, local ref.no. GEH-2014-017, I-Suite.nr. 02735). Ethical approval is not required for retrospective registry studies in Denmark. Results A total of 25 505 first-time OHCA patients 18–100 years of age from 2001 to 2014 were included in the study (Figure 1). Most arrests occurred in residential locations: 73.6% (n = 18 767) vs. 26.4% (n = 6738) in public locations. Baseline characteristics are presented in Table 1 and Supplementary material online, Table S1. Table 1 Characteristics of OHCA according to bystander CPR in public and residential location of arrest Characteristics Public (N = 6738a) Residential (N = 18 767a) No bystander CPR (n = 2471, 38.8%) Bystander CPR (n = 3901, 61.2%) P-values No bystander CPR (n = 11 166, 62.2%) Bystander CPR (n = 6799, 37.8%) P-values Age (years), median (IQR) 70 (60–80) 67 (57–77) <0.001 74 (65–82) 70 (61–80) <0.001 Male sex, n (%) 1769 (71.6) 3054 (78.3) <0.001 7201 (64.5) 4430 (65.2) <0.37 Time of day, n (%) <0.001 <0.001  Day 1294 (53.3) 2264 (58.5) 4355 (39.9) 2702 (40.3)  Evening 767 (31.6) 1204 (31.1) 3466 (31.7) 2397 (35.7)  Night 367 (15.1) 404 (10.4) 3105 (28.4) 1607 (24.0)   Data missing, n (%) 43 (1.7) 29 (0.7) 240 (2.2) 93 (1.4) Witnessed arrest, n (%) 1422 (57.9) 2858 (73.6) <0.001 4766 (42.9) 3759 (55.5) <0.001  Data missing, n (%) 13 (0.5) 17 (0.4) 50 (0.4) 29 (0.4) First defibrillation by bystanders, n (%) 14 (0.6) 382 (10.6) <0.001 18 (0.2) 143 (2.2) <0.001  Data missing, n (%) 168 (6.8) 283 (7.3) 477 (4.3) 215 (3.2) First defibrillation by EMS, n (%) 1080 (47.9) 1944 (53.7) <0.001 3563 (33.8) 2552 (38.8) <0.001  Data missing, n (%) 217 (8.8) 284 (7.3) 640 (5.7) 229 (3.4) Shockable first rhythm, n (%) 706 (29.8) 1954 (51.9) <0.001 1788 (16.8) 1946 (29.7) <0.001  Data missing, n (%) 103 (4.2) 138 (3.5) 501 (4.5) 246 (3.6) Time from arrest to first rhythm analysis by the EMS, median(IQR) 11 (7–16) 10 (5–16) <0.001 12 (7–21) 13 (8–19) <0.001  Data missing, n (%) 354 (14.3) 599 (15.4) 1559 (14.0) 967 (14.2) Income group, n (%) <0.001 <0.001  Low 747 (30.4) 790 (20.4) 2548 (22.9) 1365 (20.1)  Medium 1133 (46.1) 1724 (44.4) 6277 (56.3) 3391 (50.0)  High 579 (23.5) 1366 (35.2) 2322 (20.8) 2031 (29.9)   Data missing, n (%) 12 (0.5) 21 (0.5) 19 (0.2) 12 (0.2) CAG within 30 daysb 134 (71.7) 850 (84.0) <0.001 248 (67.8) 553 (81.1) <0.001 PCI within 30 daysb 78 (41.7) 402 (39.7) 0.67 125 (34.2) 261 (38.3) 0.21 ICD within 30 daysb 42 (22.5) 468 (46.2) <0.001 110 (30.1) 293 (43.0) <0.001 Comorbidity  Ischaemic heart disease including myocardial infarction, n (%) 629 (25.5) 1309 (33, 6) <0.001 2123 (19.0) 1240 (18.2) 0.05  Congestive heart failure, n (%) 387 (15.7) 604 (15.5) 0.88 1982 (17.8) 1174 (17.3) 0.42  Cerebral vascular disease, n (%) 237 (9.6) 322 (8.3) 0.07 987 (8.8) 625 (9.2) 0.44  Diabetes, n (%) 324 (13.1) 501 (12.8) 0.78 1705 (15.3) 1143 (16.8) 0.006  COPD, n (%) 214 (8.7) 264 (6.8) 0.006 1560 (14.0) 813 (12.0) <0.001  Malignancy, n (%) 141 (5.7) 189 (4.8) 0.15 844 (7.6) 481 (7.1) 0.24  Liver diseases, n (%) 27 (1.1) 21 (0.5) 0.02 162 (1.5) 104 (1.5) 0.72  Renal diseases, n (%) 77 (3.1) 136 (3.5) 0.47 486 (4.4) 346 (5.1) 0.03  Psychiatric diseases, n (%) 331 (13.4) 391 (10.0) <0.001 1079 (9.7) 717 (10.5) 0.06  Charlson Comorbidity Index group, n (%) 0.02 0.70   0 1442 (58.4) 2159 (55.3) 5864 (52.5) 3581 (52.7)   1 468 (18.9) 844 (21.6) 2105 (18.9) 1249 (18.4)   >1 561 (22.7) 898 (23.0) 3197 (28.6) 1969 (29.0) Cohabitant  Age of cohabitant, median (IQR) 66 (58, 74) 63 (54, 71) <0.001 71 (63, 77) 65 (57, 74) <0.001   Data missing, n (%) 13 (0.52) 10 (0.26) 31 (0.28) 29 (0.43)  Male cohabitant, n (%) 197 (17.6) 287 (13.7) 0.004 1835 (26.1) 1080 (25.7) 0.63   Data missing, n (%) 13 (0.52) 10 (0.26) 31 (0.28) 29 (0.43)  Charlson Comorbidity Index group in cohabitant, n (%) 0.001 <0.001   0 941 (83.2) 1855 (87.7) 9834 (88.1) 6176 (90.8)   1 67 (5.9) 103 (4.9) 532 (4.8) 227 (3.3)   >1 123 (10.9) 157 (7.4) 800 (7.2) 396 (5.8) Outcomes  Thirty-day survivor, n (%) 187 (7.6) 1012 (25.9) <0.001 366 (3.3) 682 (10.0) <0.001  One-year survivor, n (%) 172 (7.0) 956 (24.5) <0.001 312 (2.8) 614 (9.0) <0.001 Characteristics Public (N = 6738a) Residential (N = 18 767a) No bystander CPR (n = 2471, 38.8%) Bystander CPR (n = 3901, 61.2%) P-values No bystander CPR (n = 11 166, 62.2%) Bystander CPR (n = 6799, 37.8%) P-values Age (years), median (IQR) 70 (60–80) 67 (57–77) <0.001 74 (65–82) 70 (61–80) <0.001 Male sex, n (%) 1769 (71.6) 3054 (78.3) <0.001 7201 (64.5) 4430 (65.2) <0.37 Time of day, n (%) <0.001 <0.001  Day 1294 (53.3) 2264 (58.5) 4355 (39.9) 2702 (40.3)  Evening 767 (31.6) 1204 (31.1) 3466 (31.7) 2397 (35.7)  Night 367 (15.1) 404 (10.4) 3105 (28.4) 1607 (24.0)   Data missing, n (%) 43 (1.7) 29 (0.7) 240 (2.2) 93 (1.4) Witnessed arrest, n (%) 1422 (57.9) 2858 (73.6) <0.001 4766 (42.9) 3759 (55.5) <0.001  Data missing, n (%) 13 (0.5) 17 (0.4) 50 (0.4) 29 (0.4) First defibrillation by bystanders, n (%) 14 (0.6) 382 (10.6) <0.001 18 (0.2) 143 (2.2) <0.001  Data missing, n (%) 168 (6.8) 283 (7.3) 477 (4.3) 215 (3.2) First defibrillation by EMS, n (%) 1080 (47.9) 1944 (53.7) <0.001 3563 (33.8) 2552 (38.8) <0.001  Data missing, n (%) 217 (8.8) 284 (7.3) 640 (5.7) 229 (3.4) Shockable first rhythm, n (%) 706 (29.8) 1954 (51.9) <0.001 1788 (16.8) 1946 (29.7) <0.001  Data missing, n (%) 103 (4.2) 138 (3.5) 501 (4.5) 246 (3.6) Time from arrest to first rhythm analysis by the EMS, median(IQR) 11 (7–16) 10 (5–16) <0.001 12 (7–21) 13 (8–19) <0.001  Data missing, n (%) 354 (14.3) 599 (15.4) 1559 (14.0) 967 (14.2) Income group, n (%) <0.001 <0.001  Low 747 (30.4) 790 (20.4) 2548 (22.9) 1365 (20.1)  Medium 1133 (46.1) 1724 (44.4) 6277 (56.3) 3391 (50.0)  High 579 (23.5) 1366 (35.2) 2322 (20.8) 2031 (29.9)   Data missing, n (%) 12 (0.5) 21 (0.5) 19 (0.2) 12 (0.2) CAG within 30 daysb 134 (71.7) 850 (84.0) <0.001 248 (67.8) 553 (81.1) <0.001 PCI within 30 daysb 78 (41.7) 402 (39.7) 0.67 125 (34.2) 261 (38.3) 0.21 ICD within 30 daysb 42 (22.5) 468 (46.2) <0.001 110 (30.1) 293 (43.0) <0.001 Comorbidity  Ischaemic heart disease including myocardial infarction, n (%) 629 (25.5) 1309 (33, 6) <0.001 2123 (19.0) 1240 (18.2) 0.05  Congestive heart failure, n (%) 387 (15.7) 604 (15.5) 0.88 1982 (17.8) 1174 (17.3) 0.42  Cerebral vascular disease, n (%) 237 (9.6) 322 (8.3) 0.07 987 (8.8) 625 (9.2) 0.44  Diabetes, n (%) 324 (13.1) 501 (12.8) 0.78 1705 (15.3) 1143 (16.8) 0.006  COPD, n (%) 214 (8.7) 264 (6.8) 0.006 1560 (14.0) 813 (12.0) <0.001  Malignancy, n (%) 141 (5.7) 189 (4.8) 0.15 844 (7.6) 481 (7.1) 0.24  Liver diseases, n (%) 27 (1.1) 21 (0.5) 0.02 162 (1.5) 104 (1.5) 0.72  Renal diseases, n (%) 77 (3.1) 136 (3.5) 0.47 486 (4.4) 346 (5.1) 0.03  Psychiatric diseases, n (%) 331 (13.4) 391 (10.0) <0.001 1079 (9.7) 717 (10.5) 0.06  Charlson Comorbidity Index group, n (%) 0.02 0.70   0 1442 (58.4) 2159 (55.3) 5864 (52.5) 3581 (52.7)   1 468 (18.9) 844 (21.6) 2105 (18.9) 1249 (18.4)   >1 561 (22.7) 898 (23.0) 3197 (28.6) 1969 (29.0) Cohabitant  Age of cohabitant, median (IQR) 66 (58, 74) 63 (54, 71) <0.001 71 (63, 77) 65 (57, 74) <0.001   Data missing, n (%) 13 (0.52) 10 (0.26) 31 (0.28) 29 (0.43)  Male cohabitant, n (%) 197 (17.6) 287 (13.7) 0.004 1835 (26.1) 1080 (25.7) 0.63   Data missing, n (%) 13 (0.52) 10 (0.26) 31 (0.28) 29 (0.43)  Charlson Comorbidity Index group in cohabitant, n (%) 0.001 <0.001   0 941 (83.2) 1855 (87.7) 9834 (88.1) 6176 (90.8)   1 67 (5.9) 103 (4.9) 532 (4.8) 227 (3.3)   >1 123 (10.9) 157 (7.4) 800 (7.2) 396 (5.8) Outcomes  Thirty-day survivor, n (%) 187 (7.6) 1012 (25.9) <0.001 366 (3.3) 682 (10.0) <0.001  One-year survivor, n (%) 172 (7.0) 956 (24.5) <0.001 312 (2.8) 614 (9.0) <0.001 a Including 366 public arrests and 802 residential arrests with missing values on bystander CPR. Baseline characteristics are presented in Supplementary material online, Table S2. b 30-Day survivors. CAG, coronary angiography; COPD, chronic obstructive pulmonary disease; CPR, cardiopulmonary resuscitation; EMS, emergency medical services; ICD, implantable cardiac defibrillator; IQR, interquartile range; OHCA, out-of-hospital cardiac arrest; PCI, percutaneous coronary intervention. Table 1 Characteristics of OHCA according to bystander CPR in public and residential location of arrest Characteristics Public (N = 6738a) Residential (N = 18 767a) No bystander CPR (n = 2471, 38.8%) Bystander CPR (n = 3901, 61.2%) P-values No bystander CPR (n = 11 166, 62.2%) Bystander CPR (n = 6799, 37.8%) P-values Age (years), median (IQR) 70 (60–80) 67 (57–77) <0.001 74 (65–82) 70 (61–80) <0.001 Male sex, n (%) 1769 (71.6) 3054 (78.3) <0.001 7201 (64.5) 4430 (65.2) <0.37 Time of day, n (%) <0.001 <0.001  Day 1294 (53.3) 2264 (58.5) 4355 (39.9) 2702 (40.3)  Evening 767 (31.6) 1204 (31.1) 3466 (31.7) 2397 (35.7)  Night 367 (15.1) 404 (10.4) 3105 (28.4) 1607 (24.0)   Data missing, n (%) 43 (1.7) 29 (0.7) 240 (2.2) 93 (1.4) Witnessed arrest, n (%) 1422 (57.9) 2858 (73.6) <0.001 4766 (42.9) 3759 (55.5) <0.001  Data missing, n (%) 13 (0.5) 17 (0.4) 50 (0.4) 29 (0.4) First defibrillation by bystanders, n (%) 14 (0.6) 382 (10.6) <0.001 18 (0.2) 143 (2.2) <0.001  Data missing, n (%) 168 (6.8) 283 (7.3) 477 (4.3) 215 (3.2) First defibrillation by EMS, n (%) 1080 (47.9) 1944 (53.7) <0.001 3563 (33.8) 2552 (38.8) <0.001  Data missing, n (%) 217 (8.8) 284 (7.3) 640 (5.7) 229 (3.4) Shockable first rhythm, n (%) 706 (29.8) 1954 (51.9) <0.001 1788 (16.8) 1946 (29.7) <0.001  Data missing, n (%) 103 (4.2) 138 (3.5) 501 (4.5) 246 (3.6) Time from arrest to first rhythm analysis by the EMS, median(IQR) 11 (7–16) 10 (5–16) <0.001 12 (7–21) 13 (8–19) <0.001  Data missing, n (%) 354 (14.3) 599 (15.4) 1559 (14.0) 967 (14.2) Income group, n (%) <0.001 <0.001  Low 747 (30.4) 790 (20.4) 2548 (22.9) 1365 (20.1)  Medium 1133 (46.1) 1724 (44.4) 6277 (56.3) 3391 (50.0)  High 579 (23.5) 1366 (35.2) 2322 (20.8) 2031 (29.9)   Data missing, n (%) 12 (0.5) 21 (0.5) 19 (0.2) 12 (0.2) CAG within 30 daysb 134 (71.7) 850 (84.0) <0.001 248 (67.8) 553 (81.1) <0.001 PCI within 30 daysb 78 (41.7) 402 (39.7) 0.67 125 (34.2) 261 (38.3) 0.21 ICD within 30 daysb 42 (22.5) 468 (46.2) <0.001 110 (30.1) 293 (43.0) <0.001 Comorbidity  Ischaemic heart disease including myocardial infarction, n (%) 629 (25.5) 1309 (33, 6) <0.001 2123 (19.0) 1240 (18.2) 0.05  Congestive heart failure, n (%) 387 (15.7) 604 (15.5) 0.88 1982 (17.8) 1174 (17.3) 0.42  Cerebral vascular disease, n (%) 237 (9.6) 322 (8.3) 0.07 987 (8.8) 625 (9.2) 0.44  Diabetes, n (%) 324 (13.1) 501 (12.8) 0.78 1705 (15.3) 1143 (16.8) 0.006  COPD, n (%) 214 (8.7) 264 (6.8) 0.006 1560 (14.0) 813 (12.0) <0.001  Malignancy, n (%) 141 (5.7) 189 (4.8) 0.15 844 (7.6) 481 (7.1) 0.24  Liver diseases, n (%) 27 (1.1) 21 (0.5) 0.02 162 (1.5) 104 (1.5) 0.72  Renal diseases, n (%) 77 (3.1) 136 (3.5) 0.47 486 (4.4) 346 (5.1) 0.03  Psychiatric diseases, n (%) 331 (13.4) 391 (10.0) <0.001 1079 (9.7) 717 (10.5) 0.06  Charlson Comorbidity Index group, n (%) 0.02 0.70   0 1442 (58.4) 2159 (55.3) 5864 (52.5) 3581 (52.7)   1 468 (18.9) 844 (21.6) 2105 (18.9) 1249 (18.4)   >1 561 (22.7) 898 (23.0) 3197 (28.6) 1969 (29.0) Cohabitant  Age of cohabitant, median (IQR) 66 (58, 74) 63 (54, 71) <0.001 71 (63, 77) 65 (57, 74) <0.001   Data missing, n (%) 13 (0.52) 10 (0.26) 31 (0.28) 29 (0.43)  Male cohabitant, n (%) 197 (17.6) 287 (13.7) 0.004 1835 (26.1) 1080 (25.7) 0.63   Data missing, n (%) 13 (0.52) 10 (0.26) 31 (0.28) 29 (0.43)  Charlson Comorbidity Index group in cohabitant, n (%) 0.001 <0.001   0 941 (83.2) 1855 (87.7) 9834 (88.1) 6176 (90.8)   1 67 (5.9) 103 (4.9) 532 (4.8) 227 (3.3)   >1 123 (10.9) 157 (7.4) 800 (7.2) 396 (5.8) Outcomes  Thirty-day survivor, n (%) 187 (7.6) 1012 (25.9) <0.001 366 (3.3) 682 (10.0) <0.001  One-year survivor, n (%) 172 (7.0) 956 (24.5) <0.001 312 (2.8) 614 (9.0) <0.001 Characteristics Public (N = 6738a) Residential (N = 18 767a) No bystander CPR (n = 2471, 38.8%) Bystander CPR (n = 3901, 61.2%) P-values No bystander CPR (n = 11 166, 62.2%) Bystander CPR (n = 6799, 37.8%) P-values Age (years), median (IQR) 70 (60–80) 67 (57–77) <0.001 74 (65–82) 70 (61–80) <0.001 Male sex, n (%) 1769 (71.6) 3054 (78.3) <0.001 7201 (64.5) 4430 (65.2) <0.37 Time of day, n (%) <0.001 <0.001  Day 1294 (53.3) 2264 (58.5) 4355 (39.9) 2702 (40.3)  Evening 767 (31.6) 1204 (31.1) 3466 (31.7) 2397 (35.7)  Night 367 (15.1) 404 (10.4) 3105 (28.4) 1607 (24.0)   Data missing, n (%) 43 (1.7) 29 (0.7) 240 (2.2) 93 (1.4) Witnessed arrest, n (%) 1422 (57.9) 2858 (73.6) <0.001 4766 (42.9) 3759 (55.5) <0.001  Data missing, n (%) 13 (0.5) 17 (0.4) 50 (0.4) 29 (0.4) First defibrillation by bystanders, n (%) 14 (0.6) 382 (10.6) <0.001 18 (0.2) 143 (2.2) <0.001  Data missing, n (%) 168 (6.8) 283 (7.3) 477 (4.3) 215 (3.2) First defibrillation by EMS, n (%) 1080 (47.9) 1944 (53.7) <0.001 3563 (33.8) 2552 (38.8) <0.001  Data missing, n (%) 217 (8.8) 284 (7.3) 640 (5.7) 229 (3.4) Shockable first rhythm, n (%) 706 (29.8) 1954 (51.9) <0.001 1788 (16.8) 1946 (29.7) <0.001  Data missing, n (%) 103 (4.2) 138 (3.5) 501 (4.5) 246 (3.6) Time from arrest to first rhythm analysis by the EMS, median(IQR) 11 (7–16) 10 (5–16) <0.001 12 (7–21) 13 (8–19) <0.001  Data missing, n (%) 354 (14.3) 599 (15.4) 1559 (14.0) 967 (14.2) Income group, n (%) <0.001 <0.001  Low 747 (30.4) 790 (20.4) 2548 (22.9) 1365 (20.1)  Medium 1133 (46.1) 1724 (44.4) 6277 (56.3) 3391 (50.0)  High 579 (23.5) 1366 (35.2) 2322 (20.8) 2031 (29.9)   Data missing, n (%) 12 (0.5) 21 (0.5) 19 (0.2) 12 (0.2) CAG within 30 daysb 134 (71.7) 850 (84.0) <0.001 248 (67.8) 553 (81.1) <0.001 PCI within 30 daysb 78 (41.7) 402 (39.7) 0.67 125 (34.2) 261 (38.3) 0.21 ICD within 30 daysb 42 (22.5) 468 (46.2) <0.001 110 (30.1) 293 (43.0) <0.001 Comorbidity  Ischaemic heart disease including myocardial infarction, n (%) 629 (25.5) 1309 (33, 6) <0.001 2123 (19.0) 1240 (18.2) 0.05  Congestive heart failure, n (%) 387 (15.7) 604 (15.5) 0.88 1982 (17.8) 1174 (17.3) 0.42  Cerebral vascular disease, n (%) 237 (9.6) 322 (8.3) 0.07 987 (8.8) 625 (9.2) 0.44  Diabetes, n (%) 324 (13.1) 501 (12.8) 0.78 1705 (15.3) 1143 (16.8) 0.006  COPD, n (%) 214 (8.7) 264 (6.8) 0.006 1560 (14.0) 813 (12.0) <0.001  Malignancy, n (%) 141 (5.7) 189 (4.8) 0.15 844 (7.6) 481 (7.1) 0.24  Liver diseases, n (%) 27 (1.1) 21 (0.5) 0.02 162 (1.5) 104 (1.5) 0.72  Renal diseases, n (%) 77 (3.1) 136 (3.5) 0.47 486 (4.4) 346 (5.1) 0.03  Psychiatric diseases, n (%) 331 (13.4) 391 (10.0) <0.001 1079 (9.7) 717 (10.5) 0.06  Charlson Comorbidity Index group, n (%) 0.02 0.70   0 1442 (58.4) 2159 (55.3) 5864 (52.5) 3581 (52.7)   1 468 (18.9) 844 (21.6) 2105 (18.9) 1249 (18.4)   >1 561 (22.7) 898 (23.0) 3197 (28.6) 1969 (29.0) Cohabitant  Age of cohabitant, median (IQR) 66 (58, 74) 63 (54, 71) <0.001 71 (63, 77) 65 (57, 74) <0.001   Data missing, n (%) 13 (0.52) 10 (0.26) 31 (0.28) 29 (0.43)  Male cohabitant, n (%) 197 (17.6) 287 (13.7) 0.004 1835 (26.1) 1080 (25.7) 0.63   Data missing, n (%) 13 (0.52) 10 (0.26) 31 (0.28) 29 (0.43)  Charlson Comorbidity Index group in cohabitant, n (%) 0.001 <0.001   0 941 (83.2) 1855 (87.7) 9834 (88.1) 6176 (90.8)   1 67 (5.9) 103 (4.9) 532 (4.8) 227 (3.3)   >1 123 (10.9) 157 (7.4) 800 (7.2) 396 (5.8) Outcomes  Thirty-day survivor, n (%) 187 (7.6) 1012 (25.9) <0.001 366 (3.3) 682 (10.0) <0.001  One-year survivor, n (%) 172 (7.0) 956 (24.5) <0.001 312 (2.8) 614 (9.0) <0.001 a Including 366 public arrests and 802 residential arrests with missing values on bystander CPR. Baseline characteristics are presented in Supplementary material online, Table S2. b 30-Day survivors. CAG, coronary angiography; COPD, chronic obstructive pulmonary disease; CPR, cardiopulmonary resuscitation; EMS, emergency medical services; ICD, implantable cardiac defibrillator; IQR, interquartile range; OHCA, out-of-hospital cardiac arrest; PCI, percutaneous coronary intervention. Figure 1 View largeDownload slide Population selection. CPR, cardiopulmonary resuscitation; EMS, emergency medical services. Figure 1 View largeDownload slide Population selection. CPR, cardiopulmonary resuscitation; EMS, emergency medical services. Bystander cardiopulmonary resuscitation Throughout the study period, a significant increase in the proportion of patients receiving bystander-initiated CPR was observed in both locations: from 36.4% [95% confidence interval (CI) 30.6–42.6%] to 83.1% (95% CI 80.0–85.8%) in public locations (P for trend <0.001) and from 16.0% (95% CI 13.2–19.3%) to 61.0% (95% CI 58.7–63.2%) in residential locations (P for trend <0.001) (Take home figure). Including witnessed arrest only produced a similar pattern, Supplementary material online, Figure S1. Take home figure View largeDownload slide Proportion of out-of-hospital cardiac arrest patients receiving bystander cardiopulmonary resuscitation according to location of arrest. The vertical bars represent 95% confidence interval. CPR, cardiopulmonary resuscitation. Take home figure View largeDownload slide Proportion of out-of-hospital cardiac arrest patients receiving bystander cardiopulmonary resuscitation according to location of arrest. The vertical bars represent 95% confidence interval. CPR, cardiopulmonary resuscitation. 30-Day and 1-year survival Overall, 30-day survival increased significantly during the study period in both locations: from 6.4% (95% CI 4.0–10.0%) to 25.2% (95% CI 22.1–28.7%) in public locations, and from 2.9% (95% CI 1.8–4.5%) to 10.0% (95% CI 8.7–11.4%) in residential locations (P for trend <0.001), Figure 2. One-year survival is also depicted in Figure 2 and mimicked the results of 30-day survival. In both locations, 30-day and 1-year survival increased in patients who received bystander CPR as well as in patients who did not receive bystander CPR, Supplementary material online, Figure S2. Figure 2 View largeDownload slide Thirty-day and 1-year survival in out-of-hospital cardiac arrest patients according to location of arrest. The vertical bars represent 95% confidence interval. Figure 2 View largeDownload slide Thirty-day and 1-year survival in out-of-hospital cardiac arrest patients according to location of arrest. The vertical bars represent 95% confidence interval. All-cause mortality, anoxic brain damage, and nursing home admission among 30-day survivors A total of 2348 patients survived the first 30 days following OHCA; 67 patients were excluded due to a diagnosis of brain damage or nursing home admission before OHCA. The majority of 30-day survivors had arrest in public locations (53.4%, n = 1217/2281) compared with residential locations (46.6%, n = 1064/2281). Throughout the study period, the 1-year incidences of brain damage or nursing home admission decreased from 18.8% (95% CI 6.6–43.0%) to 6.8% (95% CI 3.9–11.8%) in public arrests (P for trend < 0.001), whereas the change among 30-day survivors from arrests in residential locations from 11.8% (95% CI 3.3–34.3) to 17.6% (95% CI 12.7–23.9%) (P for trend = 0.52), was insignificant (Figure 3). Similarly, 1-year all-cause mortality incidences in 30-day survivors decreased from 12.5% (95% CI 3.5–36.0%) to 5.0% (95% CI 2.5–9.4%) in public (P for trend = 0.03), whereas the change from 29.4% (95% CI 13.3–53.1%) to 13.6% (95% CI 9.3–19.5%) in residential arrests was insignificant (P for trend = 0.75). Bystander CPR was significantly associated with a lower risk of brain damage or nursing home admission in arrests in both public and residential locations; hazard ratio (HR) 0.55 (95% CI 0.36–0.84) and HR 0.60 (95% CI 0.43–0.84), respectively, but not with all-cause mortality; HR 1.13 (95% CI 0.60–2.12) and HR 0.78 (95% CI 0.52–1.17), respectively, Figure 4 (for complete case analyses, see Supplementary material online, Figure S3). Figure 3 View largeDownload slide Anoxic brain damage or nursing home admission and all-cause mortality in 30-day survivors from out-of-hospital cardiac arrest according to location of arrest. The vertical bars represent 95% confidence interval. Figure 3 View largeDownload slide Anoxic brain damage or nursing home admission and all-cause mortality in 30-day survivors from out-of-hospital cardiac arrest according to location of arrest. The vertical bars represent 95% confidence interval. Figure 4 View largeDownload slide The associations between bystander cardiopulmonary resuscitation and brain damage/nursing home admission and death from any cause in 30-day survivors from out-of-hospital cardiac arrest according to location of arrest. *Adjusted for age, sex, year (2001–2004, 2005–2011, 2012–2014), Charlson Comorbidity Index (0, 1, >1), witnessed status, and time from arrest to arrival of EMS (<5, 5–11, >11 min). CPR, cardiopulmonary resuscitation. Figure 4 View largeDownload slide The associations between bystander cardiopulmonary resuscitation and brain damage/nursing home admission and death from any cause in 30-day survivors from out-of-hospital cardiac arrest according to location of arrest. *Adjusted for age, sex, year (2001–2004, 2005–2011, 2012–2014), Charlson Comorbidity Index (0, 1, >1), witnessed status, and time from arrest to arrival of EMS (<5, 5–11, >11 min). CPR, cardiopulmonary resuscitation. Factors associated with not receiving bystander cardiopulmonary resuscitation Factors associated with not receiving bystander CPR according to location of arrest are depicted in Figures 5 and 6 (for complete case analyses, see Supplementary material online, Figure S4). Non-witnessed arrest, high age, low income, and arrest at night were associated with a low probability of receiving bystander CPR in both locations. In public locations, female gender, liver disease, and psychiatric disease were associated with an increased risk of not receiving bystander CPR. Figure 5 View largeDownload slide Factors associated with not receiving bystander cardiopulmonary resuscitation in out-of-hospital cardiac arrest in public locations. Figure 5 View largeDownload slide Factors associated with not receiving bystander cardiopulmonary resuscitation in out-of-hospital cardiac arrest in public locations. Figure 6 View largeDownload slide Factors associated with not receiving bystander cardiopulmonary resuscitation in out-of-hospital cardiac arrest in residential locations. Figure 6 View largeDownload slide Factors associated with not receiving bystander cardiopulmonary resuscitation in out-of-hospital cardiac arrest in residential locations. We tested the robustness of our results by including year as different time intervals (2001–2007, 2008–2014) and including with similar results, Supplementary material online, Figure S5. Including only cardiac arrest patients with witnessed arrest mimicked the main results, Supplementary material online, Figure S6. Throughout the study period, the proportions receiving bystander CPR increased independently of Charlson Comorbidity Index (0, 1, >1) and age group (18–65, 66–80, >80 years) in both locations, Supplementary material online, Figures S7 and S8. Discussion In this nationwide study investigating 25 505 OHCAs during 2001–2014, bystander CPR rates more than doubled in both public and residential locations and 30-day survival rates more than tripled in both locations. Yet, bystander CPR and survival rates remained markedly lower in residential cardiac arrests. In both locations, bystander CPR was significantly associated with a decreased risk of brain damage or admission to nursing home among 30-day survivors. However, the risk of brain damage or nursing home admission in 30-day survivors decreased only significantly over time in public arrests. Non-witnessed arrest, older age, arrest during night-time, and low income were the strongest factors associated with not receiving bystander CPR in both locations; all factors were more prominent in residential locations. Out-of-hospital cardiac arrests in residential locations differ from OHCAs in public locations in several ways: patients are older, fewer have witnessed arrest, and arrest occurs more often during night-time.9,11 All of these factors are associated with lower probability of receiving bystander CPR as shown in this study. The most recent guidelines from the American Heart Association note that patients with OHCA in residential locations are much less likely to receive bystander CPR than patients with OHCA in public locations.14 Although our overall results concur with this statement, our findings of an absolute increase of 45% in the proportion of OHCA patients receiving bystander CPR in residential areas compared with an absolute increase of 47% in public locations are impressive. These results indicate that it is possible through national initiatives and campaigns not only to increase bystander CPR rates in public locations, but also in residential areas, and show that despite unfavourable conditions for resuscitation in residential locations, efforts to increase bystander CPR seem equally beneficial in public and residential locations.4 Throughout the study period, survival increased concomitantly with CPR; however, the absolute increase in survival was greater in public locations compared to residential locations. In addition, we observed an increase in survival also among patients not receiving bystander CPR, indicating that other parts in the management of OHCAs may have improved, e.g. the immediate referral of OHCA patients to cardiac centres, early heart catheterizations and percutaneous interventions, as well as targeted temperature management.15,16 Among 30-day survivors, bystander CPR was equally associated with a decreased risk of anoxic brain damage and nursing home admission in both public and residential locations. Surprisingly, anoxic brain damage or nursing home admission rates decreased only significantly over time in public arrests. This is in line with a recent study and may indicate that the unfavourable conditions connected to arrests in residential locations (e.g. unwitnessed arrest, older patients, and older and more often alone bystanders), limit the impact of improved bystander CPR rates.8,17 All-cause mortality in 30-day survivors was not associated with bystander CPR in both locations, suggesting that if patients do survive the first 30 days the probability of 1-year survival is no longer affected by bystander CPR, but on the patients’ pre-morbid conditions, while the risk of anoxic brain damage continuously is closely associated with bystander CPR. We have previously reported a marked improvement in bystander defibrillation rates during 2001–2012 in Denmark from 1.2% to 15.3% in public arrests, while bystander defibrillation remained limited and unchanged in residential arrests at 1.3%.12 In this study, we found that amongst patients receiving bystander CPR, only 2.2% were bystander defibrillated in residential locations compared with 10.6% in public locations. The low bystander defibrillation rates may partly explain the lack of improvement in rates of anoxic brain damage or nursing home admission in residential arrests, despite bystander CPR being significantly associated with lower risk of anoxic brain damage or nursing home admission in both locations. Collectively, bystanders seem to be very willing to attempt resuscitation, but other barriers prevent them from retrieving the nearest AED. However, the major improvements in bystander CPR rates in both locations are fundamental in also improving bystander defibrillation rates in both locations, which is of great importance if more patients are to survive an OHCA.18 Another key factor to survival in OHCA patients is the EMS response time, which we in line with other studies found to be 2 min longer in residential OHCAs compared with public OHCAs.2,17 One alternative way to improve survival after OHCA may be the activation of nearby laypersons by smart-phones and sending them to assist in the resuscitation attempt: one running directly to the OHCA site performing bystander CPR and the other retrieving a nearby AED, thereby potentially improving bystander CPR rates even further and most importantly minimize time to defibrillation. This has been initiated in several larger cities in Europe with promising results, particularly in cardiac arrests in residential locations, where the majority of the mobile phone activated bystanders arrived before the EMS.19,20 We found that the cohabitants to the group of patients not receiving CPR were older, which indicates that bystanders may not be capable of performing CPR, and would benefit from assisting bystanders, again particularly in residential arrests. Nonetheless, we also observed that arrests in residential locations were characterized by more un-modifiable factors, such as unwitnessed arrests, which could be associated with the very consistent difference in 20% between bystander CPR rates in public and residential locations throughout the study period. Overall, comorbidity was not associated with a significant risk of not receiving bystander CPR. Accordingly, we found an equal increase in bystander CPR independent of comorbidity and age in both locations. However, in public locations liver disease and psychiatric disease were associated with a decreased probability of bystander CPR. Liver disease may be an indicator of alcohol abuse, and alcohol abuse and psychiatric diseases are probably over-represented among home-less and destitute people. Some studies have reported that disagreeable characteristics and fear of personal health are barriers to prevent bystanders to perform CPR on strangers.21 The decreased probability of receiving bystander CPR amongst women with arrests only in public locations indicates a gender specific bystander related barrier, which is in line with other recent results.22 That the gender disparities associated with bystander CPR is limited to public locations points to gender specific biases among strangers, perhaps reluctance to remove clothes, or push hard on a woman’s chest. The marked association between the low-income group and decreased probability of bystander CPR in arrests in both locations in this study suggests a social imbalance, which concurs with previous reporting of low socioeconomic status being associated with a lower probability of bystander CPR and survival.23 As barriers related to gender or socioeconomic status could be influenced through CPR training and national CPR campaigns directly targeted to e.g. people with low socioeconomic status, these factors should be investigated further. Limitations Some of the registered cardiac arrest cases had missing data (see Supplementary material online, Table S2). We accounted for this by performing both multiple imputation analyses and complete case analyses. We found no indication that missing data had introduced bias into the complete case analyses of our study. Five patients were lost to follow-up due to emigration. We had no information on quality, timelines or length of CPR, which could have had an effect on survival and differ according to location of arrest. The time from arrest to first rhythm analysis by the EMS is only an approximation and should be interpreted cautiously. As we apart from diabetes rely on discharge diagnoses from hospitals and outpatient clinics in identifying comorbidity, patients with comorbidities who were only followed by their general practitioner and not previously admitted for such conditions were not identified. Additionally, it was not possible for us to assess whether the diagnosis of anoxic brain damage was withdrawn in the follow-up period due to improvement in neuro-cognitive functions. Due to a limited number of 30-day survivors of arrest in the beginning of our study period, our trend analyses involving anoxic brain damage or nursing home admissions should be interpreted with caution, particularly in residential arrests, see Supplementary material online, Table S3 for number of observations per year. Finally, the observational nature of the study makes any conclusion on causality impossible and we cannot identify the causal impact of each national initiative on bystander CPR and survival. Conclusion Bystander CPR and 30-day survival rates more than doubled in both residential and public locations during 2001–2014. Bystander CPR was significantly associated with decreased risk of anoxic brain damage or nursing home admission in both locations. However, amongst 30-day survivors the rates of anoxic brain damage or nursing home admission decreased only significantly during the study period in public locations. These findings indicate that nationwide initiatives to increase bystander resuscitation have equal effect in residential and public locations. To further improve survival, particularly amongst residential arrests, alternative methods may be warranted, such as layperson first-responders or targeted health campaigns. However, the less favourable patient characteristics in cardiac arrest patients in residential locations may limit CPR rates and survival rates to achieve the levels in public locations. Supplementary material Supplementary material is available at European Heart Journal online. Acknowledgements We extend our sincere thanks to the Danish Emergency Medical personnel who completed the case report forms for the Danish Cardiac Arrest Registry. Funding This work was supported by research grants from the Danish Foundation Trygfonden, who also supports the Danish Cardiac Arrest Registry with no commercial interest in the field of cardiac arrest. G.H.G. is supported by an unrestricted clinical research scholarship from the Novo Nordisk Foundation. F.F. and F.K.L. have both received unrestricted funding from The Laerdal Foundation. Conflict of interest: K.K. reports having received research grant from The Laerdal Foundation, and speaker’s honoraria from Novartis. S.M.H. reports receiving support from The Danish Heart Foundation, The Danish Foundation Trygfonden, and The Laerdal Foundation. L.K. reports receiving support from The Danish Foundation Trygfonden. References 1 Hasselqvist-Ax I , Riva G , Herlitz J , Rosenqvist M , Hollenberg J , Nordberg P , Ringh M , Jonsson M , Axelsson C , Lindqvist J , Karlsson T , Svensson L. Early cardiopulmonary resuscitation in out-of-hospital cardiac arrest . N Engl J Med 2015 ; 372 : 2307 – 2315 . Google Scholar Crossref Search ADS PubMed 2 Rajan S , Wissenberg M , Folke F , Hansen SM , Gerds TA , Kragholm K , Hansen CM , Karlsson L , Lippert FK , Køber L , Gislason GH , Torp-Pedersen C. Association of bystander cardiopulmonary resuscitation and survival according to ambulance response-times after out-of-hospital cardiac arrest . Circulation 2016 ; 134 : 2095 – 2104 . Google Scholar Crossref Search ADS PubMed 3 Kragholm K , Wissenberg M , Mortensen RN , Hansen SM , Malta Hansen C , Thorsteinsson K , Rajan S , Lippert F , Folke F , Gislason G , Køber L , Fonager K , Jensen SE , Gerds TA , Torp-Pedersen C , Rasmussen BS. Bystander efforts and 1-year outcomes in out-of-hospital cardiac arrest . N Engl J Med 2017 ; 376 : 1737 – 1747 . Google Scholar Crossref Search ADS PubMed 4 Wissenberg M , Lippert FK , Folke F , Weeke P , Hansen CM , Christensen EF , Jans H , Hansen PA , Lang-Jensen T , Olesen JB , Lindhardsen J , Fosbol EL , Nielsen SL , Gislason GH , Kober L , Torp-Pedersen C. 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 . Google Scholar Crossref Search ADS PubMed 5 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 . Google Scholar Crossref Search ADS PubMed 6 Kitamura T , Iwami T , Kawamura T , Nitta M , Nagao K , Nonogi H , Yonemoto N , Kimura T. Nationwide improvements in survival from out-of-hospital cardiac arrest in Japan . Circulation 2012 ; 126 : 2834 – 2843 . Google Scholar Crossref Search ADS PubMed 7 Iwami T , Hiraide A , Nakanishi N , Hayashi Y , Nishiuchi T , Uejima T , Morita H , Shigemoto T , Ikeuchi H , Matsusaka M , Shinya H , Yukioka H , Sugimoto H. Outcome and characteristics of out-of-hospital cardiac arrest according to location of arrest: a report from a large-scale, population-based study in Osaka, Japan . Resuscitation 2006 ; 69 : 221 – 228 . Google Scholar Crossref Search ADS PubMed 8 Weisfeldt ML , Everson-Stewart S , Sitlani C , Rea T , Aufderheide TP , Atkins DL , Bigham B , Brooks SC , Foerster C , Gray R , Ornato JP , Powell J , Kudenchuk PJ , Morrison LJ. Ventricular tachyarrhythmias after cardiac arrest in public versus at home . N Engl J Med 2011 ; 364 : 313 – 321 . Google Scholar Crossref Search ADS PubMed 9 Folke F , Gislason GH , Lippert FK , Nielsen SL , Weeke P , Hansen ML , Fosbøl EL , Andersen SS , Rasmussen S , Schramm TK , Køber L , Torp-Pedersen C. Differences between out-of-hospital cardiac arrest in residential and public locations and implications for public-access defibrillation . Circulation 2010 ; 122 : 623 – 630 . Google Scholar Crossref Search ADS PubMed 10 Herlitz J , Eek M , Holmberg M , Engdahl J , Holmberg S. Characteristics and outcome among patients having out of hospital cardiac arrest at home compared with elsewhere . Heart 2002 ; 88 : 579 – 582 . Google Scholar Crossref Search ADS PubMed 11 Swor RA , Jackson RE , Compton S , Domeier R , Zalenski R , Honeycutt L , Kuhn GJ , Frederiksen S , Pascual RG. Cardiac arrest in private locations: different strategies are needed to improve outcome . Resuscitation 2003 ; 58 : 171 – 176 . Google Scholar Crossref Search ADS PubMed 12 Hansen SM , Hansen CM , Folke F , Rajan S , Kragholm K , Ejlskov L , Gislason G , Køber L , Gerds TA , Hjortshøj S , Lippert F , Torp-Pedersen C , Wissenberg M. Bystander defibrillation for out-of-hospital cardiac arrest in public vs residential locations . JAMA Cardiol 2017 ; 2 : 507 – 514 . Google Scholar Crossref Search ADS PubMed 13 R Core Team . R: A Language and Environment for Statistical Computing . Vienna, Austria : R Foundation for Statistical Computing ; 2017 . https://www.R-project.org/. 14 Kronick SL , Kurz MC , Lin S , Edelson DP , Berg RA , Billi JE , Cabanas JG , Cone DC , Diercks DB , Foster JJ , Meeks RA , Travers AH , Welsford M. Part 4: systems of care and continuous quality improvement: 2015 American Heart Association Guidelines update for cardiopulmonary resuscitation and emergency cardiovascular care . Circulation 2015 ; 132 (18 Suppl 2): S397 – S413 . Google Scholar Crossref Search ADS PubMed 15 Patel N , Patel NJ , Macon CJ , Thakkar B , Desai M , Rengifo-Moreno P , Alfonso CE , Myerburg RJ , Bhatt DL , Cohen MG. Trends and outcomes of coronary angiography and percutaneous coronary intervention after out-of-hospital cardiac arrest associated with ventricular fibrillation or pulseless ventricular tachycardia . JAMA Cardiol 2016 ; 1 : 890 – 899 . Google Scholar Crossref Search ADS PubMed 16 Hassager C , Nagao K , Hildick-Smith D. Out-of-hospital cardiac arrest: in-hospital intervention strategies . Lancet 2018 ; 391 : 989 – 998 . Google Scholar Crossref Search ADS PubMed 17 Fordyce CB , Hansen CM , Kragholm K , Dupre ME , Jollis JG , Roettig ML , Becker LB , Hansen SM , Hinohara TT , Corbett CC , Monk L , Nelson RD , Pearson DA , Tyson C , Diepen S , van Anderson ML , McNally B , Granger CB. Association of public health initiatives with outcomes for out-of-hospital cardiac arrest at home and in public locations . JAMA Cardiol 2017 ; 2 : 1226 – 1235 . Google Scholar Crossref Search ADS PubMed 18 Priori SG , Blomström-Lundqvist C , Mazzanti A , Blom N , Borggrefe M , Camm J , Elliott PM , Fitzsimons D , Hatala R , Hindricks G , Kirchhof P , Kjeldsen K , Kuck K-H , Hernandez-Madrid A , Nikolaou N , Norekvål TM , Spaulding C , Van Veldhuisen DJ. 2015 ESC Guidelines for the management of patients with ventricular arrhythmias and the prevention of sudden cardiac death: the Task Force for the Management of Patients with Ventricular Arrhythmias and the Prevention of Sudden Cardiac Death of the European Society of Cardiology (ESC)Endorsed by: Association for European Paediatric and Congenital Cardiology (AEPC) . Eur Heart J 2015 ; 36 : 2793 – 2867 . Google Scholar Crossref Search ADS PubMed 19 Ringh M , Rosenqvist M , Hollenberg J , Jonsson M , Fredman D , Nordberg P , Järnbert-Pettersson H , Hasselqvist-Ax I , Riva G , Svensson L. Mobile-phone dispatch of laypersons for CPR in out-of-hospital cardiac arrest . N Engl J Med 2015 ; 372 : 2316 – 2325 . Google Scholar Crossref Search ADS PubMed 20 Zijlstra JA , Stieglis R , Riedijk F , Smeekes M , Worp WE , van der Koster RW. Local lay rescuers with AEDs, alerted by text messages, contribute to early defibrillation in a Dutch out-of-hospital cardiac arrest dispatch system . Resuscitation 2014 ; 85 : 1444 – 1449 . Google Scholar Crossref Search ADS PubMed 21 Axelsson Å , Herlitz J , Ekström L , Holmberg S. Bystander-initiated cardiopulmonary resuscitation out-of-hospital. A first description of the bystanders and their experiences . Resuscitation 1996 ; 33 : 3 – 11 . Google Scholar Crossref Search ADS PubMed 22 Blewer AL , McGovern SK , Schmicker R , Morrison LJ , May S , Aufderheide TP , Daya M , Idris AH , Callaway CW , Kudenchuk PJ , Vilke GM , Abella BS. Abstract 16409: gender disparities among patients receiving bystander cardiopulmonary resuscitation in the United States . Circulation 2017 ; 136 : A16409 . 23 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 . Google Scholar Crossref Search ADS PubMed Published on behalf of the European Society of Cardiology. All rights reserved. © The Author(s) 2018. For permissions, please email: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png European Heart Journal Oxford University Press

Bystander cardiopulmonary resuscitation and long-term outcomes in out-of-hospital cardiac arrest according to location of arrest

Loading next page...
 
/lp/oxford-university-press/bystander-cardiopulmonary-resuscitation-and-long-term-outcomes-in-out-iigrnSjIGE
Publisher
Oxford University Press
Copyright
Published on behalf of the European Society of Cardiology. All rights reserved. © The Author(s) 2018. For permissions, please email: journals.permissions@oup.com.
ISSN
0195-668X
eISSN
1522-9645
D.O.I.
10.1093/eurheartj/ehy687
Publisher site
See Article on Publisher Site

Abstract

Abstract Aims Bystander cardiopulmonary resuscitation (CPR) has increased in several countries following nationwide initiatives to facilitate bystander resuscitative efforts in out-of-hospital cardiac arrest (OHCA). We examined the importance of public or residential location of arrest on temporal changes in bystander CPR and outcomes. Methods and results From the nationwide Danish Cardiac Arrest Registry, all OHCAs from 2001 to 2014 of presumed cardiac cause and between 18 and 100 years of age were identified. Arrests witnessed by emergency medical services personnel were excluded. Of 25 505 OHCAs, 26.4% (n = 6738) and 73.6% (n = 18 767) were in public and residential locations, respectively. Bystander CPR increased during 2001–2014 in both locations: from 36.4% [95% confidence interval (CI) 30.6–42.6%] to 83.1% (95% CI 80.0–85.8%) in public (P < 0.001) and from 16.0% (95% CI 13.2–19.3%) to 61.0% (95% CI 58.7–63.2%) in residential locations (P < 0.001). Concurrently, 30-day survival increased in public from 6.4% (95% CI 4.0–10.0%) to 25.2% (95% CI 22.1–28.7%) (P < 0.001), and in residential from 2.9% (95% CI 1.8–4.5%) to 10.0% (95% CI 8.7–11.4%) (P < 0.001). Among 2281 30-day survivors, 1-year risk of anoxic brain damage/nursing home admission during 2001–2014 decreased from 18.8% (95% CI 6.6–43.0%) to 6.8% (95% CI 3.9–11.8%) in public (P < 0.001), whereas the corresponding change was insignificant in residential locations from 11.8% (95% CI 3.3–34.3) to 17.6% (95% CI 12.7–23.9%) (P = 0.52). Conclusion During 2001–2014, bystander CPR and 30-day survival more than doubled in both public and residential OHCA locations. A significant decrease in anoxic brain damage/nursing home admission was observed among 30-day survivors in public, but not among survivors from residential OHCAs. View largeDownload slide View largeDownload slide OHCA, Bystander CPR, Long-term outcomes, Location of arrest See page 319 for the editorial comment on this article (doi: 10.1093/eurheartj/ehy911) One-sentence summary Nationwide initiatives to improve out-of-hospital cardiac arrest pre-hospital care seem equally beneficial on improving bystander cardiopulmonary resuscitation and survival rates in public and residential location of arrest. Introduction Early bystander cardiopulmonary resuscitation (CPR) is pivotal for survival and long-term outcomes in survivors from out-of-hospital cardiac arrest (OHCA).1–3 Consequently, numerous countries have made national efforts to improve pre-hospital resuscitative efforts, including CPR training, which has led to increased CPR rates and a concomitant increase in survival.4–6 However, survival from OHCA varies greatly with location of arrest, and arrests in residential locations have much lower survival rates compared with arrests in public locations.7–9 It has previously been observed that OHCAs in residential locations have lower CPR rates compared with arrests elsewhere10,11; and it was recently discovered that bystander defibrillation rates from onsite automated external defibrillators (AEDs) increased in public locations from 2001 to 2012, while bystander defibrillation remained unchanged and limited in residential locations.12 In Denmark, bystander CPR and survival rates more than doubled between 2001 and 2010 following several national initiatives to improve cardiac arrest management.4 Yet it remains unknown how these improvements might differ according to location of arrest. The aims of this study were: (i) to assess time trends in bystander CPR and long-term outcomes in public and residential location of arrest; (ii) to identify pre-arrest factors associated with not receiving bystander CPR separately for each location. Methods Study setting This study took place in Denmark with a population of approximately 5.6 million inhabitants and demographics varying from urban to rural areas. The Danish Cardiac Arrest Registry All OHCAs in Denmark where resuscitative efforts have been initiated by bystander or emergency medical services (EMS) personnel have been reported in the Danish Cardiac Arrest Registry (http://www.isrctn.com/ISRCTN14261134) since June 2001. Out-of-hospital cardiac arrest cases with obvious late signs of death without a resuscitation attempt are not included in the registry. Case ascertainment is close to complete, as EMS is dispatched to all emergency calls in Denmark and it is mandatory for the dispatched EMS personnel to complete a short report form with detailed information describing the OHCA case according to the Utstein criteria. Location of arrest was divided in public (public spaces, public areas, e.g. hallways and roads, places in the nature, shops, workplaces, and institutions) and residential (private homes) location of arrest. Out-of-hospital cardiac arrests of presumed cardiac cause were identified using diagnosis codes from death certificates and discharge diagnoses.4 The Danish nationwide registries Danish healthcare data are extensively recorded in multiple national registries. Cross-referencing of the registries is possible, as Danish citizens are issued a unique and permanent civil registration number at birth or immigration. Information on sex, age, and vital status are recorded in the Danish Civil Registration System. All hospitalizations are recorded in the National Patient Registry and coded according to the 10th edition of the International Classification of Diseases (ICD10). The Danish Register of Medicinal Product Statistics contains recordings of all redeemed prescriptions in Denmark classified according to the Anatomical Therapeutic Classification system. Income and civil status are recorded on an annual level by Statistics Denmark as well as information on nursing home admissions. Deaths and causes of death are recorded in the Danish Cause of Death Register. Study population, comorbidity and pharmacotherapy The study population consisted of all OHCAs 18–100 years of age in Denmark from June 2001 to 2014, excluding patients with arrest of presumed non-cardiac cause and arrests witnessed by EMS personnel. Information on comorbidity was obtained from discharge diagnoses up to 5 years prior to OHCA, with exception of diabetes, where the redemption of a prescription for antidiabetic medication up to 180 days prior to cardiac arrest defined diabetes. Anoxic brain damage was established from the diagnosis code G93.1 at any time during the 1-year follow-up period, as previously described.3 Daytime, evening, and night-time was defined as 8 AM to 3.59 PM, 4 PM to 11.59 PM, and 12 PM to 7.59 AM, respectively. Income was based on household earnings the year before the cardiac arrest and indexed according to 2010. Income group was categorized in low, medium and high based on the 1st and 3rd income quartile. Study outcome The primary study outcome was bystander CPR according to location of arrest. Secondary outcomes were 30-day survival and 1-year survival, and in 30-day survivors also all-cause mortality and the cumulative risk of a composite endpoint of anoxic brain damage or admission to nursing home during a 1-year follow-up period according to location of arrest and bystander CPR. Statistics Baseline characteristics were summarized according to bystander CPR and by public or residential location of arrest. Continuous variables were compared using Kruskal–Wallis test and categorical values were compared using χ2 test. Linear calendar time trends were examined using univariate logistic regression for bystander CPR and survival, univariate Fine-Gray regression for anoxic brain damage or nursing home admission, and univariate Cox regression for all-cause mortality using calendar year of cardiac arrest as a numeric variable. The association between bystander CPR and pre-arrest conditions were analysed by multivariable logistic regression separately for each location of arrest. Risks of anoxic brain damage or nursing home admission and all-cause mortality in 30-day survivors were analysed using multivariable cause-specific Cox regression analyses adjusted for age, sex, witnessed status, time from arrest to EMS arrival, Charlson Comorbidity Index, and calendar year (2001–2005, 2006–2011, 2012–2014). When analysing the risk of the combined endpoint anoxic brain damage or nursing home admission, death was a competing risk. In the multivariable logistic regression analyses and the multivariable cause-specific Cox regression analyses, missing data were handled by multiple imputations on 200 datasets by the smcfcs package separately for public and residential arrests.3 For complete case analyses, see Supplementary material online. Analyses were carried out using SAS software version 9.4 (SAS institute Inc., NC, USA) and R version 3.3.3 (R Development Core Team).13 Statistical significance was defined as a two-sided P-value < 0.05. Ethics This project was approved by The Danish Data Protection Agency (Ref.no. 2007-58-0015, local ref.no. GEH-2014-017, I-Suite.nr. 02735). Ethical approval is not required for retrospective registry studies in Denmark. Results A total of 25 505 first-time OHCA patients 18–100 years of age from 2001 to 2014 were included in the study (Figure 1). Most arrests occurred in residential locations: 73.6% (n = 18 767) vs. 26.4% (n = 6738) in public locations. Baseline characteristics are presented in Table 1 and Supplementary material online, Table S1. Table 1 Characteristics of OHCA according to bystander CPR in public and residential location of arrest Characteristics Public (N = 6738a) Residential (N = 18 767a) No bystander CPR (n = 2471, 38.8%) Bystander CPR (n = 3901, 61.2%) P-values No bystander CPR (n = 11 166, 62.2%) Bystander CPR (n = 6799, 37.8%) P-values Age (years), median (IQR) 70 (60–80) 67 (57–77) <0.001 74 (65–82) 70 (61–80) <0.001 Male sex, n (%) 1769 (71.6) 3054 (78.3) <0.001 7201 (64.5) 4430 (65.2) <0.37 Time of day, n (%) <0.001 <0.001  Day 1294 (53.3) 2264 (58.5) 4355 (39.9) 2702 (40.3)  Evening 767 (31.6) 1204 (31.1) 3466 (31.7) 2397 (35.7)  Night 367 (15.1) 404 (10.4) 3105 (28.4) 1607 (24.0)   Data missing, n (%) 43 (1.7) 29 (0.7) 240 (2.2) 93 (1.4) Witnessed arrest, n (%) 1422 (57.9) 2858 (73.6) <0.001 4766 (42.9) 3759 (55.5) <0.001  Data missing, n (%) 13 (0.5) 17 (0.4) 50 (0.4) 29 (0.4) First defibrillation by bystanders, n (%) 14 (0.6) 382 (10.6) <0.001 18 (0.2) 143 (2.2) <0.001  Data missing, n (%) 168 (6.8) 283 (7.3) 477 (4.3) 215 (3.2) First defibrillation by EMS, n (%) 1080 (47.9) 1944 (53.7) <0.001 3563 (33.8) 2552 (38.8) <0.001  Data missing, n (%) 217 (8.8) 284 (7.3) 640 (5.7) 229 (3.4) Shockable first rhythm, n (%) 706 (29.8) 1954 (51.9) <0.001 1788 (16.8) 1946 (29.7) <0.001  Data missing, n (%) 103 (4.2) 138 (3.5) 501 (4.5) 246 (3.6) Time from arrest to first rhythm analysis by the EMS, median(IQR) 11 (7–16) 10 (5–16) <0.001 12 (7–21) 13 (8–19) <0.001  Data missing, n (%) 354 (14.3) 599 (15.4) 1559 (14.0) 967 (14.2) Income group, n (%) <0.001 <0.001  Low 747 (30.4) 790 (20.4) 2548 (22.9) 1365 (20.1)  Medium 1133 (46.1) 1724 (44.4) 6277 (56.3) 3391 (50.0)  High 579 (23.5) 1366 (35.2) 2322 (20.8) 2031 (29.9)   Data missing, n (%) 12 (0.5) 21 (0.5) 19 (0.2) 12 (0.2) CAG within 30 daysb 134 (71.7) 850 (84.0) <0.001 248 (67.8) 553 (81.1) <0.001 PCI within 30 daysb 78 (41.7) 402 (39.7) 0.67 125 (34.2) 261 (38.3) 0.21 ICD within 30 daysb 42 (22.5) 468 (46.2) <0.001 110 (30.1) 293 (43.0) <0.001 Comorbidity  Ischaemic heart disease including myocardial infarction, n (%) 629 (25.5) 1309 (33, 6) <0.001 2123 (19.0) 1240 (18.2) 0.05  Congestive heart failure, n (%) 387 (15.7) 604 (15.5) 0.88 1982 (17.8) 1174 (17.3) 0.42  Cerebral vascular disease, n (%) 237 (9.6) 322 (8.3) 0.07 987 (8.8) 625 (9.2) 0.44  Diabetes, n (%) 324 (13.1) 501 (12.8) 0.78 1705 (15.3) 1143 (16.8) 0.006  COPD, n (%) 214 (8.7) 264 (6.8) 0.006 1560 (14.0) 813 (12.0) <0.001  Malignancy, n (%) 141 (5.7) 189 (4.8) 0.15 844 (7.6) 481 (7.1) 0.24  Liver diseases, n (%) 27 (1.1) 21 (0.5) 0.02 162 (1.5) 104 (1.5) 0.72  Renal diseases, n (%) 77 (3.1) 136 (3.5) 0.47 486 (4.4) 346 (5.1) 0.03  Psychiatric diseases, n (%) 331 (13.4) 391 (10.0) <0.001 1079 (9.7) 717 (10.5) 0.06  Charlson Comorbidity Index group, n (%) 0.02 0.70   0 1442 (58.4) 2159 (55.3) 5864 (52.5) 3581 (52.7)   1 468 (18.9) 844 (21.6) 2105 (18.9) 1249 (18.4)   >1 561 (22.7) 898 (23.0) 3197 (28.6) 1969 (29.0) Cohabitant  Age of cohabitant, median (IQR) 66 (58, 74) 63 (54, 71) <0.001 71 (63, 77) 65 (57, 74) <0.001   Data missing, n (%) 13 (0.52) 10 (0.26) 31 (0.28) 29 (0.43)  Male cohabitant, n (%) 197 (17.6) 287 (13.7) 0.004 1835 (26.1) 1080 (25.7) 0.63   Data missing, n (%) 13 (0.52) 10 (0.26) 31 (0.28) 29 (0.43)  Charlson Comorbidity Index group in cohabitant, n (%) 0.001 <0.001   0 941 (83.2) 1855 (87.7) 9834 (88.1) 6176 (90.8)   1 67 (5.9) 103 (4.9) 532 (4.8) 227 (3.3)   >1 123 (10.9) 157 (7.4) 800 (7.2) 396 (5.8) Outcomes  Thirty-day survivor, n (%) 187 (7.6) 1012 (25.9) <0.001 366 (3.3) 682 (10.0) <0.001  One-year survivor, n (%) 172 (7.0) 956 (24.5) <0.001 312 (2.8) 614 (9.0) <0.001 Characteristics Public (N = 6738a) Residential (N = 18 767a) No bystander CPR (n = 2471, 38.8%) Bystander CPR (n = 3901, 61.2%) P-values No bystander CPR (n = 11 166, 62.2%) Bystander CPR (n = 6799, 37.8%) P-values Age (years), median (IQR) 70 (60–80) 67 (57–77) <0.001 74 (65–82) 70 (61–80) <0.001 Male sex, n (%) 1769 (71.6) 3054 (78.3) <0.001 7201 (64.5) 4430 (65.2) <0.37 Time of day, n (%) <0.001 <0.001  Day 1294 (53.3) 2264 (58.5) 4355 (39.9) 2702 (40.3)  Evening 767 (31.6) 1204 (31.1) 3466 (31.7) 2397 (35.7)  Night 367 (15.1) 404 (10.4) 3105 (28.4) 1607 (24.0)   Data missing, n (%) 43 (1.7) 29 (0.7) 240 (2.2) 93 (1.4) Witnessed arrest, n (%) 1422 (57.9) 2858 (73.6) <0.001 4766 (42.9) 3759 (55.5) <0.001  Data missing, n (%) 13 (0.5) 17 (0.4) 50 (0.4) 29 (0.4) First defibrillation by bystanders, n (%) 14 (0.6) 382 (10.6) <0.001 18 (0.2) 143 (2.2) <0.001  Data missing, n (%) 168 (6.8) 283 (7.3) 477 (4.3) 215 (3.2) First defibrillation by EMS, n (%) 1080 (47.9) 1944 (53.7) <0.001 3563 (33.8) 2552 (38.8) <0.001  Data missing, n (%) 217 (8.8) 284 (7.3) 640 (5.7) 229 (3.4) Shockable first rhythm, n (%) 706 (29.8) 1954 (51.9) <0.001 1788 (16.8) 1946 (29.7) <0.001  Data missing, n (%) 103 (4.2) 138 (3.5) 501 (4.5) 246 (3.6) Time from arrest to first rhythm analysis by the EMS, median(IQR) 11 (7–16) 10 (5–16) <0.001 12 (7–21) 13 (8–19) <0.001  Data missing, n (%) 354 (14.3) 599 (15.4) 1559 (14.0) 967 (14.2) Income group, n (%) <0.001 <0.001  Low 747 (30.4) 790 (20.4) 2548 (22.9) 1365 (20.1)  Medium 1133 (46.1) 1724 (44.4) 6277 (56.3) 3391 (50.0)  High 579 (23.5) 1366 (35.2) 2322 (20.8) 2031 (29.9)   Data missing, n (%) 12 (0.5) 21 (0.5) 19 (0.2) 12 (0.2) CAG within 30 daysb 134 (71.7) 850 (84.0) <0.001 248 (67.8) 553 (81.1) <0.001 PCI within 30 daysb 78 (41.7) 402 (39.7) 0.67 125 (34.2) 261 (38.3) 0.21 ICD within 30 daysb 42 (22.5) 468 (46.2) <0.001 110 (30.1) 293 (43.0) <0.001 Comorbidity  Ischaemic heart disease including myocardial infarction, n (%) 629 (25.5) 1309 (33, 6) <0.001 2123 (19.0) 1240 (18.2) 0.05  Congestive heart failure, n (%) 387 (15.7) 604 (15.5) 0.88 1982 (17.8) 1174 (17.3) 0.42  Cerebral vascular disease, n (%) 237 (9.6) 322 (8.3) 0.07 987 (8.8) 625 (9.2) 0.44  Diabetes, n (%) 324 (13.1) 501 (12.8) 0.78 1705 (15.3) 1143 (16.8) 0.006  COPD, n (%) 214 (8.7) 264 (6.8) 0.006 1560 (14.0) 813 (12.0) <0.001  Malignancy, n (%) 141 (5.7) 189 (4.8) 0.15 844 (7.6) 481 (7.1) 0.24  Liver diseases, n (%) 27 (1.1) 21 (0.5) 0.02 162 (1.5) 104 (1.5) 0.72  Renal diseases, n (%) 77 (3.1) 136 (3.5) 0.47 486 (4.4) 346 (5.1) 0.03  Psychiatric diseases, n (%) 331 (13.4) 391 (10.0) <0.001 1079 (9.7) 717 (10.5) 0.06  Charlson Comorbidity Index group, n (%) 0.02 0.70   0 1442 (58.4) 2159 (55.3) 5864 (52.5) 3581 (52.7)   1 468 (18.9) 844 (21.6) 2105 (18.9) 1249 (18.4)   >1 561 (22.7) 898 (23.0) 3197 (28.6) 1969 (29.0) Cohabitant  Age of cohabitant, median (IQR) 66 (58, 74) 63 (54, 71) <0.001 71 (63, 77) 65 (57, 74) <0.001   Data missing, n (%) 13 (0.52) 10 (0.26) 31 (0.28) 29 (0.43)  Male cohabitant, n (%) 197 (17.6) 287 (13.7) 0.004 1835 (26.1) 1080 (25.7) 0.63   Data missing, n (%) 13 (0.52) 10 (0.26) 31 (0.28) 29 (0.43)  Charlson Comorbidity Index group in cohabitant, n (%) 0.001 <0.001   0 941 (83.2) 1855 (87.7) 9834 (88.1) 6176 (90.8)   1 67 (5.9) 103 (4.9) 532 (4.8) 227 (3.3)   >1 123 (10.9) 157 (7.4) 800 (7.2) 396 (5.8) Outcomes  Thirty-day survivor, n (%) 187 (7.6) 1012 (25.9) <0.001 366 (3.3) 682 (10.0) <0.001  One-year survivor, n (%) 172 (7.0) 956 (24.5) <0.001 312 (2.8) 614 (9.0) <0.001 a Including 366 public arrests and 802 residential arrests with missing values on bystander CPR. Baseline characteristics are presented in Supplementary material online, Table S2. b 30-Day survivors. CAG, coronary angiography; COPD, chronic obstructive pulmonary disease; CPR, cardiopulmonary resuscitation; EMS, emergency medical services; ICD, implantable cardiac defibrillator; IQR, interquartile range; OHCA, out-of-hospital cardiac arrest; PCI, percutaneous coronary intervention. Table 1 Characteristics of OHCA according to bystander CPR in public and residential location of arrest Characteristics Public (N = 6738a) Residential (N = 18 767a) No bystander CPR (n = 2471, 38.8%) Bystander CPR (n = 3901, 61.2%) P-values No bystander CPR (n = 11 166, 62.2%) Bystander CPR (n = 6799, 37.8%) P-values Age (years), median (IQR) 70 (60–80) 67 (57–77) <0.001 74 (65–82) 70 (61–80) <0.001 Male sex, n (%) 1769 (71.6) 3054 (78.3) <0.001 7201 (64.5) 4430 (65.2) <0.37 Time of day, n (%) <0.001 <0.001  Day 1294 (53.3) 2264 (58.5) 4355 (39.9) 2702 (40.3)  Evening 767 (31.6) 1204 (31.1) 3466 (31.7) 2397 (35.7)  Night 367 (15.1) 404 (10.4) 3105 (28.4) 1607 (24.0)   Data missing, n (%) 43 (1.7) 29 (0.7) 240 (2.2) 93 (1.4) Witnessed arrest, n (%) 1422 (57.9) 2858 (73.6) <0.001 4766 (42.9) 3759 (55.5) <0.001  Data missing, n (%) 13 (0.5) 17 (0.4) 50 (0.4) 29 (0.4) First defibrillation by bystanders, n (%) 14 (0.6) 382 (10.6) <0.001 18 (0.2) 143 (2.2) <0.001  Data missing, n (%) 168 (6.8) 283 (7.3) 477 (4.3) 215 (3.2) First defibrillation by EMS, n (%) 1080 (47.9) 1944 (53.7) <0.001 3563 (33.8) 2552 (38.8) <0.001  Data missing, n (%) 217 (8.8) 284 (7.3) 640 (5.7) 229 (3.4) Shockable first rhythm, n (%) 706 (29.8) 1954 (51.9) <0.001 1788 (16.8) 1946 (29.7) <0.001  Data missing, n (%) 103 (4.2) 138 (3.5) 501 (4.5) 246 (3.6) Time from arrest to first rhythm analysis by the EMS, median(IQR) 11 (7–16) 10 (5–16) <0.001 12 (7–21) 13 (8–19) <0.001  Data missing, n (%) 354 (14.3) 599 (15.4) 1559 (14.0) 967 (14.2) Income group, n (%) <0.001 <0.001  Low 747 (30.4) 790 (20.4) 2548 (22.9) 1365 (20.1)  Medium 1133 (46.1) 1724 (44.4) 6277 (56.3) 3391 (50.0)  High 579 (23.5) 1366 (35.2) 2322 (20.8) 2031 (29.9)   Data missing, n (%) 12 (0.5) 21 (0.5) 19 (0.2) 12 (0.2) CAG within 30 daysb 134 (71.7) 850 (84.0) <0.001 248 (67.8) 553 (81.1) <0.001 PCI within 30 daysb 78 (41.7) 402 (39.7) 0.67 125 (34.2) 261 (38.3) 0.21 ICD within 30 daysb 42 (22.5) 468 (46.2) <0.001 110 (30.1) 293 (43.0) <0.001 Comorbidity  Ischaemic heart disease including myocardial infarction, n (%) 629 (25.5) 1309 (33, 6) <0.001 2123 (19.0) 1240 (18.2) 0.05  Congestive heart failure, n (%) 387 (15.7) 604 (15.5) 0.88 1982 (17.8) 1174 (17.3) 0.42  Cerebral vascular disease, n (%) 237 (9.6) 322 (8.3) 0.07 987 (8.8) 625 (9.2) 0.44  Diabetes, n (%) 324 (13.1) 501 (12.8) 0.78 1705 (15.3) 1143 (16.8) 0.006  COPD, n (%) 214 (8.7) 264 (6.8) 0.006 1560 (14.0) 813 (12.0) <0.001  Malignancy, n (%) 141 (5.7) 189 (4.8) 0.15 844 (7.6) 481 (7.1) 0.24  Liver diseases, n (%) 27 (1.1) 21 (0.5) 0.02 162 (1.5) 104 (1.5) 0.72  Renal diseases, n (%) 77 (3.1) 136 (3.5) 0.47 486 (4.4) 346 (5.1) 0.03  Psychiatric diseases, n (%) 331 (13.4) 391 (10.0) <0.001 1079 (9.7) 717 (10.5) 0.06  Charlson Comorbidity Index group, n (%) 0.02 0.70   0 1442 (58.4) 2159 (55.3) 5864 (52.5) 3581 (52.7)   1 468 (18.9) 844 (21.6) 2105 (18.9) 1249 (18.4)   >1 561 (22.7) 898 (23.0) 3197 (28.6) 1969 (29.0) Cohabitant  Age of cohabitant, median (IQR) 66 (58, 74) 63 (54, 71) <0.001 71 (63, 77) 65 (57, 74) <0.001   Data missing, n (%) 13 (0.52) 10 (0.26) 31 (0.28) 29 (0.43)  Male cohabitant, n (%) 197 (17.6) 287 (13.7) 0.004 1835 (26.1) 1080 (25.7) 0.63   Data missing, n (%) 13 (0.52) 10 (0.26) 31 (0.28) 29 (0.43)  Charlson Comorbidity Index group in cohabitant, n (%) 0.001 <0.001   0 941 (83.2) 1855 (87.7) 9834 (88.1) 6176 (90.8)   1 67 (5.9) 103 (4.9) 532 (4.8) 227 (3.3)   >1 123 (10.9) 157 (7.4) 800 (7.2) 396 (5.8) Outcomes  Thirty-day survivor, n (%) 187 (7.6) 1012 (25.9) <0.001 366 (3.3) 682 (10.0) <0.001  One-year survivor, n (%) 172 (7.0) 956 (24.5) <0.001 312 (2.8) 614 (9.0) <0.001 Characteristics Public (N = 6738a) Residential (N = 18 767a) No bystander CPR (n = 2471, 38.8%) Bystander CPR (n = 3901, 61.2%) P-values No bystander CPR (n = 11 166, 62.2%) Bystander CPR (n = 6799, 37.8%) P-values Age (years), median (IQR) 70 (60–80) 67 (57–77) <0.001 74 (65–82) 70 (61–80) <0.001 Male sex, n (%) 1769 (71.6) 3054 (78.3) <0.001 7201 (64.5) 4430 (65.2) <0.37 Time of day, n (%) <0.001 <0.001  Day 1294 (53.3) 2264 (58.5) 4355 (39.9) 2702 (40.3)  Evening 767 (31.6) 1204 (31.1) 3466 (31.7) 2397 (35.7)  Night 367 (15.1) 404 (10.4) 3105 (28.4) 1607 (24.0)   Data missing, n (%) 43 (1.7) 29 (0.7) 240 (2.2) 93 (1.4) Witnessed arrest, n (%) 1422 (57.9) 2858 (73.6) <0.001 4766 (42.9) 3759 (55.5) <0.001  Data missing, n (%) 13 (0.5) 17 (0.4) 50 (0.4) 29 (0.4) First defibrillation by bystanders, n (%) 14 (0.6) 382 (10.6) <0.001 18 (0.2) 143 (2.2) <0.001  Data missing, n (%) 168 (6.8) 283 (7.3) 477 (4.3) 215 (3.2) First defibrillation by EMS, n (%) 1080 (47.9) 1944 (53.7) <0.001 3563 (33.8) 2552 (38.8) <0.001  Data missing, n (%) 217 (8.8) 284 (7.3) 640 (5.7) 229 (3.4) Shockable first rhythm, n (%) 706 (29.8) 1954 (51.9) <0.001 1788 (16.8) 1946 (29.7) <0.001  Data missing, n (%) 103 (4.2) 138 (3.5) 501 (4.5) 246 (3.6) Time from arrest to first rhythm analysis by the EMS, median(IQR) 11 (7–16) 10 (5–16) <0.001 12 (7–21) 13 (8–19) <0.001  Data missing, n (%) 354 (14.3) 599 (15.4) 1559 (14.0) 967 (14.2) Income group, n (%) <0.001 <0.001  Low 747 (30.4) 790 (20.4) 2548 (22.9) 1365 (20.1)  Medium 1133 (46.1) 1724 (44.4) 6277 (56.3) 3391 (50.0)  High 579 (23.5) 1366 (35.2) 2322 (20.8) 2031 (29.9)   Data missing, n (%) 12 (0.5) 21 (0.5) 19 (0.2) 12 (0.2) CAG within 30 daysb 134 (71.7) 850 (84.0) <0.001 248 (67.8) 553 (81.1) <0.001 PCI within 30 daysb 78 (41.7) 402 (39.7) 0.67 125 (34.2) 261 (38.3) 0.21 ICD within 30 daysb 42 (22.5) 468 (46.2) <0.001 110 (30.1) 293 (43.0) <0.001 Comorbidity  Ischaemic heart disease including myocardial infarction, n (%) 629 (25.5) 1309 (33, 6) <0.001 2123 (19.0) 1240 (18.2) 0.05  Congestive heart failure, n (%) 387 (15.7) 604 (15.5) 0.88 1982 (17.8) 1174 (17.3) 0.42  Cerebral vascular disease, n (%) 237 (9.6) 322 (8.3) 0.07 987 (8.8) 625 (9.2) 0.44  Diabetes, n (%) 324 (13.1) 501 (12.8) 0.78 1705 (15.3) 1143 (16.8) 0.006  COPD, n (%) 214 (8.7) 264 (6.8) 0.006 1560 (14.0) 813 (12.0) <0.001  Malignancy, n (%) 141 (5.7) 189 (4.8) 0.15 844 (7.6) 481 (7.1) 0.24  Liver diseases, n (%) 27 (1.1) 21 (0.5) 0.02 162 (1.5) 104 (1.5) 0.72  Renal diseases, n (%) 77 (3.1) 136 (3.5) 0.47 486 (4.4) 346 (5.1) 0.03  Psychiatric diseases, n (%) 331 (13.4) 391 (10.0) <0.001 1079 (9.7) 717 (10.5) 0.06  Charlson Comorbidity Index group, n (%) 0.02 0.70   0 1442 (58.4) 2159 (55.3) 5864 (52.5) 3581 (52.7)   1 468 (18.9) 844 (21.6) 2105 (18.9) 1249 (18.4)   >1 561 (22.7) 898 (23.0) 3197 (28.6) 1969 (29.0) Cohabitant  Age of cohabitant, median (IQR) 66 (58, 74) 63 (54, 71) <0.001 71 (63, 77) 65 (57, 74) <0.001   Data missing, n (%) 13 (0.52) 10 (0.26) 31 (0.28) 29 (0.43)  Male cohabitant, n (%) 197 (17.6) 287 (13.7) 0.004 1835 (26.1) 1080 (25.7) 0.63   Data missing, n (%) 13 (0.52) 10 (0.26) 31 (0.28) 29 (0.43)  Charlson Comorbidity Index group in cohabitant, n (%) 0.001 <0.001   0 941 (83.2) 1855 (87.7) 9834 (88.1) 6176 (90.8)   1 67 (5.9) 103 (4.9) 532 (4.8) 227 (3.3)   >1 123 (10.9) 157 (7.4) 800 (7.2) 396 (5.8) Outcomes  Thirty-day survivor, n (%) 187 (7.6) 1012 (25.9) <0.001 366 (3.3) 682 (10.0) <0.001  One-year survivor, n (%) 172 (7.0) 956 (24.5) <0.001 312 (2.8) 614 (9.0) <0.001 a Including 366 public arrests and 802 residential arrests with missing values on bystander CPR. Baseline characteristics are presented in Supplementary material online, Table S2. b 30-Day survivors. CAG, coronary angiography; COPD, chronic obstructive pulmonary disease; CPR, cardiopulmonary resuscitation; EMS, emergency medical services; ICD, implantable cardiac defibrillator; IQR, interquartile range; OHCA, out-of-hospital cardiac arrest; PCI, percutaneous coronary intervention. Figure 1 View largeDownload slide Population selection. CPR, cardiopulmonary resuscitation; EMS, emergency medical services. Figure 1 View largeDownload slide Population selection. CPR, cardiopulmonary resuscitation; EMS, emergency medical services. Bystander cardiopulmonary resuscitation Throughout the study period, a significant increase in the proportion of patients receiving bystander-initiated CPR was observed in both locations: from 36.4% [95% confidence interval (CI) 30.6–42.6%] to 83.1% (95% CI 80.0–85.8%) in public locations (P for trend <0.001) and from 16.0% (95% CI 13.2–19.3%) to 61.0% (95% CI 58.7–63.2%) in residential locations (P for trend <0.001) (Take home figure). Including witnessed arrest only produced a similar pattern, Supplementary material online, Figure S1. Take home figure View largeDownload slide Proportion of out-of-hospital cardiac arrest patients receiving bystander cardiopulmonary resuscitation according to location of arrest. The vertical bars represent 95% confidence interval. CPR, cardiopulmonary resuscitation. Take home figure View largeDownload slide Proportion of out-of-hospital cardiac arrest patients receiving bystander cardiopulmonary resuscitation according to location of arrest. The vertical bars represent 95% confidence interval. CPR, cardiopulmonary resuscitation. 30-Day and 1-year survival Overall, 30-day survival increased significantly during the study period in both locations: from 6.4% (95% CI 4.0–10.0%) to 25.2% (95% CI 22.1–28.7%) in public locations, and from 2.9% (95% CI 1.8–4.5%) to 10.0% (95% CI 8.7–11.4%) in residential locations (P for trend <0.001), Figure 2. One-year survival is also depicted in Figure 2 and mimicked the results of 30-day survival. In both locations, 30-day and 1-year survival increased in patients who received bystander CPR as well as in patients who did not receive bystander CPR, Supplementary material online, Figure S2. Figure 2 View largeDownload slide Thirty-day and 1-year survival in out-of-hospital cardiac arrest patients according to location of arrest. The vertical bars represent 95% confidence interval. Figure 2 View largeDownload slide Thirty-day and 1-year survival in out-of-hospital cardiac arrest patients according to location of arrest. The vertical bars represent 95% confidence interval. All-cause mortality, anoxic brain damage, and nursing home admission among 30-day survivors A total of 2348 patients survived the first 30 days following OHCA; 67 patients were excluded due to a diagnosis of brain damage or nursing home admission before OHCA. The majority of 30-day survivors had arrest in public locations (53.4%, n = 1217/2281) compared with residential locations (46.6%, n = 1064/2281). Throughout the study period, the 1-year incidences of brain damage or nursing home admission decreased from 18.8% (95% CI 6.6–43.0%) to 6.8% (95% CI 3.9–11.8%) in public arrests (P for trend < 0.001), whereas the change among 30-day survivors from arrests in residential locations from 11.8% (95% CI 3.3–34.3) to 17.6% (95% CI 12.7–23.9%) (P for trend = 0.52), was insignificant (Figure 3). Similarly, 1-year all-cause mortality incidences in 30-day survivors decreased from 12.5% (95% CI 3.5–36.0%) to 5.0% (95% CI 2.5–9.4%) in public (P for trend = 0.03), whereas the change from 29.4% (95% CI 13.3–53.1%) to 13.6% (95% CI 9.3–19.5%) in residential arrests was insignificant (P for trend = 0.75). Bystander CPR was significantly associated with a lower risk of brain damage or nursing home admission in arrests in both public and residential locations; hazard ratio (HR) 0.55 (95% CI 0.36–0.84) and HR 0.60 (95% CI 0.43–0.84), respectively, but not with all-cause mortality; HR 1.13 (95% CI 0.60–2.12) and HR 0.78 (95% CI 0.52–1.17), respectively, Figure 4 (for complete case analyses, see Supplementary material online, Figure S3). Figure 3 View largeDownload slide Anoxic brain damage or nursing home admission and all-cause mortality in 30-day survivors from out-of-hospital cardiac arrest according to location of arrest. The vertical bars represent 95% confidence interval. Figure 3 View largeDownload slide Anoxic brain damage or nursing home admission and all-cause mortality in 30-day survivors from out-of-hospital cardiac arrest according to location of arrest. The vertical bars represent 95% confidence interval. Figure 4 View largeDownload slide The associations between bystander cardiopulmonary resuscitation and brain damage/nursing home admission and death from any cause in 30-day survivors from out-of-hospital cardiac arrest according to location of arrest. *Adjusted for age, sex, year (2001–2004, 2005–2011, 2012–2014), Charlson Comorbidity Index (0, 1, >1), witnessed status, and time from arrest to arrival of EMS (<5, 5–11, >11 min). CPR, cardiopulmonary resuscitation. Figure 4 View largeDownload slide The associations between bystander cardiopulmonary resuscitation and brain damage/nursing home admission and death from any cause in 30-day survivors from out-of-hospital cardiac arrest according to location of arrest. *Adjusted for age, sex, year (2001–2004, 2005–2011, 2012–2014), Charlson Comorbidity Index (0, 1, >1), witnessed status, and time from arrest to arrival of EMS (<5, 5–11, >11 min). CPR, cardiopulmonary resuscitation. Factors associated with not receiving bystander cardiopulmonary resuscitation Factors associated with not receiving bystander CPR according to location of arrest are depicted in Figures 5 and 6 (for complete case analyses, see Supplementary material online, Figure S4). Non-witnessed arrest, high age, low income, and arrest at night were associated with a low probability of receiving bystander CPR in both locations. In public locations, female gender, liver disease, and psychiatric disease were associated with an increased risk of not receiving bystander CPR. Figure 5 View largeDownload slide Factors associated with not receiving bystander cardiopulmonary resuscitation in out-of-hospital cardiac arrest in public locations. Figure 5 View largeDownload slide Factors associated with not receiving bystander cardiopulmonary resuscitation in out-of-hospital cardiac arrest in public locations. Figure 6 View largeDownload slide Factors associated with not receiving bystander cardiopulmonary resuscitation in out-of-hospital cardiac arrest in residential locations. Figure 6 View largeDownload slide Factors associated with not receiving bystander cardiopulmonary resuscitation in out-of-hospital cardiac arrest in residential locations. We tested the robustness of our results by including year as different time intervals (2001–2007, 2008–2014) and including with similar results, Supplementary material online, Figure S5. Including only cardiac arrest patients with witnessed arrest mimicked the main results, Supplementary material online, Figure S6. Throughout the study period, the proportions receiving bystander CPR increased independently of Charlson Comorbidity Index (0, 1, >1) and age group (18–65, 66–80, >80 years) in both locations, Supplementary material online, Figures S7 and S8. Discussion In this nationwide study investigating 25 505 OHCAs during 2001–2014, bystander CPR rates more than doubled in both public and residential locations and 30-day survival rates more than tripled in both locations. Yet, bystander CPR and survival rates remained markedly lower in residential cardiac arrests. In both locations, bystander CPR was significantly associated with a decreased risk of brain damage or admission to nursing home among 30-day survivors. However, the risk of brain damage or nursing home admission in 30-day survivors decreased only significantly over time in public arrests. Non-witnessed arrest, older age, arrest during night-time, and low income were the strongest factors associated with not receiving bystander CPR in both locations; all factors were more prominent in residential locations. Out-of-hospital cardiac arrests in residential locations differ from OHCAs in public locations in several ways: patients are older, fewer have witnessed arrest, and arrest occurs more often during night-time.9,11 All of these factors are associated with lower probability of receiving bystander CPR as shown in this study. The most recent guidelines from the American Heart Association note that patients with OHCA in residential locations are much less likely to receive bystander CPR than patients with OHCA in public locations.14 Although our overall results concur with this statement, our findings of an absolute increase of 45% in the proportion of OHCA patients receiving bystander CPR in residential areas compared with an absolute increase of 47% in public locations are impressive. These results indicate that it is possible through national initiatives and campaigns not only to increase bystander CPR rates in public locations, but also in residential areas, and show that despite unfavourable conditions for resuscitation in residential locations, efforts to increase bystander CPR seem equally beneficial in public and residential locations.4 Throughout the study period, survival increased concomitantly with CPR; however, the absolute increase in survival was greater in public locations compared to residential locations. In addition, we observed an increase in survival also among patients not receiving bystander CPR, indicating that other parts in the management of OHCAs may have improved, e.g. the immediate referral of OHCA patients to cardiac centres, early heart catheterizations and percutaneous interventions, as well as targeted temperature management.15,16 Among 30-day survivors, bystander CPR was equally associated with a decreased risk of anoxic brain damage and nursing home admission in both public and residential locations. Surprisingly, anoxic brain damage or nursing home admission rates decreased only significantly over time in public arrests. This is in line with a recent study and may indicate that the unfavourable conditions connected to arrests in residential locations (e.g. unwitnessed arrest, older patients, and older and more often alone bystanders), limit the impact of improved bystander CPR rates.8,17 All-cause mortality in 30-day survivors was not associated with bystander CPR in both locations, suggesting that if patients do survive the first 30 days the probability of 1-year survival is no longer affected by bystander CPR, but on the patients’ pre-morbid conditions, while the risk of anoxic brain damage continuously is closely associated with bystander CPR. We have previously reported a marked improvement in bystander defibrillation rates during 2001–2012 in Denmark from 1.2% to 15.3% in public arrests, while bystander defibrillation remained limited and unchanged in residential arrests at 1.3%.12 In this study, we found that amongst patients receiving bystander CPR, only 2.2% were bystander defibrillated in residential locations compared with 10.6% in public locations. The low bystander defibrillation rates may partly explain the lack of improvement in rates of anoxic brain damage or nursing home admission in residential arrests, despite bystander CPR being significantly associated with lower risk of anoxic brain damage or nursing home admission in both locations. Collectively, bystanders seem to be very willing to attempt resuscitation, but other barriers prevent them from retrieving the nearest AED. However, the major improvements in bystander CPR rates in both locations are fundamental in also improving bystander defibrillation rates in both locations, which is of great importance if more patients are to survive an OHCA.18 Another key factor to survival in OHCA patients is the EMS response time, which we in line with other studies found to be 2 min longer in residential OHCAs compared with public OHCAs.2,17 One alternative way to improve survival after OHCA may be the activation of nearby laypersons by smart-phones and sending them to assist in the resuscitation attempt: one running directly to the OHCA site performing bystander CPR and the other retrieving a nearby AED, thereby potentially improving bystander CPR rates even further and most importantly minimize time to defibrillation. This has been initiated in several larger cities in Europe with promising results, particularly in cardiac arrests in residential locations, where the majority of the mobile phone activated bystanders arrived before the EMS.19,20 We found that the cohabitants to the group of patients not receiving CPR were older, which indicates that bystanders may not be capable of performing CPR, and would benefit from assisting bystanders, again particularly in residential arrests. Nonetheless, we also observed that arrests in residential locations were characterized by more un-modifiable factors, such as unwitnessed arrests, which could be associated with the very consistent difference in 20% between bystander CPR rates in public and residential locations throughout the study period. Overall, comorbidity was not associated with a significant risk of not receiving bystander CPR. Accordingly, we found an equal increase in bystander CPR independent of comorbidity and age in both locations. However, in public locations liver disease and psychiatric disease were associated with a decreased probability of bystander CPR. Liver disease may be an indicator of alcohol abuse, and alcohol abuse and psychiatric diseases are probably over-represented among home-less and destitute people. Some studies have reported that disagreeable characteristics and fear of personal health are barriers to prevent bystanders to perform CPR on strangers.21 The decreased probability of receiving bystander CPR amongst women with arrests only in public locations indicates a gender specific bystander related barrier, which is in line with other recent results.22 That the gender disparities associated with bystander CPR is limited to public locations points to gender specific biases among strangers, perhaps reluctance to remove clothes, or push hard on a woman’s chest. The marked association between the low-income group and decreased probability of bystander CPR in arrests in both locations in this study suggests a social imbalance, which concurs with previous reporting of low socioeconomic status being associated with a lower probability of bystander CPR and survival.23 As barriers related to gender or socioeconomic status could be influenced through CPR training and national CPR campaigns directly targeted to e.g. people with low socioeconomic status, these factors should be investigated further. Limitations Some of the registered cardiac arrest cases had missing data (see Supplementary material online, Table S2). We accounted for this by performing both multiple imputation analyses and complete case analyses. We found no indication that missing data had introduced bias into the complete case analyses of our study. Five patients were lost to follow-up due to emigration. We had no information on quality, timelines or length of CPR, which could have had an effect on survival and differ according to location of arrest. The time from arrest to first rhythm analysis by the EMS is only an approximation and should be interpreted cautiously. As we apart from diabetes rely on discharge diagnoses from hospitals and outpatient clinics in identifying comorbidity, patients with comorbidities who were only followed by their general practitioner and not previously admitted for such conditions were not identified. Additionally, it was not possible for us to assess whether the diagnosis of anoxic brain damage was withdrawn in the follow-up period due to improvement in neuro-cognitive functions. Due to a limited number of 30-day survivors of arrest in the beginning of our study period, our trend analyses involving anoxic brain damage or nursing home admissions should be interpreted with caution, particularly in residential arrests, see Supplementary material online, Table S3 for number of observations per year. Finally, the observational nature of the study makes any conclusion on causality impossible and we cannot identify the causal impact of each national initiative on bystander CPR and survival. Conclusion Bystander CPR and 30-day survival rates more than doubled in both residential and public locations during 2001–2014. Bystander CPR was significantly associated with decreased risk of anoxic brain damage or nursing home admission in both locations. However, amongst 30-day survivors the rates of anoxic brain damage or nursing home admission decreased only significantly during the study period in public locations. These findings indicate that nationwide initiatives to increase bystander resuscitation have equal effect in residential and public locations. To further improve survival, particularly amongst residential arrests, alternative methods may be warranted, such as layperson first-responders or targeted health campaigns. However, the less favourable patient characteristics in cardiac arrest patients in residential locations may limit CPR rates and survival rates to achieve the levels in public locations. Supplementary material Supplementary material is available at European Heart Journal online. Acknowledgements We extend our sincere thanks to the Danish Emergency Medical personnel who completed the case report forms for the Danish Cardiac Arrest Registry. Funding This work was supported by research grants from the Danish Foundation Trygfonden, who also supports the Danish Cardiac Arrest Registry with no commercial interest in the field of cardiac arrest. G.H.G. is supported by an unrestricted clinical research scholarship from the Novo Nordisk Foundation. F.F. and F.K.L. have both received unrestricted funding from The Laerdal Foundation. Conflict of interest: K.K. reports having received research grant from The Laerdal Foundation, and speaker’s honoraria from Novartis. S.M.H. reports receiving support from The Danish Heart Foundation, The Danish Foundation Trygfonden, and The Laerdal Foundation. L.K. reports receiving support from The Danish Foundation Trygfonden. References 1 Hasselqvist-Ax I , Riva G , Herlitz J , Rosenqvist M , Hollenberg J , Nordberg P , Ringh M , Jonsson M , Axelsson C , Lindqvist J , Karlsson T , Svensson L. Early cardiopulmonary resuscitation in out-of-hospital cardiac arrest . N Engl J Med 2015 ; 372 : 2307 – 2315 . Google Scholar Crossref Search ADS PubMed 2 Rajan S , Wissenberg M , Folke F , Hansen SM , Gerds TA , Kragholm K , Hansen CM , Karlsson L , Lippert FK , Køber L , Gislason GH , Torp-Pedersen C. Association of bystander cardiopulmonary resuscitation and survival according to ambulance response-times after out-of-hospital cardiac arrest . Circulation 2016 ; 134 : 2095 – 2104 . Google Scholar Crossref Search ADS PubMed 3 Kragholm K , Wissenberg M , Mortensen RN , Hansen SM , Malta Hansen C , Thorsteinsson K , Rajan S , Lippert F , Folke F , Gislason G , Køber L , Fonager K , Jensen SE , Gerds TA , Torp-Pedersen C , Rasmussen BS. Bystander efforts and 1-year outcomes in out-of-hospital cardiac arrest . N Engl J Med 2017 ; 376 : 1737 – 1747 . Google Scholar Crossref Search ADS PubMed 4 Wissenberg M , Lippert FK , Folke F , Weeke P , Hansen CM , Christensen EF , Jans H , Hansen PA , Lang-Jensen T , Olesen JB , Lindhardsen J , Fosbol EL , Nielsen SL , Gislason GH , Kober L , Torp-Pedersen C. 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 . Google Scholar Crossref Search ADS PubMed 5 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 . Google Scholar Crossref Search ADS PubMed 6 Kitamura T , Iwami T , Kawamura T , Nitta M , Nagao K , Nonogi H , Yonemoto N , Kimura T. Nationwide improvements in survival from out-of-hospital cardiac arrest in Japan . Circulation 2012 ; 126 : 2834 – 2843 . Google Scholar Crossref Search ADS PubMed 7 Iwami T , Hiraide A , Nakanishi N , Hayashi Y , Nishiuchi T , Uejima T , Morita H , Shigemoto T , Ikeuchi H , Matsusaka M , Shinya H , Yukioka H , Sugimoto H. Outcome and characteristics of out-of-hospital cardiac arrest according to location of arrest: a report from a large-scale, population-based study in Osaka, Japan . Resuscitation 2006 ; 69 : 221 – 228 . Google Scholar Crossref Search ADS PubMed 8 Weisfeldt ML , Everson-Stewart S , Sitlani C , Rea T , Aufderheide TP , Atkins DL , Bigham B , Brooks SC , Foerster C , Gray R , Ornato JP , Powell J , Kudenchuk PJ , Morrison LJ. Ventricular tachyarrhythmias after cardiac arrest in public versus at home . N Engl J Med 2011 ; 364 : 313 – 321 . Google Scholar Crossref Search ADS PubMed 9 Folke F , Gislason GH , Lippert FK , Nielsen SL , Weeke P , Hansen ML , Fosbøl EL , Andersen SS , Rasmussen S , Schramm TK , Køber L , Torp-Pedersen C. Differences between out-of-hospital cardiac arrest in residential and public locations and implications for public-access defibrillation . Circulation 2010 ; 122 : 623 – 630 . Google Scholar Crossref Search ADS PubMed 10 Herlitz J , Eek M , Holmberg M , Engdahl J , Holmberg S. Characteristics and outcome among patients having out of hospital cardiac arrest at home compared with elsewhere . Heart 2002 ; 88 : 579 – 582 . Google Scholar Crossref Search ADS PubMed 11 Swor RA , Jackson RE , Compton S , Domeier R , Zalenski R , Honeycutt L , Kuhn GJ , Frederiksen S , Pascual RG. Cardiac arrest in private locations: different strategies are needed to improve outcome . Resuscitation 2003 ; 58 : 171 – 176 . Google Scholar Crossref Search ADS PubMed 12 Hansen SM , Hansen CM , Folke F , Rajan S , Kragholm K , Ejlskov L , Gislason G , Køber L , Gerds TA , Hjortshøj S , Lippert F , Torp-Pedersen C , Wissenberg M. Bystander defibrillation for out-of-hospital cardiac arrest in public vs residential locations . JAMA Cardiol 2017 ; 2 : 507 – 514 . Google Scholar Crossref Search ADS PubMed 13 R Core Team . R: A Language and Environment for Statistical Computing . Vienna, Austria : R Foundation for Statistical Computing ; 2017 . https://www.R-project.org/. 14 Kronick SL , Kurz MC , Lin S , Edelson DP , Berg RA , Billi JE , Cabanas JG , Cone DC , Diercks DB , Foster JJ , Meeks RA , Travers AH , Welsford M. Part 4: systems of care and continuous quality improvement: 2015 American Heart Association Guidelines update for cardiopulmonary resuscitation and emergency cardiovascular care . Circulation 2015 ; 132 (18 Suppl 2): S397 – S413 . Google Scholar Crossref Search ADS PubMed 15 Patel N , Patel NJ , Macon CJ , Thakkar B , Desai M , Rengifo-Moreno P , Alfonso CE , Myerburg RJ , Bhatt DL , Cohen MG. Trends and outcomes of coronary angiography and percutaneous coronary intervention after out-of-hospital cardiac arrest associated with ventricular fibrillation or pulseless ventricular tachycardia . JAMA Cardiol 2016 ; 1 : 890 – 899 . Google Scholar Crossref Search ADS PubMed 16 Hassager C , Nagao K , Hildick-Smith D. Out-of-hospital cardiac arrest: in-hospital intervention strategies . Lancet 2018 ; 391 : 989 – 998 . Google Scholar Crossref Search ADS PubMed 17 Fordyce CB , Hansen CM , Kragholm K , Dupre ME , Jollis JG , Roettig ML , Becker LB , Hansen SM , Hinohara TT , Corbett CC , Monk L , Nelson RD , Pearson DA , Tyson C , Diepen S , van Anderson ML , McNally B , Granger CB. Association of public health initiatives with outcomes for out-of-hospital cardiac arrest at home and in public locations . JAMA Cardiol 2017 ; 2 : 1226 – 1235 . Google Scholar Crossref Search ADS PubMed 18 Priori SG , Blomström-Lundqvist C , Mazzanti A , Blom N , Borggrefe M , Camm J , Elliott PM , Fitzsimons D , Hatala R , Hindricks G , Kirchhof P , Kjeldsen K , Kuck K-H , Hernandez-Madrid A , Nikolaou N , Norekvål TM , Spaulding C , Van Veldhuisen DJ. 2015 ESC Guidelines for the management of patients with ventricular arrhythmias and the prevention of sudden cardiac death: the Task Force for the Management of Patients with Ventricular Arrhythmias and the Prevention of Sudden Cardiac Death of the European Society of Cardiology (ESC)Endorsed by: Association for European Paediatric and Congenital Cardiology (AEPC) . Eur Heart J 2015 ; 36 : 2793 – 2867 . Google Scholar Crossref Search ADS PubMed 19 Ringh M , Rosenqvist M , Hollenberg J , Jonsson M , Fredman D , Nordberg P , Järnbert-Pettersson H , Hasselqvist-Ax I , Riva G , Svensson L. Mobile-phone dispatch of laypersons for CPR in out-of-hospital cardiac arrest . N Engl J Med 2015 ; 372 : 2316 – 2325 . Google Scholar Crossref Search ADS PubMed 20 Zijlstra JA , Stieglis R , Riedijk F , Smeekes M , Worp WE , van der Koster RW. Local lay rescuers with AEDs, alerted by text messages, contribute to early defibrillation in a Dutch out-of-hospital cardiac arrest dispatch system . Resuscitation 2014 ; 85 : 1444 – 1449 . Google Scholar Crossref Search ADS PubMed 21 Axelsson Å , Herlitz J , Ekström L , Holmberg S. Bystander-initiated cardiopulmonary resuscitation out-of-hospital. A first description of the bystanders and their experiences . Resuscitation 1996 ; 33 : 3 – 11 . Google Scholar Crossref Search ADS PubMed 22 Blewer AL , McGovern SK , Schmicker R , Morrison LJ , May S , Aufderheide TP , Daya M , Idris AH , Callaway CW , Kudenchuk PJ , Vilke GM , Abella BS. Abstract 16409: gender disparities among patients receiving bystander cardiopulmonary resuscitation in the United States . Circulation 2017 ; 136 : A16409 . 23 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 . Google Scholar Crossref Search ADS PubMed Published on behalf of the European Society of Cardiology. All rights reserved. © The Author(s) 2018. For permissions, please email: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)

Journal

European Heart JournalOxford University Press

Published: Jan 14, 2019

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off