Skip to main content
Journal of Rural Medicine : JRM logoLink to Journal of Rural Medicine : JRM
. 2025 Jul 1;20(3):182–188. doi: 10.2185/jrm.2024-055

Changes in the rate of do-not-attempt-resuscitation orders in out-of-hospital cardiac arrest: a comparative study between 2019 and 2023

Yutaro Kurihara 1, Takaaki Maruhashi 1, Jun Hattori 1, Yasushi Asari 1
PMCID: PMC12237578  PMID: 40636253

Abstract

Objective

Japan has the highest aging rate globally, with 29.1% of its population aged ≥65 years; however, the concepts of do-not-attempt-resuscitation (DNAR) and advance care planning (ACP) are less prevalent than in other countries. We aimed to survey DNAR orders in out-of-hospital cardiac arrest (OHCA) and compare them with previous surveys to examine changes in end-of-life care.

Methods

This prospective study examined all OHCA cases attended to by the Sagamihara City Fire Department between May 30, 2023, and February 15, 2024. Data were collected through emergency services and compared with those of a similar 2019 survey.

Results

The 2023 OHCA survey included 513 patients, a 25.9% increase from 2019. The mean age was 75.9 ± 15.1 years, similar to 2019 (74.9±17.7 years, P=0.986). A subgroup analysis of patients with DNAR orders revealed a significant decrease in DNAR orders, from 45 patients (11.4%) in 2019 to 27 patients (5.3%) (P=0.000). No significant differences were found in ambulance treatment services (airway security, peripheral intravenous route security, and drug administration) (7.4% vs. 8.9%, P=1.000). Emergency transport differed significantly (P=0.001), with non-transport cases (death confirmed by a house-call doctor on site) increasing from 2.2% to 33.3% in 2019.

Conclusion

This study revealed a decrease in OHCA cases involving DNAR declarations during the COVID-19 pandemic, reflecting a positive shift toward respecting individual dignity owing to increased ACP. However, certain DNAR cases still undergo resuscitation and transport, highlighting the need for legal DNAR protocol implementation to reduce emergency-setting challenges.

Keywords: advance care planning, do-not-attempt-resuscitation, emergency medical service, out-of-hospital cardiac arrest

Introduction

Japan has the highest aging rate in the world, with 29.1% of the population aged 65 years or older, and is expected to reach 37.7% by 20501, 2). Providing effective emergency and terminal care for older adults is urgent. Although some guidelines for end-of-life care exist, they have not been legislated3,4,5). In emergency care, it is common for older patients with out-of-hospital cardiac arrest (OHCA) to communicate their wishes in advance regarding resuscitation, often expressing their intention to refuse resuscitation (do-not-attempt-resuscitation [DNAR]). A 2019 survey by the Sagamihara City Fire Department found that only 12% of patients had prior DNAR notifications6), and 95.6% of those were resuscitated against their will, causing confusion and conflict for paramedics7). Reducing unwanted resuscitation would not discourage paramedics.

To guide future care based on a person’s goals, values, and preferences, the Ministry of Health, Labour and Welfare has promoted advance care planning (ACP): a process of communication and documentation, including advance directives8). ACP increases patient and family satisfaction while reducing stress and anxiety9).

Between the 2019 and 2023 surveys, the coronavirus disease (COVID-19) pandemic occurred. The COVID-19 pandemic emphasized the need for ACP10, 11), as older adults faced high mortality rates. The importance of treatment planning based on patient and family wishes, including prior declarations, has been emphasized, and communication tools have evolved to support them12, 13). In addition, the COVID-19 pandemic has accelerated the use of information and communication technology in health care, limiting physical contact and promoting home care and online medical services.

We hypothesized that older adults’ responses to DNAR intentions might have changed during this period. Therefore, this study aimed to replicate our 2019 survey on prior DNAR declarations in OHCA to examine changes in end-of-life behaviors.

Materials and Methods

Study design and setting

This prospective study examined all patients with OHCA cases handled by the Sagamihara City Fire Department between May 30, 2023, and February 15, 2024. The 2019 survey examined all OHCA cases handled by the Sagamihara Fire Department between May 30, 2019, and February 15, 2020. Sagamihara City, with a population of approximately 720,000 (26.9% aged 65 years and older), is located in northern Kanagawa Prefecture14).

Our hospital, the city’s only tertiary emergency facility, plays a key role in community medical control. Ambulance teams follow predetermined regional medical control regulations for field activities and hospital selection. In addition, resuscitation procedures, including airway clearance and epinephrine administration, are performed according to direct orders from the emergency physician (online medical control). Patients with OHCA are typically transported to the hospital; patients with underlying conditions that cause cardiopulmonary arrest (CPA) are referred to primary care clinics. Terminally ill patients with clear DNAR orders may be transported to other hospitals through home doctor intervention. Ambulance personnel receive direct instructions for patient care, including chest compressions and ventilatory support using a bag-valve mask during patient transport.

Data collection

An emergency response team was dispatched to all patients with OHCA in Sagamihara City. The following information was obtained from the victims’ families and facility staff using the same methods as in the fact-finding survey conducted in 2019. Each person’s information was written on paper, collected at the Sagamihara Fire Station, and digitized and entered into a Microsoft Excel spreadsheet. The obtained data were compared with the data collected in 2019. Missing data items were labeled as “unknown” or “unlisted”. Individuals without a declaration of intention were excluded from the group with prior DNAR orders.

Emergency medical services patient-related data

The following information was obtained: patient age, site of occurrence of the healthcare event, emergency requester, underlying medical condition, home healthcare intervention, DNAR order details, ambulatory outcome, 1-month outcome (Utstein Style), cerebral performance category (CPC), and overall performance category (OPC). Paramedics confirmed the DNAR order by consulting relatives and facility staff.

Prehospital factors

Prehospital information included transport time, emergency requester details, transport emergency team, CPA witnesses, bystander cardiopulmonary resuscitation (CPR), emergency team CPR, emergency medical interventions, hospital to which the patient was transferred (hereafter referred to as “transport hospital”), and transport hospital selection criteria. CPR discontinuation was defined as cases in which CPR was interrupted or not performed in a prehospital setting. These decisions were made by online medical controls and home physicians.

Study outcomes and statistical analysis

The primary outcome assessed changes in prior DNAR orders compared with those from 2019 (G2019). For the 2023 survey (G2023), only the group with prior DNAR orders was selected, and changes in emergency treatment and hospital destinations were analyzed as secondary endpoints.

Mean values were analyzed using an unpaired t-test, medians using the Mann–Whitney U test, and qualitative data using Fisher’s exact test and residual analyses. The two-sided significance level was set at 5% for all patients. Statistical analyses were performed using IBM SPSS Statistics for Windows (version 27; IBM Corp., Armonk, NY, USA).

To compare the growth rates between OHCA incidence and crude mortality, we calculated population-adjusted rates per 100,000 residents for both metrics. The statistical significance of the difference in growth rates was assessed using a Z-test. A two-sided P-value <0.05 was considered statistically significant.

Results

The primary outcome, DNAR orders, was significantly lower in G2023 (27 cases, 5.3%) than in G2019 (45 cases, 11.4%) (P=0.000). The secondary endpoint of ambulance treatment procedures did not change (P=1.000). The other secondary endpoint, hospital destinations, changed significantly (P=0.001).

This study included 513 patients with OHCA from G2023, corresponding to a 29.5% increase compared with G2019 (396 patients). Mean age did not differ significantly between G2023 (75.9 ± 15.1 years) and G2019 (74.9 ± 17.7 years, P=0.986). Similar to G2019, most OHCA cases occurred at home (70.2%), followed by facilities (16.8%). Deaths confirmed in the emergency department increased significantly from 339 cases (85.6%) in G2019 to 486 cases (94.7%) in G2023 (Table 1) (P<0.001). For CPC at 1 month, CPC 2 was more common in G2019 (adjusted residual of 3.02), with no other significant differences observed. OPC 2 was also significantly higher in G2019 compared to OPC 2 (adjusted residual=2.76).

Table 1. Comparison of G2019 and G2023 patient characteristics.

G2019 n=396 G2023 n=513 P-value
Age (years) mean ± SD 74.9 ± 17.7 75.9 ± 15.1 0.336
Place of occurrence
Home 260 (65.7%) 360 (70.2%) 0.449
Facilities 73 (18.4%) 86 (16.8%)
Hospital 4 (1.0%) 2 (0.4%)
Outdoor 38 (9.6%) 39 (7.6%)
Workplace 4 (1.0%) 2 (0.4%)
Other 17 (4.3%) 24 (4.7%)
Emergency caller
Family 234 (59.1%) 344 (67.1%) 0.000
Facility staff 78 (19.7%) 89 (17.3%)
Passerby 27 (6.8%) 46 (9.0%)
Other 49 (12.4%) 34 (6.6%)
Unknown/no description 8 (2.0%) 0 (0.0%)
Witnessed CPA 154 (38.9%) 238 (46.4%) 0.016
Bystander CPR 200 (50.5%) 234 (45.6%) 0.114
Initiation of CPR by arriving ambulance team 384 (97.0%) 503 (98.1%) 0.370
Primary disease
Malignant tumor 61 (15.4%) 75 (14.6%) 0.779
Heart disease 51 (12.9%) 78 (15.2%) 0.339
Cerebrovascular disorders 23 (5.8%) 46 (9.0%) 0.078
Chronic obstructive pulmonary disease 7 (1.8%) 13 (2.5%) 0.500
Dementia 9 (2.3%) 56 (10.9%) 0.000
Bedridden due to chronic or orthopedic diseases 46 (11.6%) 2 (0.4%) 0.000
Other 86 (21.7%) 223 (43.5%) 0.000
DNAR order 45 (11.4%) 27 (5.3%) 0.000
Home health interventions 17 (4.3%) 20 (3.9%) 0.719
Contacting home physicians and family physicians 91 (23.0%) 55 (10.7%) 0.000
Discontinuation of CPR 36 (9.1%) 56 (10.9%) 0.378
Emergency services
BLS only 294 (74.2%) 311 (60.6%) 0.000
Endotracheal intubated or supraglottis 199 (50.3%) 262 (51.1%) 0.841
Peripheral venous access 199 (50.3%) 262 (51.1%) 0.841
Drug administration 185 (46.7%) 259 (50.5%) 0.284
Transfer destination
Tertiary hospital 310 (78.3%) 374 (72.9%) 0.006
Secondary hospital 85 (21.5%) 126 (24.6%)
No transfer 1 (0.3%) 13 (2.5%)
Outcome
Outpatient death 339 (85.6%) 486 (94.7%) 0.000
Return of spontaneous circulation and hospitalization 54 (13.6%) 26 (5.1%)
Transfer from the emergency room (family hospital, etc.) 2 (0.5%) 1 (0.2%)
Unknown/no description 1 (0.3%) 0 (0.0%)
One month outcome_CPC (brain function category)
1 or 2 16 (4.0%) 14 (2.7%) 0.041
5 368 (92.9%) 488 (95.1%)
One month outcome_OPC (general function category)
1 or 2 16 (4.0%) 14 (2.7%) 0.092
5 368 (92.9%) 488 (95.1%)
Unknown, no description, not subject to input 2 (0.5%) 1 (0.2%)

G2019: 2019 survey group; G2023: 2023 survey group; SD: standard deviation; CPA: cardiopulmonary arrest; CPR: cardiopulmonary resuscitation; CPC: cerebral performance category; DNAR: do not attempt resuscitation; BLS: basic life support; EMS: emergency medical services; OPC: overall performance category.

Only patients with DNAR orders were selected for subgroup analysis (Table 2). DNAR orders were significantly lower in G2023 (27 patients, 5.3%) than in G2019 (45 patients, 11.4%) (P=0.000). The mean ages were similar: 85 (81–89 years) in G2023 vs. 83 (79–87 years) in G2019 (P=0.363). Verbal DNAR orders increased significantly in G2023 (85.2%) compared with G2019 (60%) (adjusted residual=2.25), while written orders decreased (14.8% vs. 26.7%, respectively) (adjusted residual=1.59). The locations where OHCA occurred were similar between G2023 and G2019: 59.3% vs. 57.8% at home and 37% vs. 42.2% at facilities, respectively. No differences were observed in ambulance treatment procedures (airway security, peripheral intravenous route security, and drug administration: 7.4% vs. 8.9%, P=1.000). Emergency transport differed significantly (P=0.001), with non-transport cases (death confirmed by a house-call doctor) significantly increasing from 2.2% in G2019 to 33.3% in G2023(Table 2).

Table 2. Comparison of patient characteristics of DNAR cases in G2019 and G2023.

G2019 n=45 G2023 n=27 P-value
Age (years) mean ± SD 84.8 ± 7.62 83.5 ± 6.80 0.501
Place of occurrence
Home 26 (57.8%) 16 (59.3%) 0.436
Facilities 19 (42.2%) 10 (37.0%)
Hospital 0 (0.0%) 0 (0.0%)
Outdoor 0 (0.0%) 0 (0.0%)
Workplace 0 (0.0%) 0 (0.0%)
Other 0 (0.0%) 1 (3.7%)
Emergency caller
Person or family member 25 (55.6%) 16 (59.3%) 0.907
Staff 19 (42.2%) 10 (37.0%)
Passerby 0 (0.0%) 0 (0.0%)
Other 1 (2.2%) 1 (3.7%)
Unknown/no description 0 (0.0%) 0 (0.0%)
Witnessed CPA 16 (35.6%) 9 (33.3%) 1.000
Bystander CPR 27 (60.0%) 10 (37.0%) 0.088
Initiation of CPR by arriving ambulance team 44 (97.8%) 24 (88.9%) 0.145
Primary disease 31 (68.9%) 26 (96.3%) 0.003
Malignant tumor 11 (24.4%) 7 (25.9%) 1.000
Heart disease 4 (8.9%) 6 (22.2%) 0.161
Cerebrovascular disorders 4 (8.9%) 6 (22.2%) 0.161
Chronic obstructive pulmonary disease 2 (4.4%) 2 (7.4%) 0.628
Dementia 4 (8.9%) 2 (7.4%) 1.000
Bedridden due to chronic or orthopedic diseases 9 (20.0%) 1 (3.7%) 0.079
Other 4 (8.9%) 6 (22.2%) 0.161
DNAR order 45 (100.0%) 27 (100.0%) -
Means of expressing DNAR intent 0.106
Written records 12 (26.7%) 4 (14.8%)
Oral only 27 (60.0%) 23 (85.2%)
Other 4 (8.9%) 0 (0.0%)
Unknown/no description 2 (4.4%) 0 (0.0%)
Home health interventions 7 (15.6%) 12 (44.4%) 0.000
Contacting home physicians and family physicians 27 (60.0%) 19 (70.4%) 0.062
Discontinuation of CPR 5 (11.1%) 6 (22.2%) 0.310
Emergency services
BLS only 41 (91.1%) 25 (92.6%) 1.000
Endotracheal intubated or supraglottis 2 (4.4%) 1 (3.7%) 1.000
Peripheral venous access 2 (4.4%) 2 (7.4%) 0.628
Drug administration 2 (4.4%) 2 (7.4%) 0.628
Transfer destination 0.001
Tertiary hospital 17 (37.8%) 5 (18.5%)
Secondary hospital 27 (60.0%) 13 (48.1%)
No transfer 1 (2.2%) 9 (33.3%)
Reason for not transporting
Home health interventions 1 (100.0%) 8 (88.9%) 1.000
Apparent mortality 0 (0.0%) 1 (11.1%)
Outcome
Outpatient death 43 (95.6%) 27 (100.0%) 0.525
Return of spontaneous circulation and hospitalization 2 (4.4%) 0 (0.0%)
Transfer from the emergency room (family hospital, etc.) 0 (0.0%) 0 (0.0%)
One month outcome_CPC (brain function category) 1.000
1 or 2 1 (2.2%) 0 (0.0%)
5 43 (95.6%) 27 (100.0%)
One month outcome_OPC (general function category) 1.000
1 or 2 1 (2.2%) 0 (0.0%)
5 43 (95.6%) 27 (100.0%)

G2019: 2019 survey group; G2023: 2023 survey group; SD: standard deviation; CPA: cardiopulmonary arrest; CPR: cardiopulmonary resuscitation; CPC: cerebral performance category; DNAR: do not attempt resuscitation; BLS: basic life support; EMS: emergency medical services; OPC: overall performance category.

The incidence of OHCA increased by 30.9%, from 54.1 per 100,000 population in 2019 to 70.8 per 100,000 in 2023. During the same period, the crude mortality rate increased by 48.3%, from 742.0 to 1,100.6 per 100,000 population. Statistical analysis revealed that the difference in growth rates between OHCA incidence and crude mortality (17.4 percentage points) was statistically significant (P=0.014).

Discussion

In this study, OHCA cases were higher in 2023 than in 2019; however, prior DNAR orders decreased significantly. In the group with DNAR, no significant difference was observed in the resuscitation rate by the emergency team in 2023 compared with that in 2019; however, the number of cases not transported owing to on-site death confirmation increased.

The COVID-19 pandemic has increased ACP in Japan, yet DNAR orders in patients with OHCA decreased compared with those in 201915). This decrease might not indicate a lack of ACP. Sagamihara City saw a 46.8% increase in deaths from 5,433 in 2019 to 7,979 in 2023, with no major changes in population and the number of hospital beds (6,238 beds in 2020 and 6,195 beds planned for 2025)16). Notably, the survey only included individuals who requested emergency services. The rate of increase in OHCA cases (1.29-fold) was significantly lower than that in the total number of deaths (1.47-fold) (P=0.014), which may reflect the combined effects of an increase in the number of deaths due to population aging, COVID-19-related deaths, increased cases where transportation was difficult, and the development of care at nursing facilities or through home healthcare. Consequently, the reduced proportion of DNAR orders among patients with OHCA suggests that patients with cardiac arrest involving DNAR declarations, which ideally should not require emergency medical services, are being appropriately managed outside the emergency system. This indicates a potentially desirable shift in practice.

This possibility is also supported by the increase in confirmed deaths in the field and in non-transportation cases in this study. Although some emergency requests were made despite DNAR declarations, it appears that resuscitation was not performed according to the patient’s wishes because of the intervention of visiting medical doctors, indicating that terminal home care was effective. In Japan, where the aging population is growing, the demand for home medical treatment is high, and interventions are increasing17). Home medical treatment interventions seem to have reduced unnecessary hospital transports and ensured that final medical examinations were conducted in the field.

As mentioned above, the unnecessary transport of patients with DNAR orders has decreased; however, it is unlikely to be completely eliminated. In cases where a DNAR declarant experiences OHCA and an emergency call is made, confusion often occurs at the emergency scene in our country, where a national emergency medical service protocol for such situations has not yet been established In Japan, emergency medical teams are required to perform resuscitation; however, legal justification for stopping it is limited18). About 50% of regions have policies for handling DNAR orders19). In some areas, patients with DNAR orders receive only basic life support (BLS) during transport, while in others, BLS is stopped20). In Sagamihara City, if a DNAR order exists and the family doctor is contacted, the patient is taken to the family doctor without emergency transport, and CPR is stopped for a house call. However, the response is left to the discretion of the on-site ambulance team, who still face problems such as difficulty in responding and extended stays at the site. This increases the mental fatigue of the paramedics in the field. Based on the present findings, we strongly recommend the establishment of DNAR-directed patient transport guidelines in Japan.

Nonetheless, this study had some limitations. First, there were some missing or “unknown” data. The information obtained on the emergency activities of the ambulance team was recorded in the questionnaire. Because the information was collected during busy emergency operations, time constraints led to data loss. Second, the patients who did not require emergency services were not counted. As noted, the number of deaths increased, exceeding the rate of increase in CPA cases requiring emergency care. Whether the presence or absence of DNAR in patients who died without an emergency call was unknown in this survey. Third, this study is based on the pre- and post-COVID-19 pandemic years, 2019 and 2023. The increase in CPA cases without an emergency medical services call may be due to changes caused by multiple factors, including the impact of coronavirus transmission, an aging population, and the impact of ACP dissemination.

Conclusion

This study revealed that the number of OHCA cases involving individuals with DNAR declarations decreased during the COVID-19 pandemic, likely owing to shifts in older adults’ perceptions of life and death, as well as the increasing prevalence of ACP. Furthermore, even in the patients who were called emergency in spite of DNAR declaration, the frequency of non-transport decisions has increased. Considering the essence of DNAR, this trend reflects a growing respect for individual dignity, indicating a positive development. However, certain OHCA cases involving DNAR declarants are still being subjected to emergency calls, resuscitation efforts, and subsequent hospital transport. While these cases are expected to decrease further in the future, they are unlikely to be eliminated entirely. Such cases often cause confusion in emergency settings and impose stress on emergency responders. Therefore, the legal establishment of a DNAR protocol to address these challenges is an urgent priority.

Conflict of interest/Funding

Author J.H. received the grant from the Research Promotion Program for Science and Technology for Fire Fighting and Disaster Prevention. The funding source had no role in this study’s design, practice, or analysis.

Consent for publication

We, the authors, have reviewed the final version of this manuscript and agree to its publication in the Journal of Rural Medicine (JRM).

Ethics statement

Approval of the research protocol: This study was approved by the Kitasato University School of Medicine and Hospital Ethics Committee (approval no. B22-205). This study involved human participants and conformed to the principles and amendments of the Declaration of Helsinki. Informed consent: The need for informed consent from the patients or their families was waived by the Ethics Committee because the study involved only a fact-finding survey with no interventions.

Author contributions

Y. K. and T.M. mainly designed the study and drafted the manuscript. J.H. and Y.A. provided supervision and guidance throughout the study. All authors have read and approved the published version of the manuscript.

Acknowledgments

The statistical analysis of this study was supported in part by Suzuki Medical Statistics.

Data availability statement

Data are available upon request from the authors.

References

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

Data are available upon request from the authors.


Articles from Journal of Rural Medicine : JRM are provided here courtesy of Japanese Association of Rural Medicine

RESOURCES