When is a person dead? The Canadian public’s understanding of death and death determination: a nationwide survey

We conducted a cross-sectional web-based survey of the Canadian public. Ethics approval was granted from the Ottawa Hospital Research Ethics Board (Protocol number, 20210369-01). All respondents provided informed consent electronically before completing the survey.

Demographic characteristics of the 2,000 respondents are summarized in Table 2. Half of respondents (n = 1,015, 50.8%) were women and were representative of the Canadian population with respect to age distribution and province of residence. Most respondents were university/college educated (n = 1,181, 59.1%), married or in a common-law partnership (n = 1,080, 55.3%), self-identified as “white” in ethnicity (n = 1,570, 80.4%), subscribed to a religion or faith (n = 1,105, 56.7%). Of the 2,000 respondents, 694 (34.7%) reported having supported a family member or close friend who died in an ICU and 679 (34%) indicated that they are uncomfortable with the topic of death.

In this cross-sectional nationwide survey of the Canadian public, we found that the majority of Canadians agree that a person is dead when presented with scenarios that conform to currently accepted standards for death determination by either neurologic or circulatory criteria. There was a higher level of agreement when presented with a scenario of DCC compared with DNC. Those who were unsure or stated not dead for both scenarios were less confident in their response, and the majority did endorse various factors that could possibly change their response to accepting the person is dead.

While over two-thirds of Canadians surveyed agreed that the man is dead in scenario 1 (DNC), it is notable that about a third of respondents believed that the man is not dead or were unsure. Those who responded unsure or disagreed with the death determination were less confident in their response. Previous studies on public perspectives have shown considerable variation in their agreement with DNC.^7,11‐16 Furthermore, in scenario 1, we found that most agreed that doctors should stop intensive care treatments (80%). This is slightly higher than in the Skowronski et al. study, where 75.7% indicated the patient’s “life support” should be removed.⁷ Both studies suggest that the degree of brain damage and patient’s quality of life are important in deciding to stop intensive care treatments. The formal determination of death may be less important with respect to this decision. Of note, we intentionally did not use the term “life support,” and this may have impacted responses, as the term “life” erroneously implies that the person is still alive. Use of terms such as “life support” in the context of DNC can be perceived as conflicting information to the public and to family members in the ICU, misrepresenting individuals who are already dead.^17,18 In contrast, it could also be argued that use of terms such as “brain death” may bias the respondents toward agreement with death determination.

There is a paucity of data on public perspectives of DCC.¹ In our study, we found that a higher proportion of the public agreed that the man is dead in a scenario of DCC compared with DNC. Nevertheless, nearly one-fifth of the Canadian public were still unsure or did not believe the man was dead in this scenario, with most in this group (nearly 70%) responding that not enough time had passed (five minutes) after the heart had stopped to ensure that he was dead. A recent international study of patients dying in the ICU following withdrawal of life sustaining measures provides evidence for this time period as the primary finding was that the longest period of pulselessness that was followed by resumption of cardiac and pulsatile activity was four minutes 20 sec.¹⁹ Education and provision of information based on research data may alleviate public concerns in this group. That said, additional questions exist among this group that require attention, including concerns that "even if his heart stopped, his brain may not have died.”

An Australian survey of the public found no differences in demographics of those respondents that agreed the patient is dead, were unsure, or believed the patient not to be dead.⁷ In contrast, our data revealed multiple factors that predicted opinions about death determination. In both scenarios, those who believed the man is not dead or were unsure were younger and subscribed to a religion, with the latter variable being a stronger predictor of non-agreement in the neurologic criteria scenario than in the circulatory criteria scenario. In the DNC scenario respondents also tended to be less comfortable with the topic of death, whereas with the DCC scenario they were more likely to have a high school education and be unmarried. Province of residence was the strongest predictor of disagreement that the person is dead in the DCC scenario; compared with Ontario residents, those residing in Quebec had two times the odds of disagreeing that the man is dead, suggesting that sociocultural factors may play an important role beyond respondents’ individual demographic characteristics.

While there is a notable number of Canadians who either disagree or are unsure of the determination of death regardless of the criteria used, there are clear opportunities identified to increase understanding and acceptance of these determinations. In our study, only 8.4% and 3.2% of those who disagreed that the man is dead in scenarios 1 and 2, respectively, said that nothing would change their mind. Having more information on how death is determined, seeing the results of brain imaging/tests, a third doctor’s opinion, and more time to process the information were the top responses that would help to agree with determination of death for those who disagreed or were unsure if the man is dead. These findings are complementary to our investigation of family members’ understanding and acceptance of DNC, where we found facilitators to acceptance of the determination of death included provision of and repetition of information over time including visual information such as seeing the imaging, and witnessing the determination, in particular the apnea test.²⁰ Improving public knowledge of death and how it is determined prior to learning of a loved one’s critical illness and death may improve understanding and acceptance during a time of grief and stress, during which it can be difficult to process information,^21,22 in particular around complex concepts of death. Further, it is imperative to continue to improve health care professionals’ positive communication methods, using strategies grounded in social science research, to better engage community members in discussions about death.

Interestingly, there is a high level of general interest in learning more about how death is determined in Canada. This includes those who disagreed with the determination of death in either scenario. Respondents with a higher level of education and those who had previously supported a family member/close friend who died in the ICU were more likely to endorse interest in learning more about death and death determination, perhaps reflecting their greater understanding of the complexities in this topic. Studies have shown that the media, including televised dramas, plays an important role in shaping public views of death.^1,23,24 Unfortunately, these sources also perpetuate misinformation on how death is defined and determined along with the implications, creating confusion and distrust.¹ Access to accurate information is essential for preserving public trust. Indeed, we found that very few respondents preferred to receive information about death determination from media sources, including social media and posts by non-medical public figures. That the Canadian public prefers learning more about this topic from reputable sources provides an important opportunity to correct misinformation and build trust in the health care system.

Our study provides information on the public’s preferred modes of receiving information. About half of respondents indicated that they would prefer to receive information through their own doctor or health care professional. Unfortunately, significant barriers exist within the primary care setting around having conversations about serious illness, including system level, clinician, and patient factors.²⁴ Evidence suggests that discussions about end-of-life care do not routinely occur in the outpatient setting.^24,25 That said, programs and interventions have shown significant increases in the frequency of discussions and in documentation of end-of-life wishes.²⁵ Early discussion or provision of written materials from health care professionals, with whom patients have formed long-term relationships and trust, on death and death determination may be combined with such programs. These programs could help prepare patients for end-of-life care discussions and also encourage further dialog, improving knowledge and understanding of the issues involved, while providing opportunity for clinicians to discuss and better understand individual values and wishes.

This study has several limitations. As with all surveys, our results may be prone to selection bias, which may skew the findings toward those who are mostly comfortable with the topic of death. In addition, the survey was only distributed electronically, so members of the public without access to the internet and electronic devices were excluded. Because survey questions were presented in the same order for all respondents, it is conceivable that the order of presentation of the cases (neurologic determination first, circulatory determination second) may have influenced or primed respondents. While we cannot exclude the possibility that some respondents may not have fully understood the scenarios or questions, we attempted to minimize risk by engaging members of the public in the development of the research protocol and survey tool along with rigorous pretesting of the survey content. Finally, our sample was composed of a high number of “higher than high school” respondents, and given the small number of minority groups, we were not able to evaluate other predictors such as ethnicity or specific religious affiliations.

This study also has several strengths. This is the first study reporting on the perspectives of the Canadian public with respect to death and death determination. While numerous international studies have explored public perspectives of brain death, this is also one of the largest studies to report on public perspectives on death determination by both neurologic and circulatory criteria. The nationwide sample consisted of respondents who were representative of the Canadian population in age, gender, and geographical region. We developed the survey using established survey methodology.⁸ Furthermore, we engaged with members of the public throughout the development of the protocol and survey instrument and conducted rigorous pretesting of the survey tool.

The findings of this study reveal several areas for further research. Qualitative data will offer a richer, more in-depth picture of the perspectives of Canadians to contextualize the findings of this survey. To this end, a focus group study is currently underway to further explore public perspectives regarding the definition and determination of death in Canada. Our study focused broadly on public perspectives and understanding of death and death determination; we did not specifically include cases or questions regarding organ donation. Death determination in the context of organ donation may impact responses and hence this area requires further investigation and can be contrasted to our findings in future studies.

Our study examined perspectives of the Canadian public with respect to death determination and found that the majority of Canadians agree with death determined by both neurologic and circulatory criteria. We found differences in perspectives and understanding in these cases, including less acceptance and uncertainty with DNC. Our study also shows that Canadians have a high level of general interest in learning more about how death is determined in Canada. These findings provide important opportunities for further public engagement.