Introduction

Neurologic determination of death (NDD) is legally accepted as death in Canada and around the world.1 Once death has been declared using neurologic criteria, routine practice is to discontinue organ support (e.g., mechanical ventilation and vasopressors) unless deceased organ donation is planned. While NDD has been part of clinical practice for 50 years,2 it remains subject to misunderstandings3,4,5,6,7,8,9 due to philosophical and religious disagreements about what constitutes human death,10,11 complexities in the clinical practice of death determination,12,13 inconsistencies in terminology and language used to describe NDD,10 and cognitive dissonance involved in declaring as dead a person that continues to have some degree of biologic function.14,15,16,17 Finally, despite legal precedent supporting acceptance of neurologic criteria for determining death, NDD remains legally undefined in most provinces.

In some cases, differences in understanding can lead to situations in which families request prolonged organ support or object to discontinuation of organ support after NDD. Families may cite religious objections to the concept of NDD and the desire to continue supporting loved ones whom they regard as alive. In contrast, physicians may feel an ethical and societal obligation not to prolong the inevitable decline of the body of a deceased person.18,19 These complex scenarios can lead to legal conflict between the patient’s family and the medical team, a situation which occurred in the publicized cases of Taquisha McKitty20,21 and Shalom Ouanounou22 in Canada, and Jahi McMath19,23,24,25 in the United States. In these cases, legal conflict centred around whether and when organ support could be stopped for a patient determined dead using medically accepted criteria, and whether or not religious objections to the concept of NDD should prohibit removal of organ support. In prolonged disputes, both families and physicians face the risk of moral distress and its psychological consequences.26,27 Understanding which situations are likely to lead to conflict as well as how disagreements might be defused prior to legal involvement could help to prevent distress.

Recent surveys of physicians in the United States indicate 47% of adult and 61% of pediatric neurology practitioners have encountered requests for continued organ support following NDD.28,29 The prevalence of this phenomenon has not been explored in Canada, though evidence suggests requests for continued support may occur in up to 10% of NDD cases.30 Four American states currently offer accommodations for religious or moral objections to the use of neurologic criteria to determine death.29 Similar legislative accommodations are not presently offered in Canadian jurisdictions.21 Additionally, differences in healthcare funding models between the United States and Canada generate unique considerations about who should pay for extended organ support after NDD and may limit generalizability of findings to Canadian contexts. An understanding of whether and how Canadian physicians experience requests for continued organ support following NDD may help determine whether further guidance on this issue is required.

In this study, we sought to describe the prevalence and characteristics of requests for continued organ support following NDD, along with common strategies for dispute resolution as practiced in Canada.

Methods

We used a mixed-method design combining a survey with semi-structured interviews of Canadian critical care physicians. All participants read and acknowledged a letter of information and consent prior to completing the online, anonymous survey. The consent letter was reviewed with all interview participants. This study was approved by the Research Ethics Board of the William Osler Health System (REB File No. 18-0072). Since this project focused on the experiences of physicians, we did not engage with patients, allied healthcare workers, or the public when designing this study. Our study team included two experienced qualitative researchers (A.V.B., J.C.) as well as physicians with personal experience of a family request for continued organ support after NDD (A.H., A.L., J.M.S., S.D.).

Electronic survey

We created an anonymous, self-administered, electronic survey for Canadian critical care physicians in accordance with existing survey development guidelines.31 An initial list of potential survey domains (Appendix A) was narrowed down through an iterative process of review and testing with a small group of leaders in Canadian critical care practice, including four adult and two pediatric intensivists. To minimize survey completion time, the survey was reduced to 15–20 closed-ended questions (Appendix B). To test for construct and face validity draft surveys were pilot tested on seven critical care physicians with varying experience with NDD. Survey development and testing took place between August and November 2018.

A final version of the survey was uploaded to a secure online hosting platform (SurveyMonkey) and sent via email link in December 2018 to the Canadian Critical Care Society mailing list of critical care physicians in Canada (n = 448). An introductory email encouraging survey completion was sent in advance. The survey was also advertised during a plenary on NDD at the Canadian Critical Care Forum in Toronto in November 2018. A reminder email was sent after one week, two weeks, and one month. A question at the beginning of the survey excluded potential participants who did not practice critical care in Canada (e.g., health professionals and researchers who may have heard about the survey at the conference).

Survey data were analyzed using STATA/SE version 16 (StataCorp in College Station, TX, USA). Significance testing of categorical variables was performed using two-sided Fisher’s exact tests. Open-ended survey responses were included with interview transcripts for thematic coding.

In-depth interviews

To obtain a comprehensive understanding of physicians’ experiences with requests for continued organ support following NDD, we asked survey respondents to participate in an interview. Respondents who agreed were directed to a separate survey where they provided contact details for interview scheduling. The interview guide for the semi-structured interviews (see Appendix C) was built from questions in the online survey, but was designed to seek additional detail. Interview questions were open-ended, allowing participants to focus on aspects of requests for continued organ support after NDD most important to them, following accepted practice for qualitative methodologies.32

All interviews were conducted by phone or in person and digitally audio-recorded by A.V.B, and transcribed by A.V.B and C.M. Atlas.ti (Scientific Software Development GmbH, Berlin, Germany) was used for coding and thematic analysis. A qualitative method of constant comparative coding was used to identify recurrent themes.33,34 Thematic analysis findings were used to provide context and depth to results obtained from the anonymous electronic survey through a process of triangulation.35,36 All ten manuscript authors were involved in the final analysis of qualitative and quantitative data.

Results

The online survey was completed by 114/448 respondents (25% response rate). Five respondents indicated they did not practice critical care in Canada, and thus were ineligible for participation; and three respondents declined to participate after reading the initial study information page. The final quantitative analysis sample consisted of 106 surveys. A total of 17/106 survey respondents (16%) indicated they would be willing to participate in an interview; 12 physicians (71% of those initially interested) completed interviews with the study team (11 by phone, one in person).

Demographics

Survey respondent demographics are presented in Table 1. Respondents were predominantly male (75/106, 72%), between the ages of 35 and 44 yr (41%), had at least five years of critical care experience (96%), and worked with adult patients (86%) in academic teaching hospitals (81%). Overall, 60% of respondents were from Ontario (49%) and Quebec (11%), with 32% from Western Canada and the remaining 8% from Atlantic provinces. A third of survey respondents (31/106, 29%) held a formal role as a donation physician.37

Table 1 Demographics of survey respondents (n = 106 unless otherwise indicated)
Full size table

In-depth interview participants included physicians from across Canada, both adult (9/12, 75%) and pediatric (3/12, 25%) practitioners. Half (6/12, 50%) of interview participants held formal roles with organ donation organizations.

Requests for continued organ support and resulting legal actions

The majority (67/106, 63%) of survey respondents indicated they had declared death using neurologic criteria more than 20 times in their career (Table 2). Half of survey respondents (55/106, 52%) had personally interacted with a family request for continued support following NDD within the past two years of their clinical practice (Table 2); of these, 42/55 (76%) had encountered this type of request only once or twice within the past two years. We found no across-province differences in the reporting of family requests for continued organ support after NDD (Table 3). Of 49 respondents who answered a question about legal aspects of a request for continued organ support after NDD, 26 (53%) indicated at least one request had involved a threat of legal action (Table 4). Threats progressed to dispute resolution or tribunal in 24% (12/50) of reported cases, and to legal action in 18% (9/50) of reported cases (Table 4).

Table 2 Frequency of experience with family requests for continued organ support following NDD* (n = 106)
Full size table
Table 3 Frequency of experience with family requests for continued organ support across provinces
Full size table
Table 4 Threats of legal action related to family requests for continued organ support following NDD
Full size table

Of 12 in-depth interview participants, ten (83%) indicated they had personal experience with families who requested continued organ support following NDD. Interview participants described conflict as uncommon, but noted that addressing conflict caused substantial distress (Electronic Supplementary Material [ESM] eTable 1). Two participants described requests for organ support following NDD as very common (“as high as 50% of situations”). When asked for clarification, these participants indicated that these requests were usually to provide family members with additional time to be present when the ventilator was removed, and not because they disagreed with determination of death.

No interview participants had been formally involved in a court case surrounding requests for continued organ support following NDD, but one reported going to the Ontario Consent and Capacity Board, and one faced legal proceedings initiated by a family which were abandoned once the patient’s heart stopped spontaneously. Two participants described that part of their resolution process involved suggesting families obtain a legal injunction to prevent discontinuation of organ support. Other participants noted families took their own initiative to seek legal counsel (ESM eTable 2).

Perceived causes of requests for continued organ support

When prompted to choose reasons why they perceived families might refuse to discontinue organ support after NDD, 55% (56/101) of respondents felt it was due to inability to accept death rather than specific difficulty with validity of NDD testing (Table 5). There were no differences in perceived causes of requests for continued support between those with and without personal experience with this issue.

Table 5 Comparison of perceived reasons for family requests for continued organ support following NDD
Full size table

Thematic analysis of interviews with physicians about experiences with specific cases revealed perceptions that a loss of trust in the medical system, the influence of misinformation, and hope for miracles contributed most often to family requests for continued organ support after NDD (ESM eTable 3). Several interview participants also reflected that refusal to discontinue organ support following NDD might be related to “cultural issues” faced by immigrants and non-white families (ESM eTable 3).

Mitigation strategies for resolving disagreements and misunderstandings

The most commonly chosen strategies to address requests for continued organ support after NDD were delay of one to three days to allow families to accept the declaration (49/100, 49%), consultation with another physician within the hospital (55/101, 54%), and consultation with spiritual care services (41/100, 41%) (Table 6). Respondents who had personal experience with a request were more likely to indicate they would not provide an ancillary blood flow test (P = 0.004) and would not consult with their hospital bioethics service (P = 0.004) if they encountered a similar experience again.

Table 6 Strategies for resolving requests for continued organ support after NDD
Full size table

In the qualitative analysis, physicians who had experienced requests for continued organ support after NDD suggested that ancillary testing was unlikely to help with resolution in situations where families were “waiting for a miracle” or could not accept that death had occurred (ESM eTable 4). As one participant described, “…to a certain extent to get any type of ancillary testing became kind of a moot point because they [the family] were not… I could’ve done fifty ancillary tests, they were not going to accept that [the patient] had a fatal prognosis,” (Interview #7). Providing additional information about NDD was felt to have little impact in situations where families were perceived as unable to accept the patient’s death at all.

Participants also explained that while clinical ethics services were “helpful to talk to”, they did not offer new approaches for resolving disputes, prolonged the length of the dispute through ongoing discussions, and were perceived by families in some cases as representing hospital interests (ESM eTable 4).

Additional emergent themes from interviews

During open-ended and exploratory parts of the interviews, participants were encouraged to discuss what they felt was most pertinent to the issue of family requests for continued organ support following NDD. Discussion themes included: the need for a clear legal definition of criteria for determining death in Canada, the idea that requests for organ support following NDD may be part of a broader trend of increasing requests for continued life-support of patients with non-recoverable injuries, and the need for a more assertive approach in communication to relieve guilt surrounding end of life decisions (ESM eTable 5). Finally, two participants discussed cases in which families seemed to mount a resistance to NDD over the course of clinical care, describing families as “initially on board” but later disputing NDD (ESM eTable 5).

Discussion

This mixed-methods study is the first to broadly engage Canadian intensivists about their experiences with requests for continued organ support following NDD. Physicians with and without personal experience of the issue were willing to complete the survey and share opinions through interviews. The demographic profile of our respondents closely matches those recently reported by the Canadian Medical Association for the specialty of critical care.38 Our results illustrate that requests for continued support following NDD occur in Canada, with more than half (52%) of survey respondents from across the country indicating they had experienced some form of this issue within the past two years. Lewis et al. reported a similar prevalence in a survey of American neurologists.28 Together, these surveys suggest that requests for continued organ support following NDD are widespread throughout the United States and Canada. Evidence from in-depth interviews shows perceptions of what constitutes a request for continued support following NDD ranges from appeals for time for family to gather before ventilator removal to formal disagreement with the concept of NDD; this ambiguity may contribute to an overestimation of conflict. Differences in case presentation also underscore the need for multifaceted and contextual approaches to mitigation.

The emergent qualitative theme of the perceived need for a national definition of death highlights the added value of our study in the Canadian context. Many participants voiced hope that legal clarification of the definition of death would help to reduce the incidence of family requests for continued support after NDD. At present, the legal definition of death in Canada is “according to accepted medical practice”.1,20 Manitoba offers the single exception, with the explicit definition of “irreversible cessation” of “brain function” included in the Vital Statistics Act.39 Despite this definition, we found physicians from Manitoba had experience with family requests for continued organ support following NDD. In addition, Lewis et al.28,29 provide evidence that requests for continued organ support after NDD persist in the context of the Uniform Determination of Death Act,40 which enacted NDD as a legal basis for determining death across the United States in 1980. These findings imply family requests for continued organ support are likely to continue in Canada even if the legal definition of death is clarified. Consequently, it may be more beneficial to clarify practical and fair legal processes for religious and moral accommodations rather than attempting to circumvent requests altogether.

Our findings establish that within a sample of Canadian physicians experiencing requests for continued organ support following NDD, only 24% of cases moved to formal dispute resolution (e.g., through a local judge or special ethics board), and 18% proceeded to formal court cases. Although the absolute number is small, court cases surrounding conflicts over the declaration of death tend to be sensationalized in the media and engage large numbers of bedside staff as well as hospital administration. As a result, these situations are highly taxing for physicians, families, hospitals, and communities involved. Though legal clarification of the definition of death in Canada may not reduce the number of requests for continued support after NDD, it may help to prevent rare, but distressing, prolonged legal cases. Qualitative themes including underlying tensions between shared decision-making and paternalism, the potential influence of increased misinformation and mistrust in medicine, and the perception that those with lived experience of being religious or cultural minorities may struggle with end of life scenarios suggest that, in addition to legal clarification, broader and more complex strategies for ongoing social engagement surrounding the definition of death and the limits of life sustaining therapies are required.

Finally, our results show that physicians with experience with families requesting continued organ support after NDD were less likely to suggest ancillary testing, consultation with external specialists, and ethics consultations as approaches for dispute resolution. Though we did not explicitly ask respondents the maximum amount of time acceptable to provide families prior to discontinuing organ support, 47% were uncomfortable with providing more than three days. These findings are supported by in-depth interviews and indicate that while academic discussion is ongoing about how best to manage these difficult clinical situations,41 there is growing practical expertise in what not to do. We suggest further consensus-building, which also solicits input from family members to outline possible approaches for Canadian clinicians faced with requests for continued organ support following NDD.

Limitations

While providing a useful foray into a topic yet unexplored in Canada, this survey includes a number of limitations. First, our response rate was 25%. Though this is on par with the 21% response rate of a similar survey of American neurologists,28 our findings may not be representative of all Canadian critical care physicians. Additionally, participation rate in interviews was low and we did not reach thematic saturation; nonetheless interviews provide nuance and depth to our survey findings. It is important to note that this is a survey of intensivists about perceptions of family beliefs and motivators for decision-making, not a survey of families who have objected to discontinuation of organ support after NDD. The impact of this issue on nurses and other allied health professionals was not explored. This study focused on physician perspectives and thus did not engage with members of the public or allied health professionals as part of its design.

Conclusions

This survey provides the first description of Canadian intensive care physicians’ experiences with family requests for continued organ support following NDD. Family requests for continued somatic support following NDD were encountered by 52% of our sample and ranged from requests for time for family members to gather, to denial of death. Mitigation strategies should consider the multifaceted social contexts of these scenarios rather than focus only on lack of information or family understanding. Further research should seek to understand requests for continued organ support after NDD from family perspectives and to elucidate practical suggestions for ensuring high quality end of life care in these situations.