“Dying with a Smile,Just Knowing that Somebody’s Listened to Me”: End-Of-Life Care and Medical Assistance in Dying in Canadian Prisons

Abstract

Medical assistance in dying (MAiD) has been legal in Canada since 2016 and some incarcerated patients who are at the end of their lives are eligible for the procedure. Interviews with nine incarcerated men at a federal penitentiary in Canada provide insight into some of the ways that people who are navigating aging and end-of-life in prison think about MAiD. Interview themes are organized around: experience with death and dying; possibilities and barriers related to applications for release from prison at end-of-life; experiences of peer-caregiving in a prison palliative care program; support for MAiD and the expansion of eligibility criteria; what a good death looks like. Themes are contextualized alongside federal guidelines related to end-of-life care (EOLC) and MAiD for prisoners, highlighting that sound policy requires both generalizable principles and attention to nuance. MAiD rests on patient voluntariness, and thus autonomy over EOLC decisions is paramount for prisoners.

Increasing numbers of incarcerated people around the world are older adults and require end-of-life care (EOLC) (Burles et al., 2016; Hanson, 2017; Stensland & Sanders, 2016; Wion & Loeb, 2016). In Canada, individuals serving life sentences are admitted at the same rate, yet a stacking effect takes place as they age, where they account for an increasing proportion of the prison population. The federal institution, Correctional Service of Canada (CSC), is responsible under the Corrections and Conditional Release Act (CCRA) for those serving sentences of 2 years or greater (Corrections & Conditional Release, 1992). According to the Office of the Correctional Investigator (OCI) and Canadian Human Rights Commission (CHRC), people serving life sentences make up 26.4% of the entire federal prison population, double the amount since the year 2000, while only accounting for 4% of admissions per year (OCI & CHRC, 2019). Half of those aged 50 and older are serving indeterminate sentences (meaning there is no court-determined end date), including four in 10 women over 50 and six in 10 Indigenous people over 50.

Since 2016, medical assistance in dying (MAiD) has been legally accessible to Canadians covered under public health insurance. In this procedure, a physician or nurse practitioner can administer or supervise a patient self-administering a series of medications, at the patient’s request, that culminate in that patient’s death (Bill C-14, 2016).

When MAiD legislation was first adopted, the Government of Canada committed to establish a process for monitoring and reporting on MAiD to ensure that it is both adequately accessible and appropriately restricted. As a result of this monitoring process and court cases in the interim, the legislation was updated in 2021 to expand eligibility criteria to a) remove the requirement that death has become “reasonably foreseeable,” and b) to allow Canadians who have been assessed and approved for the procedure, but who risk losing capacity, to be able to sign a waiver of final consent. (Government of Canada, 2021). Additionally, an expert panel was convened to research MAiD for groups considered especially vulnerable, namely: mature minors, people for whom a mental disorder is the sole underlying condition (MD-SUMC), and people who desire to create an advanced request for MAiD (Council of Canadian Academies, 2018).

Concerns have been raised about vulnerable people being pressured into choosing an assisted death (Lerner & Caplan, 2015). Demographic research from jurisdictions that have had assisted death for decades indicates that there is no evidence of this happening. On the contrary, in jurisdictions where MAiD has been legal for years report that people who choose an assisted death are most likely white, highly educated, urban, and financially secure (Steck et al., 2014; Wineberg & Werth, 2003). Similar data has emerged in Ontario, Canada, where it was also found that MAiD recipients are less likely to reside in an institution (Downar et al., 2020). It is important to understand what people who are marginalized know about MAiD and what they perceive to be current issues related to it in order to both help prevent any pressure to hasten death, and to prevent any barriers to accessing assisted death. While the implementation of MAiD for those in Canadian prisons is fraught with ethical dilemmas, in order to align with federal policy stakeholders needed to determine how—not whether—MAiD ought to be offered to federally imprisoned individuals.

While conducting research for Shaw et al. (2019), which focused on perceptions and experiences of MAiD among highly marginalized populations, such as illicit drug users and people experiencing homelessness, a physician who worked with incarcerated people contacted the research team to ask whether prisoners could be considered marginalized within the context of our research. Given that prisoners have many of their rights and freedoms restricted, and because they often enter the prison system having suffered from poverty, systemic racism, and other negative social determinants of health, and that the health profile of older prisoners often mirror that of Canadian individuals 10 years older (OCI &CHRC, 2019), we concluded that as a cohort they are highly marginalized and ought to be included in our work. This paper offers insight into the thoughts of nine incarcerated men as they considered the morality and possibility of MAiD for Canadian prisoners.

Methods

After conducting a literature review, our methodological approach to investigating MAiD in Canadian prisons was two-fold: 1) we reviewed related Canadian guidelines and reports on MAiD in prison and 2) conducted interviews with men who were currently serving life-sentences in a Canadian prison. Although this pilot-project was built upon the work that we did for Shaw et al. (2019) we amended our research protocol to be prison-focused and obtained additional ethics approvals from both the University of Calgary (REB17-0191) and CSC.

Document Review

Beyond reviewing the limited scholarly literature that exists on assisted dying for prisoners in Canada (Downie et al., 2019) and globally (Devolder, 2016; Handtke & Bretschneider, 2015; Harrington, 2004; Messinger, 2019; Reichstein, 2020; Schildkraut, 2013; Shaw & Elger, 2016; Urwyler & Noll, 2020), we acquired and reviewed Canadian documentation related to EOLC and MAiD that was available publicly online (CSC 2017b Guideline 800-9: Medical Assistance in Dying; OCI reports for the years 2017, 2018, 2019, 2020; OCI and CHRC 2019 report Aging and Dying in Prison: An Investigation Into the Experiences of Older Individuals in Federal Custody), available through Access to Information and Privacy (ATIP) requests (CSC data on applications and approvals; CSC 2009 Hospice Palliative Care Guidelines for Correctional Services Canada), and made available through direct release to us (Parole Board of Canada [PBC] data on applications and approvals for release). In this paper, while we include a brief overview of some of these documents because they compose the policy context of MAiD and Canadian prisons, we focus our findings and the majority of our analysis and discussion on the primary research data that we elicited from the prisoner interviews.

Interview Sampling and Recruitment

Men who were serving long-term prison sentences at Pacific Institution/Regional Treatment Centre (PI/RTC)—a federal penitentiary and prison hospital in Abbotsford, BC—were interviewed about EOLC, including MAiD. PI/RTC is one of the only federal institutions that has an established palliative care program within a prison, where health and social service teams support incarcerated people at end-of-life, and where healthy prisoners are involved in the daily care of sick and/or elderly prisoners. The sampling method used for this research study was purposive, self-selected sampling, where CSC staff posted recruitment posters for an information session about the study at PI/RTC, and an in-person information session was offered by the researchers to potential interested participants. People who attended the information session, and anyone who had seen the posters but had not been able to attend the information session were able to sign up for an interview at a later date. Staff in the RTC also identified men who they believed would be willing and able to do an interview and shared the research invitation to them.

Interviews and Analysis

In August 2018, nine men were interviewed at PI/RTC using an interview guide that was developed by Shaw et al. (2019) and amended and approved for use in prison by CSC. Security clearance was required prior to entry, and Shaw conducted all of the interviews. Additional security arrangements (e.g., wearing an emergency call button) were required to permit Shaw to interview the men in a private room without the supervision of a prison guard, thus protecting the confidentiality of the participants, even though their identity was known to CSC staff. Shaw refrained from reviewing any files related to the participants to prevent bringing any potential bias into the interview room. Some of the participants assumed that the interviewer knew why they were in prison, and while they were not asked about their convictions, some disclosed details that they felt were relevant to their interview. The details of their convictions are intentionally not included in analysis.

The semi-structured interviews were transcribed, anonymized, and analyzed in Dedoose using an inductive thematic coding framework influenced by Gale et al. (2013) for working with multidisciplinary teams in health and social science research. The analysis steps included: initial comparative coding; discussion and agreement on codes (parent and child) and code definitions; testing the framework on a sample of transcripts; coding of all transcripts, where codes were added and clarified as needed. All quotations and references to the interviewees are attributed to pseudonyms.

Research Interruptions and COVID-19

The intention was to return to PI/RTC to conduct additional interviews with both prisoners and CSC staff. In 2019, Shaw applied for and received funding from the Social Sciences and Humanities Research Council of Canada for the project “Exploring Palliative Care and Medical Assistance in Dying in Canadian Prisons” in order to conduct more interviews with incarcerated people. Despite initial interest to expand our research protocol through external funding, CSC denied access to prisoners for further interviews, citing concerns about operational impacts at the site level, including the (in)ability to offer follow-up supports after the interviews. Interviews with CSC staff were supported and were scheduled to take place in March—April 2020.

When COVID-19 restrictions were imposed on CSC (and the rest of the country), the scheduled interviews were temporarily suspended. In March of 2020, CSC suspended all research activities within its institutions indefinitely. With the inability to conduct further interviews with either incarcerated people or staff of CSC, we extended the scope of our study to include MAiD providers who had assessed and/or provided MAiD for someone who was within a correctional facility at the time of their request. Those interviews are complete and are in the process of being written for publication.

Results

Document Review

We identify challenges in delivering quality care that appear in relevant policy and guidelines. CSC’s current guidelines (800-9) on MAiD present several points where incarcerated patients may not receive comparable quality of care to patients requesting MAiD in the general Canadian public, thus rendering them in contravention of both federal policy (Corrections & Conditional Release, 1992; Supreme Court of Canada, 2016) and international prison human rights norms (United Nations, 2005, 2015). For example, the guidelines do not provide information on possible recourse for a patient should there be a finding of ineligibility by either assessor and expressly prevent a patient from seeking a further opinion if either of the two assessors finds them ineligible. In contrast, Canadian residents outside the prison system have the option of seeking another assessment should their healthcare provider decline eligibility, and some Canadian jurisdictions require a provider who refuses to assess based on conscientious objection to refer the patient to someone who is willing to do an assessment. The guidelines assume that MAiD would be provided outside of CSC, requiring a successful parole application or a Temporary Absence to a community hospital while under CSC supervision. Any request from a patient to remain in place for MAiD provision requires approval from senior officials. For a detailed overview of our analysis of Guideline 800-9, we direct the reader to Driftmier & Shaw, (in press).

The CSC 2009 document Hospice Palliative Care Guidelines for Correctional Services Canada—which is the only health-related CSC document that is not publicly available and must be requested via an ATIP request—endorses the Canadian Hospice Palliative Care guidelines on palliative care, but with a strong focus on security as a baseline from which patient-centered care can be offered “whenever possible.” Furthermore, as Hudson et al. (2019) have highlighted, “no scholarly research has been published that specifically addresses dying or palliative care among Canadian prisoners, either at the federal or provincial level, making it difficult to know how these guidelines are actually implemented in practice” (p. 7).

An ATIP request revealed that as of August 2020, there had been 11 requests for MAiD from 11 prisoners in CSC, of which three were granted and occurred (according to CSC) in the patient’s preferred location of death. These three MAiD cases were documented with varying degrees of detail in reports by the OCI. The first occurred in a community hospital after a year of waiting for approval of parole by exception, which was denied prior to the MAiD request (2018). The next occurred in custody at the prisoner’s request, with a prior application for standard parole denied and a subsequent appeal in process (2019; 2020). The last case occurred in a community hospital for an individual deemed by the OCI as unlikely to access release due to his dangerous offender status (2020). The OCI affirmed that, in their opinion, the medical professionals involved in these cases had acted compassionately, professionally, and in alignment with MAiD legislation.

The primary mechanism available to release for compassionate reasons at end-of-life are through parole by exception and through royal prerogative for mercy, though both also are available to individuals with other concerns. In March 2021, the PBC released to us the number of applications and approvals for each of the above, listed according to reporting years from 2011/2012 up to February 28, 2021 (several weeks short of the complete 2020/2021 reporting cycle: see Tables 1 - 4 below). In the 17 years leading up to CSC (2017a) most recent report on deaths in custody, 904 deaths occurred, of which 603 (66.7%) were deemed natural (not deaths due to murder, suicide, or accident). These numbers, when contrasted with the numbers of natural deaths in prison, illuminate that release at end-of-life is not readily accessible. For an in-depth discussion of compassionate release from federal prison, see Iftene (2017b).

Table 1.

Medical Assistance in Dying Eligibility in Canada as Per Bill C-14, 2016.

To Receive MAiD in Canada from 2016 to March 2021, a Patient Must Have met Each of the Below Criteria
Be eligible for health services funded by government in Canada (or would be but for minimum period of residence or waiting period)
Be at least 18 years old
Be capable of making decisions with respect to their health
Have made a voluntary request
Have given informed consent to receive medical assistance in dying after having been informed of means available to relieve suffering, including palliative care
Have a grievous and irremediable medical condition. This means
- Have a serious and incurable illness, disease or disability
- Are in an advance state of irreversible decline in capability
- That illness, disease or disability or that state of decline causes them enduring physical or psychological suffering that is intolerable to them and that cannot be relieved under conditions that they consider acceptable; and—their natural death has become reasonably foreseeable, considering all of their medical circumstances, without a prognosis necessarily having been made as to the specific length of time that they have remaining

Table 2.

Typical Process for Approval and Procedure of MAiD: CSC Guideline 800-9 versus Canada’s 2016 Bill C-14.

Steps	Canada	Correctional Service Canada (Guideline 800-9)
Request	A patient signs a provincial/territorial MAiD request form. Upon receipt by a health provider, an effective referral must be made in accordance with the local jurisdiction’s process	A patient signs a provincial/territorial MAiD request form
		The Institution’s chief health services has 5 days to meet with the patient and determine why MAiD was requested, provide information, and consider security intelligence that could have impact on request
1st assessment	Physician or nurse practitioner assesses if patient meets criteria for MAiD under Bill C-14	A physician or nurse practitioner internal to CSC physician has 7 days to complete an assessment to determine if patient meets criteria for MAiD under Bill C-14
2nd assessment	Physician or nurse practitioner assesses if patient meets criteria for MAiD under Bill C-14. If both assessors agree that patient meets criteria for MAiD, the patient is eligible for the procedure	A physician or nurse practitioner external to CSC assesses if patient meets criteria for MAiD under Bill C-14. If both assessors agree that patient meets criteria for MAiD, the patient is eligible for the procedure
	If one or both assessors determine that a patient is ineligible for MAiD, the patient may seek second opinion	If one or both assessors determine that a patient is ineligible for MAiD, the patient’s request for MAiD is denied
		MAiD may take place in a community hospital, while patient is on temporary absence
		If patient requests MAiD take place within a CSC facility, it may be granted “in exceptional circumstances, at the request of the inmate.” There must be “an exception… approved by the assistant commissioner, health services”
Provision	Provision of MAiD occurs at a date agreed upon between provider and patient. The provider may be a physician or nurse practitioner, and is routinely the second assessor	Provision external to CSC is considered default and “all release options will be considered.” Yet the policy also states: “Access to the community for MAID purposes will be facilitated via temporary absences.”
	Provision generally occurs at a place of patient’s choosing, and if not is considered a “forced transfer”	The provision in an external setting would meet policies and procedures of the hospital and the provincial requirements, completed by the second assessor
		If provision occurs within a CSC institution, “the procedure [must include a] health professional external to CSC”
Time duration	A total of 10 days must have been completed from date of signing request to date of provision	A total of 10 days must have been completed from date of signing request to date of provision
	Exceptions may occur in instances of rapid decline	Additional steps (detailed above) include a 5 -day period, a 7 -day period, and if request is made for provision within CSC, another step without a specified time limit

Note. MAiD = Medical assistance in dyingCSC = Correctional Service of Canada.

Table 3.

Parole Board of Canada Authority Parole by Exception Applications Received and Granted: 2011-04-01 to 2021-02-28.

Fiscal Year Received	Applications Received	Applications Granted
FY2011–2012	4	2
FY2012–2013	3	1
FY2013–2014	5	3
FY2014–2015	6	4
FY2015–2016	5	3
FY2016–2017	2	1
FY2017–2018	3	1
FY2018–2019	7	7
FY2019–2020	7	4
FY2020-2021 *	113	12
Total	155	38

Data Source: PBC-IDS, Data Extraction Date: 2021-03-07.

Note. Application numbers include all cases where a decision was rendered, an application was rejected or withdrawn or cases that are currently pending decision.

Each application can contain more than one decision (i.e., Day Parole and Full Parole).

In cases where there are multiple decisions in a review, a decision to grant has been prioritized (e.g., Day Parole Granted, Full Parole Denied) will be counted as a grant for the purpose of these statistics.

*

Figures for FY2020-2021 only include figures from 2020-04-01 to 2021-02-28.

Table 4.

Royal Prerogative of Mercy Remedy Requests Received and Granted: 2011-04-01 to 2021-02-28.

Fiscal Year Received	Requests Received	Remedies Granted
FY2011–2012	29	2
FY2012–2013	44	0
FY2013–2014	45	0
FY2014–2015	37	0
FY2015–2016	46	0
FY2016–2017	37	1
FY2017–2018	36	0
FY2018–2019	48	2
FY2019–2020	58	2
FY2020–2021 *	26	0
Total	406	7

Note. Excludes clemency granted to Habitual Offenders (2), as a result of the Self Defence Review (5) and Ordinary pardons granted to Wheat Farmers (10).

Excludes requests for relief from driving prohibitions under the CCRA.

*

Figures for FY2020-2021 only include figures from 2020-04-01 to 2021-02-28.

Interviews

Demographics/Participant Profiles

The nine participants included men between the ages of 49 and 72, all completing sentences in custody. They had diverse geographic backgrounds from across Canada. Two of the men (n = 2 22%) identified themselves as part of the palliative caregiver program where, as part of their prison work, they cared for elderly and dying men at the end of their lives. Three of the men (n = 3, 33%) identified that they were receiving palliative care from the program. One man mentioned that he is a cleaner in the treatment center. One man (n = 1, 11%) mentioned that he was a caregiver with a different program. The men all had varying religious beliefs, including Roman Catholicism, Judaism, Paganism, Indigenous spirituality, or belief in a Creator, and none identified as atheist or non-spiritual.

Theme: Experience with Death and Dying

They had varying degrees of prior experience with death and dying outside of prison. For example, Matthew’s wife died of cancer while he was in custody and he had faced several delays and denials of requests for leave to see her and help his son. Timothy grew up in the foster care system, and discussed knowing two other youth in the system who died of varying unspecified causes during that time. James’ father died when he was 8 years old. Suicide in prison, however, was a frequently indicated experience and concern. Some of the participants had limited experiences with death and dying outside of suicide prior to their time at PI/RTC, but now have become more accustomed to it on the unit or in prison. Tommy summarizes these experiences with death:

Couple of family members have died, but I wasn’t really super connected to them. So, it’s kind of disassociated. In here, however, I’ve seen a lot of guys die. A few guys have died right near me, one guy severed his femoral arteries down the hallway from me a few years back.

Later in the conversation he returns to the death that occurs in the PI/RTC: “and then the old guys that die, we’ve had a couple of them die here, and so thinking about it… it’s really sad to die in a cage.” Roger describes the concentration of death in a geriatric unit:

I’ve been here for two years, and I think six or seven inmates have died. That’s in one range, there are four ranges here that we have. In Ontario, we might have three die, but here it’s all concentrated, so it hits you more.

Theme: Possibilities and Barriers Related to Release

Release at end-of-life is a concern at the heart of all conversations surrounding MAiD and death in prison. The OCI has asserted that any prisoner at end-of-life, including MAiD applicants, should have expedited release under parole by exception (2017). As the interviews illuminate, there is a perception that appropriate transitional supports are not always available, that there is limited release from prison at end-of-life, and that such limited release may influence suicidality.

Matthew spoke about living on the street while on parole, and within 3 weeks was hospitalized and subsequently had his parole revoked, returning to prison for health concerns. He believed his parole was revoked due to the lack of supports that would mitigate the liabilities associated with the health condition he was in and the stigma that assumes a prisoner would sue:

I got sick very quickly, and then it was turned into a mental issue as well. I was disillusioned, or disorientated. That’s why they pulled their support, they were terribly afraid that I was gonna fall down the stair, because I was losing my mobility, body functions… They sent me to a community hospital, and I was there for – I can’t remember, a couple weeks?.. That’s one thing that upsets me, because a couple times, I’ve seen where someone’s got sick and gotten sicker, and was on the street, and they whisked through the support for them, on to the half-way house or CSC. Because they felt “we can’t do anything for them, we’re not equipped to take care of ‘em, and secondly we’re concerned about liability issues because if he falls down the stair, or something, we know a law suits coming our way.”

Just as he described challenges of poor access to adequate supports outside of prison ultimately leading to the revocation of his parole, so too does he describe fears of release:

The thought of getting out after [amount of decades]-something years, with no family out there really... Anyhow, looking at the outside, and the limited resources out there, I have very limited means of mobility issues. It’s going to be such a painful struggle to get by out there, I have nothing in common with… Say they put me in a seniors centre of some sort, what do I have in common with those people? They’re all talking about stories about growing their flowers garden, and I’m talking about sports that we had in the prison. I just didn’t feel I would fit in, and it was just easier to live out my days here with people I known all my life, a big part of my life, and you get institutionalized, you get familiar with and accustomed to your surroundings here…

Several men spoke of the high degree to which many of the aged men in the institution were no longer a risk of re-offending, and thus should no longer be in prison, including Roger:

We had one guy die here, just a couple months ago. I don’t even think he knew he was sick. We have another guy in here that’s got Parkinson’s, and there’s no reason why he should be in here. He can barely walk, and he’s not going to go out and rob a bank or anything like that. Put him in an institution, an old-age-home, and even if you have to start a prison old-age-home. One where they have a little more dignity..

Yet, when a prisoner’s death is imminent there is a perceived rush to get the individual to a community hospital just in time for death. This likely occurs as a form of temporary absence, a process that is subjected to fewer bureaucratic approvals and where the prisoner remains in the custody of CSC guards. Patrick articulated this often happening in a manner out of sight, and potentially without fellow prisoners having a chance to say goodbyes:

Do people die here? In the ward?

Generally, they get to the hospital if they’re going to die… In the 6 months that I’ve been here, no one has. No one’s passed away…. they would take them to the outside hospital because they’re that sick. Yeah. And then they bring the ambulance in through the back doors and take them out. We don’t generally see that, they lock us all up to do that kind of thing.

Is there a time for transition to say goodbye to the guys you’ve been working with?

I don’t know…

The limited approval of compassionate releases prior to an expected death is perceived to contribute to suicidal behavior within prison. Tommy describes another prisoner in advanced decline complete suicide, believing the reason to be his limited access to healthcare and limited possibility of release. “When we’re in jail, we’re not really able to get a transfusion or a… transplant”…

He committed suicide… out here. He’s been on dialysis, long-term dialysis, for quite a few years… I heard that he yanked his shunt, the (dialysis) tube that they put the fluids in…and he yanked ‘em out. That’s a hard way to go. But, I understand his pain. He’s been in for a long time. He got out on parole and then killed somebody else and came back in on a… they’re never gonna let him out. They wouldn’t even let him down to minimum for that. So that’s a lot of pain in there. So, he took that way out.

In contrast to - or in the absence of – transitions to care outside of prison, it is important to understand how the men experienced and perceived the quality of care that end-of-life patients receive inside prison.

Theme: The Palliative Care Program

Most of the participants had limited experience with EOLC prior to becoming involved with the prison palliative care program. Some were involved as patients, while others were enrolled in a peer caregiver program referred to as “the Pal Program.” Pal caregivers perform certain forms of hygiene, social care, and even administering medications in coordination with professional healthcare staff. Upon entering our research project, we had not known of the program, but learned about it through our interviews. Pal is a play on words, between palliative care and ‘pal,’ as in friend. The men who were Pal workers described the work as involving significant degrees of hands-on care, which illuminates how the peer program both enables and erects barriers to quality care. James detailed his regular duties to support the “gentlemen” he works with, including checking regularly on colostomy bags:

You have to measure how much urine is deposited in the bag so that the nursing staff knows, “okay he is urinating, he’s not really dehydrated or whatever.” You try and keep a check on whether the man uses the toilet to pass his bowel movements. You do his laundry for them, you help him dress, you help write their letters for them, or what we call canteen where we order shaving equipment, shampoos and stuff like this, or other food goods or munchie treats or whatever, depending on what it is that the person wants.

Patrick recalled a particularly traumatic incident for a Pal worker that was asked to retrieve the body of a prisoner who passed away on the toilet in his cell:

That guy, he’s not here anymore… it broke him. He became very angry and bitter, he challenged staff in ways that weren’t appropriate. They should never have asked him to move a dead person out of the cell. It just wasn’t cool. He struggled with a lot of emotional distress. Talked about having PTSD. I don’t know how you make that… thing. But when you talk to him about that, you could see that he was struggling with his emotions with regard to that. Shortly after I arrived, he was asked to leave.

The economic function of the Pal Program to lower operating expenses is clearly understood by James, contrasting the wages of those who would need to replace Pal workers were they to not be there. James said he earns $25 every 2 weeks for this work. As he and other Pal caregivers “can walk around in amongst the inmate population”, the prison authorities are “not too worried about things happening to us. In other words, we’re not escorted by guards.” However, unlike occupational therapists, nurses, or guards, Pal care workers are never truly off duty. As Patrick described, they face a unique difficulty in that they:

can’t leave the work area. So, although we’re developing relationships with these guys, that relationship is all the time. We don’t get to break from it, so the men that we are assisting are also our neighbours and our friends.

While some of the men discussed the erosion of family relationships on the outside, some shared how they forged bonds and community with their peers inside prison over their long periods of incarceration. James shared his empathy for a man he cares for in his Pal work:

He really has no family left out there in the world, or people that would like to contact and/or associate with him. So, the only people that he really knows are the people here in this institution.

Not only were relationships central to prison experience, but they were crucial factors in these men’s positive descriptions of healthcare. Several men found PI/RTC to be an exception of quality care, in contrast to other prison experiences they had. In contrast to other CSC institutions Timothy has been in, he emphasized the importance of staff:

You’ve got some remarkable people here… Some of them are outstanding human beings and professionals. And they’d be the first ones who’d probably turn red all over if they heard that. But they are. And they don’t get the recognition they deserve. They really don’t. And they should.

Will described the positive qualities of PI/RTC in contrast to barriers he witnessed at other institutions:

In other institutions… they would have needed an environment like this. Where it’s a treatment opportunity, where there’s nursing staff here, where there’s occupational therapists here, and there’s more interaction with them. There’s more hands-on.

Despite the positive qualities of care, the men commented on how PI/RTC subordinates the clinical needs of those in its custody to security needs. Patrick argued, with all the time his patients with cancer or dementia are alone in their cells, “they wouldn’t be able to say like ‘hey I need help,’ because they don’t know that that’s what that button’s function is.” When asked what resources were needed, he said:

There certainly needs to be more of a focus on the clinical aspect - the clinical needs of the men that we’re caring for, than the security needs that seem to dominate the… atmosphere and the area that we’re working in. I’m not quite sure how CSC is… pretending they’re providing care when what they’re really doing is providing security. So, a guy who no longer understands why he’s here or he’s committed to a wheelchair certainly doesn’t need officers walking around with mace on their belts. That’s not a necessary tool that they need. It doesn’t make sense to me…

As the men described, prisoners in this setting witness the strengths and the challenges of providing quality care to the aging and ill in prison.

Theme: Prisoner Support for MAiD

Access to MAID may be a small component within the continuum of EOLC in the prison context, yet it represents more than just a procedure. Bound up with it are perceptions and experiences of autonomy, agency, ethics, and choice around death. All the men were supportive of MAiD as it was legislated for the general public at the time of the interviews, as well as supported prisoners’ access to it. It was perceived of as a form of compassionate care and an appropriate intervention for suffering. “I look at this as not a violent act; this is something of peace,” said Will. Max likened MAiD to the compassion shown pets at advanced stages of decline:

It is going to go downhill, there’s a lot of pain, and a lot of suffering. We don’t do that to our pets, to our cats, to our dogs. If our cat or our dog is having a terminal disease, we’re usually compassionate enough to put them down.

The men reported significant awareness about the nature of the procedure as being for people at end-of-life. One individual, Max, knew of two high-profile figures in the fight for assisted dying in North America, and considered them “heroes.”

Further, the men were largely supportive of expanding eligibility to include some the changes later legislated in Canada in 2021, including for use of advanced directives in cases where a patient at end-of-life loses capacity to consent. Several spoke about the need to simultaneously protect individuals from pressure within their family systems to choose MAiD. We also heard expressions of support for expanding MAiD eligibility to individuals with a MD-SUMC diagnosis while assuring those individuals have access to adequate resources and that their capacity to consent is assessed and assured. Roger voiced his support for MAiD to those in pain, and was asked if it had to only be physical pain:

Emotional pain, if they’re, as I say, cognitive enough to know to say “I wanna die,” and what that meant. Some people will look at that board and say “it’s white,” but not know what that is.

Patrick articulated the need to reconsider the ethical boundaries erected between MAiD, suicide, and MAiD for an MD-SUMC of depression:

I get how people look at it and go, “well depression and giving up isn’t a reason to die.” But if you can’t find a reason to live, then how’s that different than being in pain and your quality of life is in such peril that you’re no longer able to do the things that you wanted to do? Which for that person, the person who’s committed suicide, where is that line for them? For the legal assisted side? We’re told that you have to know what you’re doing at the time you request it and know what you’re doing when you’re good with dying. So how is that… where is the difference in that?

Participants spoke directly to the concerns that prisoners may be pressured to choose MAiD. When asked if they ever felt pressured to end their lives, several participants responded with simple “no” answers. There was even a dismissing and comedic attitude expressed about this concern: Roger said “the only person I would encourage to end their life is whoever thought of this new menu system.” Yet, the same participant also identified his own suicidality related to the early portion of his prison sentence. One man, Patrick, spoke about pressures from family caregivers outside of prison. Several discussed general social inequalities in explaining concerns about pressure to choose MAiD, yet did not include prison. Matthew spoke about an interaction he had with a physician in another prison when he was first diagnosed with cancer, who proposed palliation instead of treatment because of his prisoner status and drug history:

He said, “you’ve been here long time, you’ve been a drug addict for a long time.” He said, “you don’t have much of a future,” he says, “tell you what: why don’t you just let nature take its course? And I’ll keep you happy, I’ll keep you comfortable”… He said “I’ll keep you comfortable.” I said, “what, you want me to give in on this?” He said, “oh it won’t do any good.” I said, “that’s not happening.”

Some identified suicidality as a legitimate reason to access assisted death for the general public while other participants disagreed. However, despite rejecting the concern that prisoners will be “pressured” to choose MAiD, several men identified the prevalence of prison weariness while at end-of-life as contributing to suicidality as a separate phenomenon from “pressure.” Further, four men also saw this as a legitimate reason to seek an assistance in dying. Patrick recalls when assisted dying was a significant topic in the news cycle:

It became part of the conversation in prison. And men were certainly circling that idea. I should have a chance for parole, but I’m doing a life sentence and served 12 years already, I’m eligible for parole for the last 10 years and I’m still not getting anywhere, I’m tired of doing all of this. And those conversations were happening over and over again. So, men were certainly talking about whether or not that would be okay. And wanting to know more about it and maybe making that part of their parole application. As “okay you guys win, how about you’re obviously not going to help me, let me die.”

James expressly identified his reasons for participating in the interview was to advocate for this idea, with the hope that he may access assisted death 1 day as an alternative to completing his sentence. Will also wished this option for himself, identifying a concern frequently heard from participants about the resources being used to keep them incarcerated:

I’ve considered it already, because I’ve got (decades) in. I look at it, if they could say “okay, you’re never getting out,” this is something that I would do. That I would consider.

Okay. Not because your body’s dying, but because it would be an alternative to your sentence?

Yeah… I look at: if you’re doing a long sentence and you’re never gonna get out, and the parole board’s always turning you down, and your CO and the CMTM parole officers are saying “no, no, no,” then you’re stuck doin’ time for the rest of your life. If the assisted dying is an option, it should be looked at, especially if there’s no chance of you ever getting out. There’s over 400 dangerous offenders in Canada, and a handful are out on the streets, so why not give us another option? Instead of having the tax-collectors pay millions of dollars, why not just give us that option to go to sleep?

While opposed to access to assisted death for suicidality in the general public, Will supports assistance in death as an alternative to a long sentence. He distinguishes between it and the death penalty as:

Something that the state gives you, or the government, or the courts. By us choosing, we’re choosing to end our own life. Because, who wants to do 30, 40, 50 years inside and grow old, when you can say “okay, I’ll end my time now.” It also opens up the doors for other people to come in; to do programs, to get the help they need instead of you taking up bed space.

Throughout conversations about assisted death as an alternative to prison sentences, we provided information to clarify that such an arrangement is not likely to be made legal in Canada.

We heard the men share concern about how MAiD would actually be allowed for prisoners who would qualify. Timothy described the generalbureaucratic and cultural barriers to access it:

When you come into the prison situation and you say something like assisted dying. What should happen? The first thing that you run into is the whole bureaucratic wall: “will they let, will they allow?”… It may be the right thing… But will it be bureaucratically feasible? When your poster went up, I heard a fairly senior staff member say, “oh yeah, right, not here.” Not a hospital. I heard the same, because I was around before women came into the system and I heard the same thing “oh yeah, right, that’s not going to happen.”

Three others shared concern about about prison staff treating interest in MAiD as if it were suicidality, and receiving measures seen as punitive and stigmatizing instead of helpful. For example, Patrick, a PAL caregiver, was asked about the concern someone might be put in segregation under suicide watch as a result:

Oh, yeah no, no you couldn’t (ask about MAiD). Anyone who did that would be immediately… there would be problems for them.

So how would someone ask about assisted death if they… let’s say they would be eligible on the outside, so they’ve got cancer, they’re dying, they’re going to be here for a long time. Who would they talk to?..

Yeah in here, they would not ever feel like they could have that conversation. Because they would be immediately… again the security approach would take over. And any privileges that you had would be set aside in order to ensure that you were not dying.

The participants were all articulate and thoughtful about aging, dying, palliative care, suicide, and assisted death. This culminated in responses to what they desired and how they imagined their own deaths.

Theme: A Good Death

For a specialized facility where death is more common and autonomy over one’s circumstances is limited, enabling quality EOLC is fraught. Central to such care being patient-centered is understanding participants’ perceptions of what constitutes “a good death.” When asked what a good death looks like, the men’s responses generally aligned with four themes: joy, company, comfort, and decision-making control at end-of-life. Roger emphasized the following ideal death, not for being quick, but for the enjoyment of his last moments:

For me, it would be up in a fighter aircraft, and just realizing I’m out of fuel and I have no control. What a way to go!.. One that’s - I would say, be thrilling. My idea was going into a dive, and looking out the window and saying “oops, there goes one wing”.

Others spoke of the importance of being in the company of friends, or family if they still had such connections. Matthew described providing company for another prisoner at end-of-life by reading religious scripture with him, and this other man expressed gratitude to him for it:

He managed to get to a palliative care place, and his daughter and grandchildren got to come up and see him. That was the first time he met his grandchildren, and then he died. That was where it hit me: God did bless him by going in there, that’s where he received the gift of God, and deliverance. All his word was summed up with him going there, meeting the grandchildren, and then passing away.

Three men described pain management and physical comfort as they pass, such as Will: “When you’re able to just close your eyes, and go to sleep.”

Having one’s wishes respected at end-of-life was portrayed as a form of compassion that is rare in prison life. For some, this included control of one’s own death during advanced decline and being able to “pull the plug.” Tyson spoke throughout his interview about his empathy for people who have long desired to die to be able to access an assisted death. Here, he responds to a prompt about compassion, signaling the primacy of having one’s wishes respected:

At the end of the day, it’s compassion. I think I could see a lot of people where if they are dying, if they are in a great deal of pain, whether it be mentally or physically, and then finally getting what they’ve desired for a while, I can see them dying with a smile on their face, just knowing that “somebody’s listened to me.”

Providing the care necessary to enable “a good death” lies at the heart of patient-centered care, and requires even greater attention in prison.

Discussion

This study reviewed documents and interviewed nine men from a CSC hospital unit to understand how EOLC broadly, and MAiD in particular, can be accessed by those in federal prison. Ultimately, the various concerns identified culminate in how the twin carceral and healthcare structures that administer EOLC can foster patients’ access to a good death.

The care provided at PI/RTC was described by participants in generally positive terms, especially in contrast to other CSC facilities. Positive factors included the relationships with healthcare staff and with peer caregivers, as well as the types of care accessed in the facility. Negative qualities included what was perceived to be a subordination of healthcare needs to security and carceral needs. Further, providing peer care came with traumatic and grief-laden experiences, and it was described in exploitive terms. Access to release was enmeshed with what constituted quality EOLC for the participants. There were no significant differences in responses from prisoners who were providing care through the Pal Program, versus those who were receiving care through the Pal Program.

The data released to us of applications and approvals for release demonstrate that there is limited access to these mechanisms. The number of approvals to application for parole by exception during the past 10 reporting years is 38–155, while the number of approvals to application for royal prerogative for mercy is 7 to 406. The number of individuals dying natural deaths and receiving palliative care stands in sharp contrast to the relatively smaller number of individuals applying for and being granted such release. Given that palliative patients pose minimal risk to the public due to their bodies and/or minds being incapacitated by illness, we offer that denying parole by exception for prisoners at end-of-life runs counter to the parole principle of facilitating release when there is no undue safety risk to the public. These concerns were echoed in themes raised by the men.

Significant concerns have been raised by the OCI that decision-making about MAiD is impaired if a patient has no material prospect for freedom from the prison system (2017; 2018; 2019; 2020; OCI & CHRC, 2019). If mechanisms for release are not available in practice, some incarcerated patients who are at end-of-life may ultimately request MAiD, even if their preference would be to receive palliative care in the community. Further, in MAiD cases where an individual is in hospital while on temporary absence, their voluntary consent to proceed with the procedure would be compromised since a withdrawal of consent prior to death would result in a return to prison. The concerns raised by the research participants demonstrate a perception that it is difficult to access release at end-of-life and that this unavailability may compound suicidal ideation and behavior. This is compounded by the uncertainty about whether there were “MAiD friendly” medical staff and a fear that even asking questions about MAiD might put a prisoner at risk of being placed in segregation on suicide watch.

Beyond emphasizing the importance of release protocols (whether expedited or not), the unique challenges of MAiD in the prison context extend beyond access to release. Choice and agency at end-of-life should be available within the prison itself as well, equivalent to what is offered in the community, and particularly for those who wish to die among their prison family (Handtke & Bretschneider, 2015; Urwyler & Noll, 2020). The CSC MAiD guidelines appear to present significant barriers to remain in a CSC institution if that were the patients’ preference. It appears that decisions about release or staying in place for MAiD rest with parole bodies and executives rather than health providers determining release on the basis of medical grounds (Driftmier & Shaw, in press). The practice of forced transfers is a common institutional barrier to patient-centered MAiD access, “when a healthcare facility requires a person who requests assisted dying to go off-site to receive—or even be assessed for—(MAID),” (Dying With Dignity Canada, n.d.). For those who are denied parole at end of life and are transferred to a community hospital while under the custody of CSC officers—a mechanism called Temporary Absence—a patient that rescinds consent to a MAiD procedure would then be returned to prison, compromising MAiD voluntariness (Downie et al., 2019). All the above challenges about release can compound the physical and emotional distress of a patient. The wishes of any person—including persons who are incarcerated—regarding the location of their death must serve as the primary factor for location selection.

Truly compassionate person-centered care would allow for a person who is incarcerated and dying to be able to choose either: a transfer to a community facility for care, or the ability to remain in a CSC healthcare facility that is equipped to support EOLC if they desire to do so. This runs parallel to the need to address existing barriers to access parole by exception or the royal prerogative for mercy and is not addressed by simply expanding availability and accessibility to compassionate release policies. While the above articulates concern about how MAiD may be accessible to those in prison, participants overwhelmingly expressed ambivalence that a prisoner’s wish to receive MAiD would be respected at all. Principles of self-determination, including respecting where an individual wishes to die and with whom, are paramount to the provision of patient-centered EOLC in the prison context.

Finally, participants largely supported the expansion of eligibility criteria for access to MAiD, such as those reflected in the 2021 legislative Bill C-7, An Act to amend the Criminal Code (medical assistance in dying). This is especially relevant to note given that Bill C-7 was passed after interviews with the prisoners took place, meaning that prisoners were (along with the general public) talking about the need to improve eligibility prior to such legislative debates. Of particular interest and relevance to discussions about MAiD for prisoners, is the availability of MAiD MD-SUMC. Deliberations about MAiD MD-SUMC ought to include discussions and policy decisions around the implications of the expansion of eligibility criteria for people in the custody of CSC. In addition to safeguards and capacity assessment measures that will be created for the general public, extra measures should consider both prison weariness (Handtke & Bretschneider, 2015; Handtke & Wangmo, 2014) and psychological suffering. If MAiD eligibility expands to include access to assisted dying for reasons of mental illness, consideration will have to be given to whether there ought to be a differentiation between MAiD requests due to an ongoing and incurable mental disorder, and MAiD requests due to psychological suffering that could be relieved if a person’s circumstances were different. In such cases, decisions will have to be made about the extent to which imposed suffering is an intended and inescapable consequence of incarceration.

Limitations

Interviews from this pilot study with nine incarcerated men cannot be generalized to larger populations, but the findings may assist in setting direction for future research. Since the interviews did not focus on the “Pal Program,” we did not elicit details about the program specifically, such as what training prisoners may or may not have received prior to becoming a peer caregiver. In the interviews that were scheduled with CSC staff members (but were canceled due to COVID-19), there was an expressed desire from staff who wanted to elaborate on the PI/RTC model of peer support care. There may be tensions around whether the program ought to be considered an exemplary standard of in-prison peer support, or whether it is inappropriate for care to be provided by peers. Further interviews with all: prisoners, CSC healthcare and non-healthcare staff, and CSC decision-makers are necessary in order to better understand the nuances of the peer-support program, and the possibilities, limitations, and realities of addressing MAiD requests from Canadian prisoners.

Conclusions

All of the interviewed prisoners had a basic understanding of MAiD and were supportive of it, and most were supportive of expanding eligibility requirements to access MAiD, including some supporting it for mental health reasons. A few discussed the desire for assisted death as an option for people serving life sentences who might want to end their life rather than complete a life sentence. It was clarified with participants that this is not something that is currently viable under Canadian law, nor is it something that we envision having in Canada in the near future.

In the palliative care program at PI/RTC, healthy prisoners participate in peer-support and caregiving of older/unhealthy prisoners. Most had limited experience with EOLC prior to becoming involved with the prison palliative care program as either a peer caregiver or as a patient. The peer caregiver program was described as both enabling quality care, as well as taxing in a manner potentially off-loading responsibilities better handled by professionals onto prisoners who earn less. Among the men who were interviewed, there was no perception of pressure for incarcerated people to request MAiD, but there were concerns about how asking about it might put them at risk (e.g., under suicide watch in segregation). Despite an understanding that MAiD is different from suicide, there was much discussion about the widespread prevalence of suicide and suicidal ideation. The limited approval of “compassionate release” prior to an expected death was thought to potentially contribute to suicidal behavior within prison, as was the experience of prison-weariness and psychological suffering due to incarceration.

Sound policies and guidelines require both generalizable principles and attention to nuance. When it comes to impending death, understanding and uplifting an individual’s dying wishes is central to patient-centered care. It would be difficult to justify an EOLC policy as either practicable or ethical if it denies autonomy to a patient that is otherwise mentally fit and capable of consent, on the basis of their legal status in the prison system. Especially because prisoners have had many freedoms removed, autonomy over end-of-life care decisions is paramount.