Suffering and medicalization at the end of life: The case of physician- assisted dying

‘Suffering’ is a central discursive trope for the right-to-die movement. In this article, we ask how pro- ponents of physician-assisted dying (PAD) articulate suffering with the role of medicine at the end of life within the context of a decriminalization and legalization debate. We draw upon empirical data from our study of Carter v. Canada, the landmark court case that decriminalized PAD in Canada in 2015. We conducted in-depth interviews with 42 key participants of the case and collected over 4000 pages of legal documents generated by the case. In our analysis of the data, we show the different ways pro- ponents construct relationships between suffering, mainstream curative medicine, palliative care, and assisted dying. Proponents see curative medicine as complicit in the production of suffering at the end of life; they lament a cultural context wherein life-prolongation is the moral imperative of physicians who are paternalistic and death-denying. Proponents further limit palliative care’s ability to alleviate suffering at the end of life and even go so far as to claim that in some instances, palliative care produces suffering. Proponents’ articulation of suffering with both mainstream medicine and palliative care might suggest an outright rejection of a place for medicine at the end of life. We further find, however, that proponents insist on the involvement of physicians in assisted dying. Proponents emphasize how a request for PAD can set in motion an interactive therapeutic process that alleviates suffering at the end of life. We argue that the proponents’ articulation of suffering with the role of medicine at the end of life should be understood as a discourse through which one configuration of end-of-life care comes to be accepted and another rejected, a discourse that ultimately does not challenge, but makes productive use of the larger framework of the medicalization of dying. Crown Copyright © 2016 Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND

license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

This is a momentous occasion, for my clients, for society, for this court. This case quite simply concerns matters of life and death. It may require the court…to determine if the state has the right to require family members, our friends, ourselves to endure intolerable suffering as a result of a medical conditionwhen that suffering is worse than life itself.

Joseph Arvay, At the Supreme Court of Canada, October 15, 2014

. Karsoho), jennifer.fishman@ . Wright), mary.macdonald@

vier Ltd. This is an open access art

1. Introduction

Lead counsel for the claimants in Carter v. Canada, Joseph Arvay, uttered the above as part of his opening statement to the Supreme Court of Canada (SCC). Carter was landmark litigation that chal- lenged the constitutionality of the Criminal Code prohibitions on physician-assisted dying (PAD; euthanasia and physician-assisted suicide). Less than four months after the hearing, the Justices released a unanimous decision striking down the prohibitions on PAD, giving the federal government a limited window of time to revise the law. On June 17, 2016, the Parliament passed legislation on PAD. Canada is now one of a growing number of countries in which the practice is legal.

Arvay’s statement above highlights the centrality of suffering as a discursive trope in the right-to-die movement. Indeed, Scherer and Simon (1999) have identified ‘suffering’ along with

icle under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/

H. Karsoho et al. / Social Science & Medicine 170 (2016) 188e196 189

‘autonomy’ to be the primary social movement frames used by the proponents of PAD. Considerations of suffering in the right-to-die movement, however, cannot be divorced from discussions about the role and place of medicine at the end of life. As Lavi (2001) argues, the right-to-die movement must be properly seen “in the medical context in which it arises and primarily as a solution to the problem of pain in dying” (p. 138). Implicit, too, in Arvay’s state- ment is an indictment on medicine’s failure to adequately address suffering. This article therefore aims to investigate how proponents articulate suffering with the role of medicine, particularly in the end-of-life context. We use ‘articulate’ to mean the process of forming discursive linkages between two different entities or concepts. In other words, how do the proponents construct the relationship between suffering and medical interventions at the end of life? This requires us to grapple with such questions as: What do the proponents identify as the primary causes of suffering at the end of life? Inwhat ways do they suggest medicine, including palliative care, is (un)successful in addressing suffering? In the context of PAD, how do the proponents conceive the role of med- icine in addressing suffering?

In order to answer these questions, we draw upon a set of original, empirical data from our investigation of Carter v. Canada. We begin by describing Carter in greater detail. We then describe two social phenomena that others have identified as transformative of the contemporary dying experience: the increasing use of life- extending interventions in mainstream curative medicine and the emergence and rise of palliative care as the paradigmatic end-of- life care modality. This description serves two purposes: to pro- vide readers with necessary context for many claims advanced by Carter’s proponents and to serve as a basis for discussion of our empirical data in the last section of the article. We then proceed to describe our studymethods. In our reporting of results, we find that proponents see curative medicine as complicit in the production of suffering at the end of life. Proponents draw limits around the ability of palliative care to relieve suffering; they further contend that in some instances, palliative care can actually produce addi- tional suffering. At the same time, proponents insist that physicians must be involved in any legal regime of assisted dying. Thus, we also find that proponents emphasize how a request for PAD can set in motion an interactive medical process that has the potential to alleviate suffering at the end of life. In the discussion section, we argue that proponents’ articulation of suffering with the role of medicine constitutes a discourse through which different config- urations of end-of-life care come to be rejected or accepted within the larger framework of the medicalization of dying.

2. Background

2.1. Contextualizing Carter v. Canada

Political efforts to legalize PAD date back to the late nineteenth century (Dowbiggin, 2002; Lavi, 2007). It was not until 1997, however, that the first law on physician-assisted suicide (PAS) went into effect, in Oregon. Thereafter, a quick succession of other medico-legal regimes appeared, including Netherlands in 2002 and recently California in December 2015. Although euthanasia is an ancient topic (Van Hooff, 2004), PAD as a medicolegal practice accessible to the public at large is a relatively recent phenomenon. There are now 13 jurisdictions, including Canada, that have decriminalized or legalized PAD.

Carter v. Canada is a watershed moment in the history of the global right-to-die movement. With Carter, Canada became only the second country in theworld, after Colombia, to have allowed for PAD on constitutional grounds. Moreover, Carter decriminalized not only PAS but also, for the first time in North America, euthanasia.

The case began in the Supreme Court of British Columbia (the province’s court of first instance) in 2011. It was then heard at the British Columbia Court of Appeal in 2012, and finally the country’s highest court in 2014. The claimants included Lee Carter and her husband Hollis Johnson, Gloria Taylor, William Shoichet, and the British Columbia Civil Liberties Association (BCCLA). Carter and Johnson had accompanied Carter’s mother to die at an assisted suicide clinic in Switzerland the previous year, an event that they made public immediately afterward. Taylor was a woman with Amyotrophic Lateral Sclerosis (ALS) and Shoichet was a family physician. The diversity of the claimants was meant to reflect the diversity of persons with stakes in the legalization of PAD.

The Carter claimants challenged the Canadian Criminal Code prohibitions on assisting in another person’s suicide and on con- senting to one’s death. The claimants’ legal arguments essentially advanced along the lines of autonomy and equality. The autonomy argument stated that ill patients ought to have the right to seek PAD in order to control the manner and time of their own dying. The equality argument stated that since attempting suicide was not a crime, the ban on assisting suicide had the discriminatory effect of preventing disabled persons incapable of suicide from taking their own lives. The SCC eventually agreed with the claimants’ autonomy argument and having done so, found it unnecessary to adjudicate the matter in terms of equality (for more details on the ruling, see Karsoho, 2015).

The right-to-die movement, like other social movements, developed within a socio-historical context that both enabled and constrained what could be accomplished by the proponents. In the rest of the section, we discuss in brief two important social phe- nomena that have radically transformed the dying experience in contemporary times: the growing use of life-prolonging technolo- gies in mainstream medicine and the emergence of palliative care. Many authors see these phenomena as constituting the larger process of the medicalization of dying and intersecting with the right-to-die movement in significant ways.

2.2. Mainstream curative medicine and the extension of life

Mainstream curative medicine is now ever more reliant on the sciences and technologies (Clarke et al., 2003). For persons nearing the end of life, such “technoscientization of biomedical practices” (Clarke et al., 2010) manifests itself in the normalization and routinization of life-extending technologies (Kaufman et al., 2004; Shim et al., 2006). These life-prolonging technologies have created new forms of dying (e.g., neurovegetative state) and at the same time remade the moral frameworks at the end of life (Kaufman, 2005; Kaufman and Morgan, 2005).

In Kaufman’s (2015) incisive ethnography on “ordinary medi- cine,” she notes how the biomedical research industry is producing evidence of effective therapies at historically unprecedented rate. Many of these therapies (e.g., implantable cardiac defibrillator) were originally intended as last resort options. Once insurable, however, they become standard care and “ethically necessary and therefore difficult, if not impossible, for physicians, patients, and families to refuse” (Kaufman, 2015, p. 7). Indeed, refusing these potentially life-prolonging therapies seems irrational or even downright morally wrong in a cultural context in which death is seen as bad. The problem then is that “few know when that line between life-giving therapies and too much treatment is about to be crossed…the widespread lament about where that line is located and what to do about it grows ever louder” (Kaufman, 2015, p. 2). The use of life-extending technologies reproduces and, at the same time, is made possible by the organising principle of main- stream medicine: the (mistaken) belief that life can be prolonged more or less indefinitely throughmedical interventions, a pervasive

H. Karsoho et al. / Social Science & Medicine 170 (2016) 188e196190

cultural ideology that Dumas and Turner (2007, 2013) call “pro- longevism” and which they view as producing more harms than benefits to persons at the end of life.

2.3. The rise of palliative care

Public concerns about the medical care of the dying in the developed world began to surface in the 1950s when systematic studies revealed the neglect of dying patients and under-treatment of their symptoms (Clark, 2002, 2007). These concerns about care of the dying were taken upmost notably by Cicely Saunders, who is widely acknowledged to be the founder of the modern hospice movement. Saunders founded the first modern hospice, St. Chris- topher’s Hospice, in London, UK, in 1967 (Saunders, 2000). The success of St. Christopher’s, together with Saunders’ prolific writing contributed to the development of “a new approach to the care of dying people which would harness together medical innovation in pain and symptom management with wider concerns for the practical and social needs of patients and families, as well as responsiveness to spiritual matters” (Clark and Seymour, 1999, p. 72). One of Saunders’ most important intellectual contributions to medical knowledge and practice is the concept of ‘total pain,’which argues that suffering is irreducible to physical pain and must be understood in its multiple dimensions: physical, psychological, social, and spiritual. In order to relieve suffering, care for the dying must therefore be similarly holistic (Clark, 1999).

The modern hospice movement quickly gained international following. Balfour Mount, a Montreal urologist, coined and brought the term ‘palliative care’ into wide usage, preferring its use to ‘hospice’ because in French the word ‘hospice’ referred to alms- house (a house for the poor, not the dying) (Lewis, 2007). Knowl- edge and practice of palliative care quickly spread to other countries such that 115 of the world’s 234 countries now have one or more palliative care services (Clark, 2007). Palliative care, how- ever, has not provided equal benefits to all patients. The develop- ment of palliative care is deeply rooted in oncology, “which has shaped the conceptual model of palliative care, produced some of its major leaders and innovators, and provided a population of patients with the obvious potential to benefit from a new approach to themanagement of those with advanced disease” (Clark, 2007, p. 430). This means that cancer patients are more likely than patients with other terminal illnesses to benefit from palliative care (Clark, 2007).

The ethos of palliative care is typically portrayed as antagonistic to the idea of PAD as an ethical EOLC practice (Bernheim et al., 2008). Hermsen and ten Have (2002) analyze how PAD is dis- cussed in palliative care journals and found that the “majority of the examined articles do not consider euthanasia as an ethically legit- imate act in the context of palliative care” (p. 524). Wright et al. (2015) analyze the representation of physicians’ perspectives on PAD in the Canadian print media and identify a palliative care advocacy discourse whereby physicians who self-identify as part of the palliative care community voice a strong and consistent mes- sage of opposition to PAD. Reasons given for opposing PAD include, but are not limited to, the ethical principle of respect for life, the ability of palliative care interventions to address suffering, and concern about the diversion of resources away from palliative care.

2.4. The medicalization of dying

The two phenomena just described attest to the ways in which dying in Western societies has changed radically since the Middle Ages. There are, of course, other processes (e.g., demographic transition) that underlie the changes and these have been explored elsewhere (see Ari�es, 1981; Kellehear, 2007; Seale, 1998; Walter,

1994). Starting in the mid-twentieth century, however, changes in dying are best characterised as medicalization (Howarth, 2007). As Broom (2015, p. 6) argues, “[e]ssentially, dying frommedical illness was gradually transformed over the course of the twentieth cen- tury into a medical challenge not just an existential moment… Increasingly medicalized, dying was drawn into various institutions and viewed as a site of medical expertise and professional skill” (emphases in the original).

The medicalization of dying is immediately self-evident when one considers the use of life-prolonging interventions in main- stream curative medicine. But what about palliative care? Some see palliative care as a pointed critique on the medicalization of dying. McNamara (2001) argues that palliative care “has the potential to disrupt the medicalization of death” (p. 76) and “has served as a symbolic critique of how dying people are managed in other medical settings” (p. 121). Others, however, see palliative care as contributing to the medicalization of dying. Broom (2015, p. 12), for example, views palliative care as part of the overall trend of the medicalization of dying: “Themedicalization of dyinge via hospice and specialist palliative care … had the potential to reinforce the cultural dispositions toward death, separating those who are dying from the community and their families, and further concealing this important point in the life course.” It is, of course, possible for palliative care to be a medicalizing and, at the same time, deme- dicalizing force. Indeed, Syme and Bruce (2009, p. 20) argue for a view of palliative care as a “social movement that augmented and opposed mainstream curative medicine as the hegemonic model of care” (emphases added). Such view is supported by a historical reading of Saunders’ original intentions: according to Clark and Seymour (1999, p. 64), Saunders actually wanted to “improve and extend medical care at the end of life. The debate of course was around the precise form which this medicine should take” (em- phases in the original).

The term ‘end-of-life care’ (EOLC) is best understood as a configuration of healthcare resources (i.e., people, practices, and technologies) mobilized at the last phase of a person’s life. While palliative care has become the paradigmatic EOLC modality today (Livne, 2014), it does not have a complete monopoly over EOLC. According to Connelly (1998), there exist two dominant logics in EOLC: on the one hand, there is the ‘dying well’ path inmedicine, as represented, practiced, and promoted by palliative care pro- fessionals. On the other hand, there is still the traditional ‘life- saving’ path of medicine whereby treatment is directed at curing the patient’s disease; the goal is always to prolong life and comfort care is secondary to this goal. For most dying individuals, then, the medicalization of dying has come to mean that their last days of life are strongly shaped by either or the interplay of these two EOLC logics. For proponents of PAD, the medicalization of dying serves as the platform from which to argue the moral imperative of new op- tions in EOLC. In the results section, we will show how participants in Carter v. Canada engaged both EOLC logics (the ‘dying well’ path and the ‘life-saving’ path) as inadequate in addressing, and in some cases even perpetuating, the fundamental problem of suffering in dying.

3. Materials and methods

Carter v. Canada represents for us a “critical case” (Flyvbjerg, 2001) through which we can fruitfully investigate how PAD pro- ponents articulate suffering with the role of medicine at the end of life. A critical case is not the same as a representative or a typical case; instead it is a case that is rich in information. Indeed, so extensive was the scope of evidence and actors involved in Carter that an Irish court in a subsequent PAD litigation noted that the review conducted by the Canadian trial judge was “enormously

H. Karsoho et al. / Social Science & Medicine 170 (2016) 188e196 191

detailed and comprehensive” [Fleming v. Ireland. (2013) IESC 19 (BAILII)]. By the time Carter reached the SCC, 97 witnesses and 26 interveners, along with the claimants, their legal counsels, and the Crown Counsels had participated in the case. The expert witnesses called on by the claimants and the government hailed from 7 different countries. The selection of a critical case is important if the goal of analysis is to generalize not to other cases (i.e., statistical generalisation) but to theory (Ruddin, 2006), as is our intent here.

Prior to data collection, the study received Institutional Review Board approval fromMcGill University. The data we analyze consist of all of the legal artefacts generated by the case (i.e., affidavits, trial transcripts, factums, court decisions), all amounting to over 4000 pages of texts, and in-depth interviews with key participants in the case. The first and third authors also attended the SCC hearing on October 15, 2014 and took observational notes. Our selection of interviews with participants is predicated upon the insight that not all actors are created equal in a controversy (Venturini, 2010); we therefore selected only those actors with the most impact in the case: the claimants, interveners, and witnesses whose opinions were cited by the judges in the case. In the case of the witnesses, for each participant who did not respond, declined, or was lost to follow up, we made sure to recruit another participant who could speak to similar issues. Our recruitment process resulted in 42 in- terviews. The interviews were designed to ‘speak’ directly to the legal data; the interview guide for each participant was tailored according to that participant’s legal documents. Our participants were interviewed either in person or over Skype. All interviews were audio recorded and transcribed. Informed consent, either written or verbal, was obtained from every participant. For those participants who chose to remain anonymous, we identify only the data source (e.g., interview, trial transcript). Those we name in this article have given us permission at the time of interview to identify them. Data collection for this article spanned 21 months from June 2013 to March 2015.

Analysis proceeded along an iterative process involving coding, memo writing, and literature review. We began by uploading all of the documents to Atlas.ti™. Karsoho then coded all of the docu- ments both deductively, using themes derived from the literature, and inductively for emergent themes. For this article, our analysis was initially guided by an analytical interest in the role of medicine in the debate over legalization of PAD. Karsoho reviewed all of the codes pertaining to this issue; during this process, ‘medicalization’ and ‘suffering’ emerged as “core categories” (Strauss and Corbin, 1998). Further elucidation of the relationship between these cate- gories occurred through memo writing and consultation with the literature. This initial analysis was then presented to the co-authors for multiple rounds of further analysis and clarification.

Our analytical approach to the data is guided by insights from studies of discourse in sociology of knowledge. Following Potter (1996), we understand discourse to be talk and text in action. Discourse here is understood to be part of a broader repertoire of human actions used to accomplish something. In order to under- stand how talk and text do things, we have to consider their “deployment in specific interactions and the nature of those in- teractions (Potter, 1996, p. 180).” This understanding of discourse guides our analysis in two interrelated ways. First, we interrogate the practical nature of actors’ text and talk, rather than its truth value. That is, we are constantly asking, ‘what are our study par- ticipants attempting to do here?’ What legal, moral, or political aims are advanced by describing suffering in particular ways? Second, we treat actors’ discourse not as a resource but as a topic (Gilbert and Mulkay, 1984). In other words, we do not take the words of our participants to be ‘true’ but attend to the ways in which meaning is produced and to what effects. Therefore, pro- ceeding from the assumption that language is “used to do things; it

is amedium of action” (Potter,1996, p.11) allows us to bemindful of the broader legal context in which our data is produced. We approach the data not as evidence of a ‘true’ or ‘real’ perspective on suffering and the role of medicine but rather as a strategic deployment of language by the participants to advance specific agenda.

In this article, we focus and present data on the proponents’ discourse; however, their discourse was necessarily constructed vis-�a-vis the opponents’. Thus, in a few places, we present data from the opponents’ discourse to provide greater clarity for readers. Our use of the term ‘proponents’ or ‘opponents’ is not intended to elide the diversity of opinions and positions within each ‘side’ in the debate. We recognize that our study participants may differ in the strength of their support for the claimants or the government. By proponents, we mean the claimants and all of the actors (in- terveners, witnesses) who are strategically enrolled by the claim- ants to advance their case for decriminalization. Opponents refer to all of the actors on the opposing side.

4. Results

Proponents’ discursive articulation of suffering occurs throughout the data. Expressions of suffering are most common amongste although not limited toe the claimants, lay affiants, and those expert witnesses who professionally identify as physicians. Lay affiants often speak about either their own illness or having witnessed their loved ones die, using such adjectives as “horrific,” “heartbreaking,” and “torturous” to describe their experiences. Physicians, on the other hand, often speak of their professional experience caring for patients with intractable suffering.

There is an overwhelming consensus among the proponents that only suffering arising from medically diagnosable conditions could ever justify the need for PAD, the sole exception being a representative from Right to Die Canada who told us during inter- view that she would also accept suffering from a non-medical condition as a justification (e.g., tiredness of life). Further, mental illness is discussed less as a source of primary suffering at the end of life, and more as a potential source of interference with a person’s ability to make a clear and rational decision around assisted dying. During the SCC hearing, for example, Arvay suggests that any existing mental co-morbidities (e.g., depression in a context of cancer) be treated before a patient be granted access to PAD. Finally, while intractable (physical) suffering could occur at any point in the illness trajectory, proponents emphasize suffering that occurs in the last phase of life.

Having described how suffering appears in our data, we now turn to the ways in which suffering is linked by proponents to the practices of mainstream curative medicine, palliative care, and assisted dying.

4.1. The complicity of mainstream curative medicine

Proponents evince an awareness of the larger biomedical context inwhich the contemporary dying experience is embedded: that medicine plays an increasingly important role at the end of life. Citing figures from Belgium, one EOLC researcher notes that med- ical end-of-life decisions are now implicated in half of all deaths in the country. According to this researcher, this means that “doctors are more and more responsible for decisions that have huge im- plications on the quality of life of the patients” (interview, Deliens).

While proponents acknowledge that “medicalization” prolongs life, it does not comewithout concomitant costs. In fact, proponents argue that it is the medical efforts to prolong life that render the dying experience difficult. Leslie Laforest, a lay witness with anal cancer, describes what she sees to be effects of the litany of

H. Karsoho et al. / Social Science & Medicine 170 (2016) 188e196192

medications that are needed to sustain life in the context of a ter- minal illness:

In order for people to maintain life in terminal illnesses, they very often have to be on wretched volumes of drugs that make them sick, that make them queasy, that make them extraordi- narily sad, that sink them into a depression regardless. So then you’re on this whole super highway of trying to balance the depressants; the drugs that they have to have to keep them alive is giving them, needs to be counteracted with drugs to try and lift their spirits like with [antidepressant].