The desire for hastened death in individuals with advanced cancer: A longitudinal qualitative study

Abstract

Research is needed on the desire for hastened death (DHD) in the context of advanced cancer in order to address the clinical, ethical, and legal questions that it raises. The goal of the present qualitative study was to understand the experience of the DHD as expressed by individuals with advanced cancer, and to understand how it evolves over time. Participants were 27 ambulatory patients aged 45–82 years with advanced lung or gastrointestinal cancer. Participants were recruited through theoretical sampling from outpatient clinics at a large cancer center in Toronto Canada, and were asked about the DHD in serial, open-ended interviews until the point of death. A total of 54 interviews were analyzed based on the grounded theory method. The experience of the DHD in the context of advanced cancer was found to be subsumed under three distinct categories: i) DHD as a hypothetical exit plan; ii) DHD as an expression of despair; and iii) DHD as a manifestation of letting go. Each category had unique temporal and qualitative characteristics. The identification of these categories may be important to inform future research on the DHD, the criteria for clinical intervention in individuals who express this desire, and the public debate about physician-assisted suicide and euthanasia for individuals with advanced disease.

Introduction

The desire for hastened death (DHD) has been the subject of critical enquiry, in part because it appears to violate the common social and religious expectation that the trajectory of life should not be unnaturally shortened. Indeed, the will to live has been regarded as a natural instinct of all living creatures (Carmel, 2001), and a strong association of suicidality with major psychiatric illness has been demonstrated (Conwell et al., 1996). Recent quantitative studies have found a strong DHD in patients with advanced cancer to be associated with symptoms of psychosocial distress (Breitbart et al., 2000, Rodin, Zimmermann, et al., 2007). However, the meaning and significance of this desire remain an issue of debate, including with regard to the justification for legalization of physician-assisted suicide and euthanasia (Chochinov, 2006). The DHD in patients with advanced cancer deserves further exploration, in light of the complex clinical, legal, and ethical questions that it raises (Mak, Elwyn, & Finlay, 2003). Ambiguity about what the DHD reflects may account for the absence of evidence-based guidelines to assist health professionals in responding appropriately to it (Hudson, Schofield, et al., 2006).

The presence of the DHD may be most clearly determined when it is based on the frequency of actions taken by individuals to hasten their death, i.e., attempting or completing suicide, or seeking/requesting physician-assisted suicide or euthanasia. Epidemiological studies have consistently indicated that suicide rates in cancer patients are low, although they are elevated compared to the general population. A number of studies have found suicide to be the cause of death in less than 1% of individuals with cancer (e.g., Björkenstam et al., 2005, Hem et al., 2004). Even in countries in which physician-assisted suicide (defined as death resulting from self-administered medication that was prescribed by a physician with the explicit intention of hastening death) and euthanasia (in which death results from medication administered by a physician) have been legalized, the reported incidence of these practices is still less than 8% of all cancer deaths (van der Heide et al., 2007). However, it may be argued that the documented rates of suicide, physician-assisted suicide, or euthanasia, underestimate the frequency with which such actions are taken or desired, and that indirect or covert expressions of the DHD are more common. This position is supported by studies in which 12–30% of physicians reported having received a request for assisted suicide from their patients (Back et al., 1996, Craig et al., 2007). More striking, physicians reported actually providing the relevant assistance in 24% of these cases, despite its illegality (Back et al., 1996). Further, research with patients with advanced cancer suggests that most find practices of physician-assisted suicide and euthanasia to be acceptable, and can imagine a future scenario in which they would consider utilizing these options (Emanuel et al., 2000, Wilson et al., 2007). These findings add particular urgency to the need for a more accurate understanding of the expressed desire of individuals with advanced disease to hasten their death.

The study of less overt expressions of the DHD has been advanced by the development of quantitative measures to assess this construct. Currently, the most commonly used scale is the Schedule of Attitudes toward Hastened Death (SAHD; Rosenfeld et al., 1999). Using this measure, Breitbart et al. (2000) found that 17% of cancer patients in a palliative care unit could be classified as having a high DHD. In other studies of this population, using the SAHD or similar scales, the reported prevalence of a high DHD ranged from 8 to 22% (Chochinov et al., 1995, Tiernan et al., 2002). Interestingly, the prevalence of the DHD in ambulatory patients with advanced cancer may be much lower than in patients in palliative care units. For example, Ransom, Sacco, Weitzner, Azzarello, and McMillan (2006) found that only 3% (2/60) patients with advanced cancer, recruited in an oncology outpatient clinic, had a high DHD score, as measured by the SAHD. Similarly, we found a high DHD in only 1.2% of 326 ambulatory patients with advanced cancer (Rodin, Zimmermann, et al., 2007). This large discrepancy between the prevalence of the DHD reported in ambulatory patients with advanced disease compared to patients hospitalized in palliative care settings suggests that its intensity and meaning may vary with the trajectory of the disease.

The cross-sectional quantitative research studies that have been conducted on the DHD have provided important information about its relationship to psychosocial and physical variables. A consistent finding is that scores on self-report measures of the DHD, depression, and hopelessness are correlated (Breitbart et al., 2000, Hudson & Kristjanson, et al., 2006, Rodin, Zimmermann, et al., 2007, Tiernan et al., 2002). Still, the need for longitudinal studies has been highlighted by a prospective study that showed that patients' self-reported will to live fluctuated over time, and that the strength of its associations with psychosocial and physical variables changed as a function of proximity to death (Chochinov, Tataryn, Clinch, & Dudgeon, 1999).

In several recent studies that investigated its meaning using qualitative methods in patients with advanced disease, the DHD was found to be a multidimensional construct with diverse underlying motivations (Johansen et al., 2005, Mak and Elwyn, 2005, Schroepfer, 2006). For example, in a study of seven patients who expressed to their physicians a wish to hasten death, Coyle and Sculco (2004) found this communication to contain at least one of nine messages, including a wish to draw attention to oneself as a unique individual, or a despairing cry, depicting the misery of the current situation. Similarly, in a study of six patients who were identified by their physicians as desiring a hastened death, Mak and Elwyn (2005) interpreted the DHD as multidimensional and influenced by the reality of disease progression, current suffering, a prediction of a future that is worse than death itself, a desire for good quality of life care, and the lack of a holding environment.

The available qualitative studies of the DHD provide valuable information, although they have some methodological limitations, particularly related to sampling. Some studies have recruited participants based on physicians' judgments (e.g., Coyle and Sculco, 2004, Mak and Elwyn, 2005), while others have utilized a convenience sampling strategy in which all patients in a setting were invited to participate (e.g., Johansen et al., 2005, Schroepfer, 2006). The first strategy, although commonly employed in qualitative research on the end-of-life, has been criticized on the grounds that biases of the physicians may skew the selection of potential participants (Ewing et al., 2004). The second strategy of recruiting from convenience samples is not informed by emerging themes, which may limit the validity and richness of qualitative findings (Morse, 1995). Importantly, all previous qualitative studies on the DHD were cross-sectional studies of palliative care patients and, therefore could not elucidate the emergence of the DHD at earlier stages of the disease nor how it fluctuates over time.

The goal of the present study was to contribute to the understanding of the DHD in the context of advanced cancer. It is unique in its longitudinal, qualitative design and in its use of theoretical sampling (Rennie, Phillips, & Quartaro, 1988). Ambulatory patients with advanced cancer, selectively recruited from a larger quantitative longitudinal study on the DHD (Rodin, Zimmermann, et al., 2007) were interviewed repeatedly until the time of their death, allowing us to learn about the DHD at varying self-reported levels of intensity. Also, the investigation of the DHD in the present study was enhanced by the focus on a defined population, namely individuals with advanced lung or gastrointestinal cancer, which have a progressive course, multiple physical symptoms, and an expected survival time of less than 2 years (Parkin, Bray, Ferlay, & Pisani, 2005).