INTRODUCTION
Around 56 million people die worldwide each year (World Health Organization, 2015). Some of these deaths could be categorized as “bad”—where a painful or otherwise problematic dying process takes place. It is of great concern that bad deaths could be occurring today, as “good” deaths have been a common global objective since the palliative care movement was initiated in the 1960s (Good et al., Reference Good, Gadmer and Ruopp2004). Moreover, good deaths are widely considered possible now as a result of advances in medications, healthcare delivery, and other areas (Hales et al., Reference Hales, Zimmermann and Rodin2010). Although avoiding bad deaths may be as much or more important than trying to ensure that good deaths occur, care efforts and research attention to date have mainly focused on the good death. Many good death studies have been performed, and at least four literature reviews have been conducted to establish what is known about good deaths (Cottrell & Duggleby, Reference Cottrell and Duggleby2016; Evans & Walsh, Reference Evans and Walsh2002; Meier et al., Reference Meier, Gallegos and Thomas2016; Proulx & Jacelon, Reference Proulx and Jacelon2004). These reviews have identified a wide range of factors for good deaths, although it is widely acknowledged that good deaths are highly individualized in relation to the person, family, context, and culture/society (Brazil et al., Reference Brazil, McAiney and Caron-O'Brien2004). One would assume, though, that when good death factors are present those factors would not only enhance the possibility of good deaths occurring but also reduce the occurrence of bad deaths. It is also possible that different factors cause or potentiate bad deaths.
As no published literature reviews on the topic of a bad death could be located, a review of the research literature related to “bad death” was undertaken to determine the state of the science on the bad death. Moreover, our review sought to identify the factors that have been associated with bad deaths, so as to learn what contributes to or causes a bad death. In addition, evidentiary information was sought on the incidence of bad deaths to determine how important it is to design and implement plans to prevent them. Moreover, research evidence on the outcomes or consequences of bad deaths was also sought. Outcomes information is particularly relevant, as bad deaths can have serious consequences for the dying person as well as for their family, their caregivers, their community, and society as a whole.
METHODS
A scoping literature review was performed. Like other literature reviews, scoping reviews collect and organize information to gain an understanding of the existing evidence base (Arksey & O'Malley, Reference Arksey and O'Malley2005; Armstrong et al., Reference Armstrong, Hall and Doyle2011). However, scoping reviews are specifically designed to consolidate research evidence and thus create knowledge on new or undeveloped topics (Whittemore & Knafl, Reference Whittemore and Knafl2005). Scoping reviews differ from systematic literature reviews in two ways: (1) systematic reviews focus on narrow and well-defined questions as compared to scoping reviews which seek to gain an overall understanding of a topic; and (2) systematic reviews aim to provide answers to specific questions through the use of quality-confirmed studies as compared to scoping reviews where the entire research literature base is employed (Arksey & O'Malley, Reference Arksey and O'Malley2005).
Before conducting our literature search, a university librarian was consulted to determine which library databases, medical subject headings (MESH), and keywords would identify published qualitative, quantitative, and mixed-methods research reports that focused in whole or in part on the “bad death.” The goal was to find all English-language research reports published in peer-reviewed journals over the past two decades (1995–2016). The library databases Embase, PubMed, CINAHL, SocINDEX, and Medline were identified as the most relevant. The MESH/keyword terms were “bad death” OR “good death” OR “death quality” combined with “research.” The inclusion criteria were: in the English language, had an abstract, published in a peer-reviewed journal during the years 1995–2016, was a research study with information provided on the methods used to gather and analyze data, and focused in whole or in part on the concept of a “bad death.” The search criteria excluded all non-research reports, although four “good death” literature reviews were identified and used to find research articles for review. The reference lists of all the selected articles were also read to identify articles for review.
Our search of the five library databases revealed nearly 1,000 articles for possible review. Each was assessed by way of its title and a reading of its abstract for relevance. Most were not research articles, and most studies focused solely on good deaths, which reduced the number to 212 possible articles for review. A reading of each full paper was then undertaken for relevancy, with this reducing the number to 20. Another 5 articles were found through searching the reference lists of these 20 articles and the 4 good death literature reviews (Cottrell & Duggleby, Reference Cottrell and Duggleby2016; Evans & Walsh, Reference Evans and Walsh2002; Meier et al., Reference Meier, Gallegos and Thomas2016; Proulx & Jacelon, Reference Proulx and Jacelon2004).
Each of the 25 retained articles was read by two readers independently, both seeking to determine the state of the science on bad deaths and to identify information to answer three questions:
-
1. What causes or contributes to a bad death?
-
2. How often do bad deaths occur?
-
3. What is the impact or outcomes and consequences of bad deaths?
Table 1 contains the identified information from each article divided into five columns: (1) author(s), year published, and where the study was conducted; (2) the purpose of the study and type of study; (3) research methods; (4) findings in relation to the three questions; and (5) other information relevant to this scoping review. The collected and confirmed information was then assessed through team discussions. A consensus was reached on the findings, with a report then drafted and completed by the team.
Literature review findings
RESULTS
Most investigations were qualitative (n = 20), with 2 others being anthropological studies, 2 mixed-methods studies, and 1 a quantitative study. The 25 studies were conducted in 10 developed or developing countries. Nearly half were conducted in the United States (US, n = 12), followed by England (n = 5), and 1 each in Bangladesh, Ghana, South Africa, New Guinea, Korea, Australia, Argentina and Spain combined, and Canada.
Five were published in 2004, 4 in 2002; 3 in 2014; 2 each in 2003, 2005, 2010, and 2015; and 1 each in 1996, 2007, 2008, 2009, and 2013. Most often, the perspectives or views of registered nurses about bad deaths were sought (n = 7), followed by terminally ill or dying people (n = 5), family members (n = 3), older people (n = 4), physicians (n = 3), nursing students (n = 2), the community as a whole (n = 2), clergy (n = 1), younger and middle-aged adults (n = 1), and broad-based social insights through an analysis of print newspaper media (n = 1).
The Incidence of Bad Deaths
No studies were conducted to establish the incidence of bad deaths, although three identified some relevant information. One study of hospice nurses in the US found that they considered 8 of 102 deaths (7.8%) as bad as a result of food and fluid refusal (Ganzini et al., Reference Ganzini, Goy and Miller2003). Another US study of registered nurses working in hospitals and community settings reported that they thought bad deaths were more common in hospitals than in community settings (Borbasi et al., Reference Borbasi, Wotton and Redden2005). A third US study found that 23% of family members believed that their family member had died a bad death in a hospital (Wiegand & Petri, Reference Wiegand and Petri2010). That study focused on eight decedents, with five family members reporting that the death was bad, four reporting that the death had both bad and good elements, and one unsure about the quality of the death.
Factors or Reasons for Bad Deaths
All of the reviewed studies identified one or more factors related to bad deaths, with some factors common across two or more studies. Despite considerable variability, the identified factors could be grouped into six categories: (1) physical pain, (2) suffering, (3) sudden and unexpected deaths, (4) prolonged dying processes or terminal illnesses, (5) disrespect of the dying person, and (6) dying while experiencing a lack of dignity. Among these, the most commonly identified factor was unresolved physical pain (Adesina et al., Reference Adesina, DeBellis and Zannettino2014; De Jong & Clarke, Reference De Jong and Clarke2009; Good et al., Reference Good, Gadmer and Ruopp2004; Kim & Lee, Reference Kim and Lee2003; Ko et al., Reference Ko, Cho and Perez2013; Shea et al., Reference Shea, Grossman and Wallace2010; Vig et al., Reference Vig, Davenport and Pearlman2002; Vig & Pearlman, Reference Vig and Pearlman2004). A bad death was also thought to occur in cases where the dying person in pain could not ask for pain medications, such as those with dementia dying in a nursing home who could not indicate that they were in pain (Seymour et al., Reference Seymour, Bellamy and Gott2002).
Other forms of suffering were also of concern for their potential to cause bad deaths, such as unrelieved physical symptoms other than pain (Good et al., Reference Good, Gadmer and Ruopp2004; Hanson et al., Reference Hanson, Henderson and Menon2002; LeBaron et al., Reference LeBaron, Cooke and Resmini2015; Luxardo et al., Reference Luxardo, Padros and Tripodoro2014; Payne et al., Reference Payne, Langley-Evans and Hillier1996; Pierson et al., Reference Pierson, Curtis and Patrick2002; Shea et al., Reference Shea, Grossman and Wallace2010; van der Geest, Reference van der Geest2004; Vig et al., Reference Vig, Davenport and Pearlman2002; Vig & Pearlman, Reference Vig and Pearlman2004; Wiegand & Petri, Reference Wiegand and Petri2010). Suffering could also result from nonphysical factors, such as when the dying person was lonely (Hanson et al., Reference Hanson, Henderson and Menon2002) or not at peace (van der Geest, Reference van der Geest2004). Dying alone or when isolated from people who care for them was similarly identified as a factor for bad deaths (Kim & Lee, Reference Kim and Lee2003; Ko et al., Reference Ko, Kwak and Nelson-Becker2015; LeBaron et al., Reference LeBaron, Cooke and Resmini2015; Luxardo et al., Reference Luxardo, Padros and Tripodoro2014; Pierson et al., Reference Pierson, Curtis and Patrick2002; Seale, Reference Seale2004; Shea et al., Reference Shea, Grossman and Wallace2010; van der Geest, Reference van der Geest2004; Vig et al., Reference Vig, Davenport and Pearlman2002). Existential pain and spiritual pain also identified as factors that cause bad deaths (Luxardo et al., Reference Luxardo, Padros and Tripodoro2014). In addition, fear during the dying process was associated with bad deaths (Shea et al., Reference Shea, Grossman and Wallace2010).
Another category of commonly identified factors for bad deaths was sudden and unexpected death, which was considered bad by a number of sociocultural and other groups (Borbasi et al., Reference Borbasi, Wotton and Redden2005; Costello, Reference Costello2006; Good et al., Reference Good, Gadmer and Ruopp2004; Hanson et al., Reference Hanson, Henderson and Menon2002; Joarder et al., Reference Joarder, Cooper and Zaman2014; Kim & Lee, Reference Kim and Lee2003; Ko et al., Reference Ko, Cho and Perez2013; Reference Ko, Kwak and Nelson-Becker2015; Garnett et al., Reference Garnett, Vandrevala and Hampson2008; Pierson et al., Reference Pierson, Curtis and Patrick2002). Sudden and unexpected deaths were considered bad because there was little or no opportunity to prepare for the death and ensure that a good death occurs (Adesina et al., Reference Adesina, DeBellis and Zannettino2014; Borbasi et al., Reference Borbasi, Wotton and Redden2005; Costello, Reference Costello2006; Good et al., Reference Good, Gadmer and Ruopp2004; Luxardo et al., Reference Luxardo, Padros and Tripodoro2014; van der Geest, Reference van der Geest2004; Vig et al., Reference Vig, Davenport and Pearlman2002). In some cases, signs of impending death were ignored, resulting in a sudden and unexpected death (Payne et al., Reference Payne, Langley-Evans and Hillier1996). When deaths occurred suddenly, the dying people, their families, and their formal caregivers (including hospital nurses if hospitalized or nursing home staff if residing in a nursing home) were unprepared (Adesina et al., Reference Adesina, DeBellis and Zannettino2014; Borbasi et al., Reference Borbasi, Wotton and Redden2005; Costello, Reference Costello2006). Sudden and unexpected deaths were considered premature or abhorrent, as in the case of homicide, a fatal accident, or the suicide of younger persons. Early deaths were considered neither timely nor appropriate (Adesina et al., Reference Adesina, DeBellis and Zannettino2014; Counts & Counts, Reference Counts and Counts2004; Joarder et al., Reference Joarder, Cooper and Zaman2014; Garnett et al., Reference Garnett, Vandrevala and Hampson2008; Ko et al., Reference Ko, Cho and Perez2013; Reference Ko, Kwak and Nelson-Becker2015; Payne et al., Reference Payne, Langley-Evans and Hillier1996; Posel et al., Reference Posel, Kahn and Walker2007; van der Geest, Reference van der Geest2004; Vig et al., Reference Vig, Davenport and Pearlman2002; Wiegand & Petri, Reference Wiegand and Petri2010).
In addition, prolonged dying processes and long terminal illnesses were also identified as factors causing bad deaths (Garnett et al., Reference Garnett, Vandrevala and Hampson2008; Good et al., Reference Good, Gadmer and Ruopp2004; Hanson et al., Reference Hanson, Henderson and Menon2002; Ko et al., Reference Ko, Cho and Perez2013; Shea et al., Reference Shea, Grossman and Wallace2010; Vig et al., Reference Vig, Davenport and Pearlman2002; Vig & Pearlman, Reference Vig and Pearlman2004). Vig and Pearlman's (Reference Vig and Pearlman2004) study found that dying in one's sleep was preferable to a long drawn-out dying process. Wiegand and Petri (Reference Wiegand and Petri2010) found that prolonged hospital stays with fluctuations in health that fostered and then dashed hope constituted bad deaths. Persistent vegetative states or prolonged comas were also identified with bad deaths (Kim & Lee, Reference Kim and Lee2003). Moreover, long dying processes and dependency on other people contributed to bad deaths, because the family was often overburdened with end-of-life care (Kim & Lee, Reference Kim and Lee2003; Ko et al., Reference Ko, Kwak and Nelson-Becker2015; Luxardo et al., Reference Luxardo, Padros and Tripodoro2014; Vig et al., Reference Vig, Davenport and Pearlman2002). However, bad deaths could also occur when the person's end-of-life wishes were not known, such as when Alzheimer's disease was present and no relevant conversations between the dying person and their family or caregivers could take place (Hanson et al., Reference Hanson, Henderson and Menon2002; Pierson et al., Reference Pierson, Curtis and Patrick2002; Vig et al., Reference Vig, Davenport and Pearlman2002). When no direction from the dying person was available, the healthcare team's values and beliefs were often paramount in end-of-life care (Borbasi et al., Reference Borbasi, Wotton and Redden2005).
Disrespect of the dying person was another commonly identified bad death factor. For instance, bad deaths were thought to occur when the wishes and preferences of the dying person were not followed by the care team or their family (De Jong & Clarke, Reference De Jong and Clarke2009). Similarly, a bad death was deemed to have occurred when older dying people were not listened to (Seymour et al., Reference Seymour, Bellamy and Gott2002). Bad deaths also occurred when the end-of-life or other decisions made by the dying person were not respected (Seymour et al., Reference Seymour, Bellamy and Gott2002). Yet, bad deaths were also considered to occur when the family had no involvement or opportunity for involvement in end-of-life care decisions (Borbasi et al., Reference Borbasi, Wotton and Redden2005; Ko et al., Reference Ko, Cho and Perez2013). In one developing country, bad deaths were thought to be a consequence of the person's bad actions in this or a past life (Joarder et al., Reference Joarder, Cooper and Zaman2014).
Undignified deaths were also bad deaths (Adesina et al., Reference Adesina, DeBellis and Zannettino2014; LeBaron et al., Reference LeBaron, Cooke and Resmini2015; Pierson et al., Reference Pierson, Curtis and Patrick2002; van der Geest, Reference van der Geest2004). For instance, the use of life-support technology so that the dying process was prolonged preventing a natural death from occurring, and life lacking quality was often identified as bad deaths (Garnett et al., Reference Garnett, Vandrevala and Hampson2008; Good et al., Reference Good, Gadmer and Ruopp2004; Ko et al., Reference Ko, Cho and Perez2013; Reference Ko, Kwak and Nelson-Becker2015; Payne et al., Reference Payne, Langley-Evans and Hillier1996; Pierson et al., Reference Pierson, Curtis and Patrick2002). Not dying in the place of one's choice was also considered a bad death (Payne et al., Reference Payne, Langley-Evans and Hillier1996; Pierson et al., Reference Pierson, Curtis and Patrick2002), as well as not dying at home (Ko et al., Reference Ko, Cho and Perez2013; van der Geest, Reference van der Geest2004). Negative changes in the dying person's physical appearance (e.g., disfigurement) were also identified as a factor contributing to an undignified death (Hanson et al., Reference Hanson, Henderson and Menon2002; Vig et al., Reference Vig, Davenport and Pearlman2002). The physical incapacity of the dying person (such as being bedridden and dependent on others) was similarly considered a factor leading to an undignified and therefore bad death (De Jong & Clarke, Reference De Jong and Clarke2009; Hanson et al., Reference Hanson, Henderson and Menon2002). Bad deaths also occurred when the dying person had no control over their own life (Garnett et al., Reference Garnett, Vandrevala and Hampson2008; Pierson et al., Reference Pierson, Curtis and Patrick2002). A negative emotional or mental state of the dying person constituted additional factors leading to an undignified or bad death (e.g., if the dying person was angry or frustrated) (LeBaron et al., Reference LeBaron, Cooke and Resmini2015; Pierson et al., Reference Pierson, Curtis and Patrick2002; Vig et al., Reference Vig, Davenport and Pearlman2002).
Outcomes or Consequences of Bad Deaths
No study focused on determining the outcomes or consequences of bad deaths, but half reported on at least one potential impact of a bad death. These impacts differed considerably because the dying individual, their family, the involved healthcare providers or healthcare team, and the community or society as a whole could be variably impacted. These impacts were often identified as immediate, but they could also occur over the long term.
Bad deaths were of chief concern for their impact on the dying people, who would die a less than optimal or perhaps even horrific death (Borbasi et al., Reference Borbasi, Wotton and Redden2005; Costello, Reference Costello2006). Bad deaths were also of concern as they could impact caregivers, including nurses, who get distressed over not being able to reduce or eliminate symptoms (Borbasi et al., Reference Borbasi, Wotton and Redden2005; Costello, Reference Costello2006; De Jong & Clarke, Reference De Jong and Clarke2009). Nurses were said to feel guilty when unable to prevent a bad death or ensure a good death (Costello, Reference Costello2006). Caring for dying people who were experiencing bad deaths was said to cause emotional overload and frustration among formal caregivers (Luxardo et al., Reference Luxardo, Padros and Tripodoro2014). Reduced morale among caregivers was also reported among hospital nurses both during and after a bad death (Borbasi et al., Reference Borbasi, Wotton and Redden2005; Costello, Reference Costello2006). Caregiver distress and stress among all caregivers occurred when important conversations did not take place, such as those needed to plan good end-of-life care or to promote “closure” (Costello, Reference Costello2006; Ko et al., Reference Ko, Cho and Perez2013).
Bad deaths due to sudden and unexpected deaths also meant that the wishes of the dying person were not known or not achieved, as there was not enough time to enact final wishes or preferences (Borbasi et al., Reference Borbasi, Wotton and Redden2005; Costello, Reference Costello2006). Distress and stress among healthcare providers, as well as family members, were related to an inability to start comfort-oriented care—such as when the death came too suddenly or when palliative care was not started early enough for a chronically or terminally ill person, so that their pain and other symptoms could not be prevented or successfully addressed (Borbasi et al., Reference Borbasi, Wotton and Redden2005; Costello, Reference Costello2006; Luxardo et al., Reference Luxardo, Padros and Tripodoro2014). Sudden deaths also prevented important end-of-life religious and cultural rituals (Costello, Reference Costello2006). Low-quality end-of-life care has thus been identified as a consequence of bad deaths.
Conflict between and among those involved in the dying process was another commonly identified outcome of bad deaths (Counts & Counts, Reference Counts and Counts2004; De Jong & Clarke, Reference De Jong and Clarke2009; Garnett et al., Reference Garnett, Vandrevala and Hampson2008; Good et al., Reference Good, Gadmer and Ruopp2004; Luxardo et al., Reference Luxardo, Padros and Tripodoro2014). This conflict could be between nurses and physicians when the nurses believed that a physician had ordered inappropriate in-hospital care (Borbasi et al., Reference Borbasi, Wotton and Redden2005; Costello, Reference Costello2006; De Jong & Clarke, Reference De Jong and Clarke2009). Conflict among family members was also cited (Counts & Counts, Reference Counts and Counts2004; Good et al., Reference Good, Gadmer and Ruopp2004; Luxardo et al., Reference Luxardo, Padros and Tripodoro2014), as well as conflict between family members and the physician or nurses who were providing care for their dying family member (Costello, Reference Costello2006; Garnett et al., Reference Garnett, Vandrevala and Hampson2008; Good et al., Reference Good, Gadmer and Ruopp2004; Luxardo et al., Reference Luxardo, Padros and Tripodoro2014).
Community-wide problems and social disruption or societal upheaval were additional possible outcomes of bad deaths (Counts & Counts, Reference Counts and Counts2004; Posel et al., Reference Posel, Kahn and Walker2007). For instance, AIDS deaths in South Africa were considered bad deaths, as young people were dying, and these deaths led community members to think that their country's social values and customs were changing for the worse (Posel et al., Reference Posel, Kahn and Walker2007). Shame and blame were also associated with bad deaths, with this resulting in distrust between and among community members (Counts & Counts, Reference Counts and Counts2004; Posel et al., Reference Posel, Kahn and Walker2007). In one case, the murder of a presumed offender believed to have caused a bad death was an outcome of community strife over that bad death (Counts & Counts, Reference Counts and Counts2004). Bad deaths also have led to charges of professional incompetence when healthcare providers were unable to prevent deaths or adequately explain why a sudden or premature death had occurred (Counts & Counts, Reference Counts and Counts2004; Posel et al., Reference Posel, Kahn and Walker2007). The suspicion that family members had been incompetent or uncaring was another outcome of bad deaths, as well as suspicion of foul play (Counts & Counts, Reference Counts and Counts2004).
DISCUSSION
Our attempt to gather research evidence on bad deaths revealed only 25 research articles published in the last two decades for inclusion in our review. Although this is not a large number, and most were qualitative studies with findings that cannot be generalized, these studies were helpful in demonstrating that bad deaths are a valid concern. These studies were also helpful in identifying probable causes or contributing factors and possible outcomes and consequences of bad deaths.
Although these 25 studies were conducted in 10 different countries, which indicates widespread recognition of the possibility of bad deaths, no escalation in research has occurred since publication of the first such article in 1996. With only two research reports published in 2015 and none in 2016, current research attention is not apparent. As such, an undeveloped state of the science exists, a state that would be appropriate if the palliative care movement and other developments had rendered all death and dying processes good. However, two of the reviewed studies revealed that 7.8 to 23% of deaths were considered bad (Ganzini et al., Reference Ganzini, Goy and Miller2003; Wiegand & Petri, Reference Wiegand and Petri2010). Moreover, bad deaths were thought more likely to occur in hospitals than in community settings (Borbasi et al., Reference Borbasi, Wotton and Redden2005). It is also important to recognize that the 25 studies included in our review were conducted out of a concern that bad deaths are possible. Consequently, more research is needed on bad deaths, including quantitative studies to establish their incidence. Until this incidence is known, bad deaths may be considered rare or unusual, and thus of little or no consequence.
Only 8 of our 25 studies focused on the perspectives of dying persons and/or their family members, with the formal caregiver perspective more often obtained. Although formal care providers are an understandable focus of research attention, since burdening dying people and their families with research participation has ethical and practical concerns, research needs to determine what makes a death bad for the dying person (Ternestedt et al., Reference Ternestedt, Andershed and Eriksson2002). Another essential focus is the family, as they are often involved in end-of-life care and are the ones who are left to grieve.
Research is also needed to determine the extent of individual and sociocultural differences in relation to factors causing or potentiating bad deaths. Vig et al. (Reference Vig, Davenport and Pearlman2002) found that common bad death factors existed according to 16 older Americans attending a geriatric clinic for nonterminal heart disease or cancer, although some differences among them also existed. Similarly, Vig and Pearlman (Reference van der Geest2004) found common factors but heterogeneous views about good and bad deaths among 26 men diagnosed with terminal heart disease or cancer. Payne et al.'s (Reference Payne, Langley-Evans and Hillier1996) study found that major differences in bad death views existed between patients and staff. Kim and Lee (Reference Kim and Lee2003) found that bad death views varied among Korean nurses. Shih's (Reference Shih2010) discussion of bad death in Taiwan illustrated the importance of determining how culture and religion influence individual, family, and social views about the “bad death.”
Regardless, unrelieved pain was revealed as a prime factor in bad deaths. Correspondingly, all four good death literature reviews identified an absence of pain as being essential for a good death (Cottrell & Duggleby, Reference Cottrell and Duggleby2016; Evans & Walsh, Reference Evans and Walsh2002; Meier et al., Reference Meier, Gallegos and Thomas2016; Proulx & Jacelon, Reference Proulx and Jacelon2004). Other bad death factors appear to mirror those needed for good deaths (Adesina et al., Reference Adesina, DeBellis and Zannettino2014; van der Geest, Reference van der Geest2004), although our literature review revealed some salient bad death attributes, ones that could be avoided to prevent bad deaths from occurring.
Our review was also helpful in revealing that bad deaths have consequences and so should be avoided. These consequences are concerning, as the dying person is often impacted by a bad death and dying process. For instance, McPherson and colleagues (Reference McPherson, Wilson and Murray2007) found that dying persons were distressed when believing that they are a burden to their family members. In fact, dying persons who are suffering may ask for assisted suicide to end such burdens (Quill et al., Reference Quill, Lo and Brock.1997). Family members, caregivers, communities, and even society as a whole can also be impacted by bad deaths. As only one long-term study was found for our review (Counts & Counts, Reference Counts and Counts2004), it is advisable for longitudinal or multi-year studies to be performed in order to assess the ongoing impact of bad deaths. These studies are important, as Counts and Counts' (Reference Counts and Counts2004) study in a developing country showed that a bad death can have serious consequences for many years after a death, as it resulted in continuing conflicts, fissures in society, and possibly more deaths. Their study also revealed that a murder took place a few years after a young woman committed suicide, the crime being a result of the social turmoil in the community arising from the bad death (Counts & Counts, Reference Counts and Counts2004).
Although murder may seem a highly unusual or rare outcome of a bad death, a recent criminal conviction in the province of Alberta in Canada demonstrates otherwise. In 2016, Steven Vollrath was found guilty of kidnapping and torturing an Edmonton man who had driven a vehicle into a restaurant in 2013 and killed a 2-year-old boy. He was convicted on four charges—kidnapping, aggravated assault, possession of a dangerous weapon, and impersonating a police officer—and sentenced to 12 years in prison (Johnston, Reference Johnston2016). These criminal actions were thought to be committed out of revenge for the child's death and the anguish it caused his family (Johnston, Reference Johnston2016).
Clearly, research needs to focus on the immediate and long-term outcomes of bad deaths. These outcomes should focus on the family, as research is revealing bereavement grief to be more severe and longer in duration when the mourners consider the death and dying process to have been a bad one (King, Reference King2004). Mourners can experience extremely severe and prolonged or even permanent grief (Holland et al., Reference Holland, Currier and Gallagher-Thompson2009; He et al., Reference He, Tang and Yu2014; Middleton et al., Reference Middleton, Burnett and Raphael1996; Zhang et al., Reference Zhang, El-Jawahri and Prigerson2006). This grief can lead to physical and/or psychological illnesses and, indeed, early death (Shah et al., Reference Shah, Carey and Harris2013).
Other outcomes are also important to assess, as our review showed that nurses are impacted by bad deaths. This is not the first time that nurses have been found to be impacted by their difficult work—moral distress among nurses is an established concern (Austin, Reference Austin2012). Moral distress arising from bad deaths can lead to nurse depression and job loss, long-term toxic conflict among care team members, and other outcomes that impact the workplace (Austin, Reference Austin2016).
The wide range of factors identified as causing or potentiating bad deaths and the possible impact of bad deaths suggests that routine assessment of terminally ill people and their end-of-life contexts are needed to prevent bad deaths and promote the possibility of good deaths. These assessments must not be burdensome. Our six categories of contributing factors could form the template for this assessment, with one question each asked about concerns or needs in relation to: (1) physical pain, (2) suffering, (3) sudden and unexpected deaths, (4) prolonged dying processes or terminal illnesses, (5) disrespect of the dying person, and (6) dying while experiencing a lack of dignity. This assessment could reveal what must be avoided and what must be done to prevent bad deaths and enable good ones.
IMPLICATIONS AND RECOMMENDATIONS
This scoping review of research literature revealed only 25 studies published in the previous two decades that focused in whole or in part on the “bad death.” Although much more research is needed, some information has been gained in relation to the incidence of bad deaths, what may cause them, and what may be their possible consequences. Above all, our review indicates that bad deaths can occur whenever a good death has not been achieved, as well as when bad death factors are present. These bad death factors must be understood and addressed further.
ACKNOWLEDGMENTS
No funding was received for this project. We appreciate the researchers who understood the need for studies on the “bad” death, and all the people who are doing so much to prevent bad deaths and potentiate good deaths.
DISCLOSURES
The authors hereby declare that they have no conflicts of interest to disclose.
