Abstract
In conceptualizing vulnerability, it is common for researchers to assume that some participants are more vulnerable on the basis of their membership of a particular group or because they exhibit particular characteristics. Older people are often viewed as inherently more vulnerable by ethics committees and the ethical guidelines committees construct. Because age alone does not confer or cause vulnerability, risk of harm to older research participants is not purely associated with their intrinsic connection to a vulnerable group, and classifying older research participants as vulnerable may not necessarily protect them from harm. Drawing on the preliminary findings of a qualitative study of older people who had survived a suicide attempt, we reflect on how the specific context of our study had the potential of framing older people as vulnerable, and describe ways in which these were managed and resolved. Specifically, we discuss potential for harm through the ethical principles of coercion and distress.
Introduction
Certain categories of people are presumed to be more likely than others to be misled, mistreated or otherwise taken advantage of as participants in research. This sets up the expectation for researchers, review committees and regulators to provide special protections for them (Ketefian, 2015; Levine et al., 2004). Extant literature locates several groups of individuals as generally more vulnerable. Such groups often include foetuses, children, pregnant women, students, employees and older adults (Moore and Miller, 1999). A fundamental assumption underpinning the concept of vulnerability is the likelihood of physical harm. The term originates from the Latin word
So far there has been little concrete engagement with the question of the particular vulnerability of older persons (as opposed to those with experience of suicidality), and relevant issues relating to ageing itself do not receive dedicated attention in the research literature (Holm, 2013). In response, the following section will engage with conceptual and ethical concerns regarding the relationship between ageing and vulnerability – specifically, engagement with those concerns from the bioethical literature. We begin by addressing the notion and ethical significance of vulnerability in ageing before exploring this in the context of suicide research.
Background
Bioethics has been defined as a ‘disciplined examination of value dilemmas in health care’ (Moody, 1992: 395). Of the most significant histrionic debates are the ethical dilemmas associated with birth, reproduction, and death and dying; by comparison, ageing and old age are under-theorized in bioethics (Holm, 2013) and, where work has been done, bioethicists have tended towards a lifespan approach (Agich, 2001). Such an approach offers a deficit lens, focusing attention towards debility and loss associated with changes in patterns of ageing (Agich, 2001: 25). The idea of the natural lifespan within the accepted normative framework of bioethics values instrumental reason, emphasizing the (re)productive purpose of human life, and because older people have achieved their functions of parenting and are no longer viewed as economically productive, they become a natural group to nominate for so-called ‘natural’ death. Such biologized readings of death give rise to the cultural belief that other deaths are not acceptable or natural (Agich, 2001). In the research context, then, how might this impact the way we come to see vulnerability in older people who have attempted suicide?
In conducting suicide research, ‘life and death are potentially at stake’ (Mishara and Weisstub, 2008: 353). Suicide is related to vulnerability in the sense that many persons associated with the act fall into the classification of vulnerable by international councils and bodies governing ethical guidelines (Mishara and Weisstub, 2008). Suicidal populations also overlap with other vulnerable groups, such as the mentally disordered and older people, which presents specific challenges (Mishara and Weisstub, 2008) and some particular groups of older adults are more vulnerable to suicide than others. Worldwide, men aged 70 and older are at significantly greater risk of dying by suicide than older females (World Health Organization, 2014). For example, in Australia, men aged 85 and over are three times more likely to die by suicide than are those in other age groups (Australian Bureau of Statistics, 2012). Furthermore, between July 2000 and December 2011, the deaths by suicide of 2816 men aged 65 years and over were reported to an Australian Coroner, compared with 830 women aged 65 years and over (National Coronial Information System, 2014).
Moore and Miller define doubly vulnerable individuals as persons who ‘simultaneously experience more than one factor that diminishes autonomy’ (Moore and Miller, 1999: 1035). Given that a previous suicide attempt across the lifespan is the strongest single clinical indicator for a future attempt (Lönnqvist, 2006), does this then infer a double vulnerability where older people are concerned? There is currently little evidence to support this. Crocker et al. (2006) conducted a qualitative study with 15 people aged 65 and over to explore subjective experiences pre- and post-suicide attempt. In terms of participant vulnerability, they argued that there was no ‘correct’ time to capture their experiences. For example, some participants only agreed to participate in the research once they felt less vulnerable and had ‘moved on’ from their attempt, whilst others talked about their experience when it was still current (Crocker et al., 2006: 646). More recently, Gutierrez et al. (2015) also discussed the subjective experiences of older people with suicidal ideation and/or suicide attempts. In describing the vulnerability of their participants, they refer to ‘capacity’, highlighting a demarcation between real-world and research contexts:
Although [the] study is of old people who are fragile and weakened in their capacity to act in the world, one finds in them a capacity for criticism and reflection about life that should be given value, supported and enhanced, for the purpose of a process of personal reconstruction and recomposition of meaning (Gutierrez et al., 2015: 1739).
Although vulnerability is relevant to conceptualizing, designing and conducting research (Lange et al., 2013), it is a concept that has focused on populations rather than situations, particularly the changeable nature of situations (Aultman, 2014). For example, when asked to describe their own vulnerability, older people in the study by Abley et al. (2011) did not see themselves as vulnerable in general. Rather, they spoke of feelings associated with being in certain ‘situations of vulnerability’. In this way, vulnerability involved ‘an emotional response which was dependent on being in a particular situation’ (Abley et al., 2011: 361). In research situations, individual needs for enhanced protection not only depend on their group, but rather on the specific features of the research project and the setting within which it is carried out (Levine et al., 2004). The resolution of ethical issues in suicide research is not simply based upon a direct application of existing guidelines (Mishara and Weisstub, 2008: 353). As Jenkins highlights:
it is for the researcher to reflect on in what way the specific context of their proposed study might create vulnerability, rather than for the research team to assume that a particular group is vulnerable per se on the basis of their membership of a particular group or because they have particular characteristics (Jenkins, 2012: 28).
What now follows is a description of a multiple-case study research design which aimed to explore the nature and scope of protective factors that provide older people reasons and experiences to live following a suicide attempt. Within this context, we reflect on how we conceptualized the potential vulnerabilities of our participants. The discussion that follows centres on two key ethical principles – coercion and distress – which were of particular concern to the ethics committees who reviewed our proposal. Specifically, we demonstrate how the potential for harm was managed and resolved during the recruitment and interview stages of the study.
Method
Ethics
Prior to commencing the study, ethical approval was gained from the Human Research Ethics Committees of the University of South Australia and the South Australian Department of Health.
Setting
We conducted a community-based study in metropolitan Adelaide. Participants were recruited through one non-Government and one Government organization: (i) Minimization of Suicide Harm (MOSH), a not-for-profit organization which aims to cater for the needs of those at risk of self-harm and suicide and those bereaved by suicide; and (ii) Older Persons Mental Health Service (OPMHS), a free state Government service for people aged 65 years and over who have developed mental illness later in life or have longstanding mental health problems complicated by age-related issues.
Participants
Participants were men and women between 60 and 82 years of age who had survived a suicide attempt, but not within 12 months of the study. Burless and De Leo (2001) argue that the intensification of any suicidal manifestation is highest in the year after it first occurs. Participants possessed an ability to speak and understand English and articulate their experience. We excluded persons with psychosis and dementia and persons who may have been put at increased risk of suicide by participating in an interview. Two males and five females participated in the study. Three of the participants had made one suicide attempt, three participants had made multiple suicide attempts, and two participants had a history of self-harming behaviour, with one serious attempt.
Selection of cases and recruitment
A combination of purposive and snowball sampling techniques were employed to identify participants. We used the purposeful technique to identify information-rich cases for study in-depth (Patton, 2002), whilst the snowball strategy aimed to identify the cases through referrals made among people who possessed characteristics relevant to the research (Biernacki and Waldorf, 1981). According to Stake (2006), the benefits of multiple-case study are optimal with no fewer than four and no more than 15 cases. In the present study, a maximum of ten cases was targeted to enhance external validity. Owing to the replication inherent in the design, multiple-case study evidence is typically viewed as more compelling, enhancing confidence in the generalizability of the findings (Yin, 2014). Recruitment was carried out in two ways. The first stage of recruitment was facilitated by the Chair of MOSH, who sent a preliminary email to known senior stakeholders from other suicide prevention and support services and community organizations to introduce them to the study and invite them to identify potential participants. In the second stage, the researcher delivered a presentation about the study to four regional groups of psychiatrists, case workers and allied health staff from OPMHS. In both stages, interested persons who met inclusion criteria were invited to contact the researcher to discuss the project further with them and/or organize an interview time.
Screening
Screening was carried out in two stages. During recruitment, only those participants identified by case workers and psychiatrists as not being put at increased risk of suicide were referred to the study. Secondly, because there was a time delay between recruitment and interview, all participants were screened on the day of their scheduled interview. A highly qualified and experienced mental health nurse (MHN) conducted a Mental State Examination and administered self-report questions to assess the person’s suitability for an interview at that particular time.
Data collection
Semi-structured individual interviews were conducted with participants. An interview questioning framework was used to explore (i) what led up to the person’s suicide attempt (what was happening in their life at the time of the attempt), (ii) what things gave them a sense of meaning and purpose immediately post-attempt and during their recovery, and (iii) what keeps them going now. Semi-structured interview transcripts were also supplemented with the participants’ discharge summaries. The aim was to enhance data quality based on the principles of idea convergence and confirmation of findings.
Discussion
The project was described by the ethics review committees as having a ‘sensitive indication’. In their response to the original application, two main concerns were voiced. The first concern pertained to the consumer participation pathway, specifically how participants would be identified and contacted and by whom. It was suggested that the researcher needed to address the potential for coercion as a consequence of dependent relationships. Secondly the committees asked that the researcher demonstrate a greater awareness of the risks to participants, as well as identify strategies to manage and/or minimize them. Despite adequately addressing these concerns, none of our study participants decided to take up the extra support provided pre- or post-interview. None indicated that they felt distressed as a result of discussing their experiences – in fact all participants indicated that they found it therapeutic to share their feelings and experiences.
One of the fundamental ethico-political challenges in research is the potential for coercion and exploitation of vulnerable groups (Russell, 1999). Participants should be able to ‘understand the nature of the research process, comprehend their role and not be under duress when making decisions’ (Tee and Lathlean, 2004: 538). Questions such as how the approach would be made, who made the approach and how to avoid any sense of coercion were critical considerations. A tension existed between demonstrating a safe and supportive approach and adopting overly paternalistic measures (Tee and Lathlean, 2004). According to Fineman, the risk is that:
vulnerability may be seen as inherent to the status of being an older person, rather than something that has roots in the life-courses and environments of some older persons and can be used to justify heavy-handed and paternalistic intervention in the lives of older adults (Fineman, 2012: 93).
Gaining access to participants in this study presented a challenge to the researcher. It was likely that some participants may have experienced – and were continuing to experience – mental illness. Although participants with mental illness were not seen as any less entitled to or capable of participating in the consent process, traditional methods of recruiting participants such as through newspaper or radio announcements and flyer postings in public domains were deemed inappropriate and not likely to reach those who were most representative of the target population (Flaskerud and Winslow, 1998). Although research involving these people need not be limited to their particular impairment, disability or illness, their distinctive vulnerabilities as research participants should be taken into account. The capacity of a person with any of these conditions to consent to research, and to have the ability to participate in it, can vary for many reasons, including the nature of the condition, the person’s medication or treatment, the person’s discomfort or distress, the complexity of the research project, and fluctuations in the condition. For example, mental illness can often be experienced as temporary or episodic. Even when capable of giving consent and participating, people with these conditions may be more than usually vulnerable to various forms of discomfort and stress (National Health and Medical Research Council, 2007: 53). The decision was reached to carry out recruitment through persons who were both trusted by the potential participants (Gueldner and Hanner, 1989) and supportive of the research. In this way, accessibility was potentially enhanced (Moore and Miller, 1999) and participants were deemed more likely to meet the inclusion criteria. Trusted persons in this study were key members of a range of community support groups and public service mental health clinicians and case workers. Whilst arguably there may be a potential conflict of interest in employing existing relationships to recruit and support participants, we emphasized to the ethics committees that potential risks to voluntary consent were outweighed by participants’ rights to be involved in the study and the provision of adequate stand-by support (Lees, 2013). Fostering these partnerships with staff that participants knew and trusted was integral in helping them understand the relevance of the study and reduce suspicion towards the researcher (Provencher et al., 2014).
Being interviewed, often at length, in depth and on a sensitive topic, can be experienced as intrusive or distressing (Gibson et al., 2012). Furthermore, some people may become distressed in response to certain circumstances and not in others because of the timing of a research study. In the field of suicide research, this might occur while participants give accounts of past suicidal thoughts or actions. Equally important is that distress may not necessarily be experienced as harmful. In the field of qualitative suicide research, ethics review committees have been described as conservative and paternalistic, with the tendency to overestimate risks associated with participation (Biddle et al., 2013; Corbin and Morse, 2003; Lakeman and Fitzgerald, 2009). According to Morse and Field, review committees:
sometimes express concern that the interview process may be stressful to some participants. This is not generally the case, even though the participants may become upset during the interview. They generally express appreciation that someone has at last listened [original emphasis] to their stories. (Morse and Field, 1995: 93)
Indeed, there is now substantial evidence that emotional disclosure across a range of settings has the potential to be beneficial; even validating and therapeutic (Biddle et al., 2013; Lakeman et al., 2013). For example, on reflection after their interview with the researcher, one of the participants said:
I’ve enjoyed this … with you.
Thank you very much. It’s a privilege.
I think unless people like me … get out … and talk about what they’ve done … and how glad they are that they’re still here …
It’s critical.
It is. And I am.
According to Browne (2003: 64), ‘suicide attempters have unique needs; they need to talk, to be heard, and to look back with hindsight to see the “why” of the attempt’. Having accepted an invitation to be interviewed, another participant said:
I think it will probably be helpful to talk and give the experience a bit of an airing again four years down the track.
Empirical evidence also suggests that asking about thoughts of suicide is not associated with increases in suicidal ideation (Mathias et al., 2012). Thus, inquiring about a participant’s level of suicidal thoughts or behaviour was built into the screening interview. Whilst the majority of participants provided minimal detail of their suicide attempt – tending to focus more on recovery and the context in which the attempt occurred – one of the participants expressed ambivalence about the possibility of a future attempt:
You feel so down. I just wanted to go to sleep. And most of the time I think I might have done that just to go to sleep. And then other times … but … I hadn’t done it for a while … so … whether … I won’t say it’s the last time. Put it that way. There could be more. Eventually.
Our point is that some distress is to be reasonably expected in discussing suicide with research participants. However, as Thompson and Chambers emphasize, ‘we need to be mindful of the potential for distress and be practiced at managing it – but not necessarily to avoid it’ (Thompson and Chambers, 2012: 30). Thus we aimed to be transparent in the study materials that participants received. In line with recommendations by Henderson and Jorm (1990), our detailed information sheet posed and responded to the question, ‘how will I be supported if I become distressed during the interview?’ Should this have occurred, the researcher would have facilitated an environment where the participant could comfortably choose whether or not to proceed (Thompson and Chambers, 2012).
Concern was also raised by the supervision team about the appropriateness and safety of the researcher conducting interviews with participants alone in their homes. Because of the possibility of eliciting emotional distress during an interview, the research team decided that MOSH House would provide a safer and more supportive environment for both researcher and participant. The premises house a charity shop, communal areas and private consulting rooms. Trained volunteers, known by MOSH as ‘active listeners’, were also on hand post-interview. Generally, it is good practice to offer sufficient post-interview space for reflection on the process to occur (Fossey et al., 2002). Participants were provided with a space at the end of the interview to debrief with the MHN about any emotional distress experienced during the interview and discuss where they could access additional external support if required.
Conclusion
Describing a group as vulnerable suggests that further protections – above and beyond those required for the general population – are warranted. As we have argued, having a starting position of seeing a group as inherently vulnerable does not automatically protect participants from harm through coercion or distress, nor does it encourage opportunities for older people to participate in and influence research agendas. Importantly, ethically sound research involving people who are older and have survived a suicide attempt requires thoroughly developed procedures pertaining to informed consent, assessment of risk, and access to competent pre- and post-interview de-briefing and support.
Recruitment pathways and pre- and post-interview support structures were put in place to address the potential for coercion and distress from the participants, researchers, MHN and support organizations’ perspectives. However, participation in suicide research does not differentially affect older people in comparison to people of other ages, and older people who have survived a suicide attempt are no more or less vulnerable than younger people who have survived an attempt. In this way our initial conceptions of vulnerability and vulnerability management of the participants in our study were less significant than the emerging issues actually borne out by the data collected during the research process. It is imperative that researchers, human research ethics committees and those developing ethical policy guidelines reflect on how the specific context of a research study might create vulnerability for research participants of all ages to effectively critique and address the potential for harm. As Fineman states:
The concept of vulnerability reflects the fact that we are all born, live, and die within a fragile materiality that renders all of us constantly susceptible to destructive external forces and internal disintegration (Fineman, 2012: 71).
Footnotes
Declaration of conflicting interests
The authors declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship and/or publication of this article.