|Home | About | Journals | Submit | Contact Us | Français|
While bereavement is considered by many to be among one of the most stressful life events, it becomes even more distressing when it is related to the suicide of a loved one. A synopsis of psychosocial outcomes of suicide survivor bereavement is presented along with an overview of group interventions designed to help these survivors cope with their grief. The structure of an ongoing eight-week bereavement support group is described to lay a foundation for the application of narrative theory within the group process. Using narrative theory and structural analysis, the discourse of group members is presented and various themes are discussed in an effort to contribute to the task of developing effective psychotherapeutic group interventions for survivors of suicide.
In the United States, suicide is the ninth leading cause of death, accounting for about 30,000 deaths annually (Anderson, Kochanck, & Murphy, 1997). For each person who commits suicide, there may be as many as six survivors (persons close to the suicide victim) left to cope with the loss (American Association of Suicidology, 1997), increasing the number of newly bereaved survivors by as much as 180,000 every year. From a public health perspective, this presents an incredible cost to the quality of life, human productivity, and health service expenditures in terms of the personal and clinical impact of suicide.
The sudden, often violent, unnatural death of an individual is frequently conceptualized as a stressful life event with negative mental health outcomes among bereaved survivors. Studies suggest that the psychological effects of sudden, unnatural deaths by suicide may persist for years (Colt, 1991; Lehman, Wortman, & Williams, 1987; Lukas & Seiden, 1987; Parkes & Weiss, 1983) and include such emotional reactions as shock, anxiety, depression, anger, feelings of hopelessness, shame, self-blame, and guilt related to real or perceived culpability (Brown, Reynolds, Monk, & Prigerson, 1996; Cain, 1972; Jacobs, 1990; McIntosh & Wrobleski, 1988; Shneidman, 1981; Smith, Mitchell, Bruno, & Constantino, 1995; Van Dongen, 1990). This information strongly suggests a need for the development and evaluation of supportive interventions for individuals experiencing suicide-related bereavement (referred to as bereaved survivors or bereaved suicide survivors). However, very little investigative effort has been devoted to understanding or assisting bereaved survivors, either individually or in groups (Dunne, McIntosh, & Dunne-Maxim, 1987).
In addition to the emotional reactions associated with the unexpected tragedy, bereaved survivors of suicide are often confronted with negative social outcomes. For example, Wallace (1977) conducted interviews with women whose husbands committed suicide and found that bereaved spousal survivors suffered from a lack of social support and an inability to talk about their grief because the loss was associated with social stigma. Wallace’s findings were supported more than 20 years later when Farberow and associates investigated the social consequences of suicide and found that bereaved suicide survivors receive significantly less emotional support for their feelings of depression and grief than individuals bereaved by natural causes (Farberow, Gallegher-Thompson, Gilewski, & Thompson, 1992). These authors also found that suicide survivors do not confide in persons within their social networks any more than nonbereaved individuals (Farberow et al., 1992). Further, when Dunne and associates surveyed bereaved suicide survivors, they found marked distortions in their perceptions of time, emotional estrangements from other family members, a pessimistic outlook on life (including the idea of an early death), and suicide ideation (Dunne et al., 1987).
Anecdotal evidence suggests that bereavement following suicide is very different from bereavement following other types of death. However, scientific investigations comparing suicide survivor bereavement with other forms of bereavement show mixed results (for reviews, see Farberow, 1991; McIntosh, 1993; Ness & Pfeffer, 1990; Rudestam, 1992; Seguin, Lesage, & Kiely, 1995; Van der Wal, 1989). It is beyond the scope of this paper to address grief reactions to various types of death. Thus, the reader is referred to an essay by Ellenbogen and Gratton (2001) in which the authors offer plausible explanations for the contradictory findings and suggest areas that warrant further investigation.
While investigators refine methods to explore the differences between the various types of bereavement, it is important for psychiatric nurses and other mental health professionals to have a basic understanding of suicide survivor bereavement because of both its prevalence and potential negative health outcomes. This statement is supported by a study by Barrett and Scott (1990), where data suggested that approximately 65% of adult suicide survivors seek mental health services, even though mental health professionals have received little education on how to intervene effectively (Dunne, 1992).
Given the above, it is important for psychiatric nurses to develop supportive interventions to assist individuals having difficulty coping with suicide-related bereavement. As a starting point, it is important for psychotherapeutic group facilitators to be informed of the various themes that emerge as bereaved suicide survivors grieve (Jordan, 2001). Therefore, the purpose of this paper is to contribute to the ongoing task of providing appropriate support for survivors of suicide.
Awareness of the negative psychological and social effects of bereavement underscores the necessity for designing and evaluating interventions to facilitate healthy grieving among bereaved suicide survivors. To date, several investigators have examined the effects of group psychotherapeutic interventions among suicide survivors. The following is a brief overview of these studies.
More than two decades ago, Hatton and Valente (1981) described a heterogeneous group of six bereaved parental suicide survivors (two husband/wife dyads and two mothers). The group met a total of ten times over a 14-week period. When the group process began, participants often reported feeling isolated from friends and family. At conclusion of the group intervention, participants reported a significant decline in symptoms of depression and improved levels of concentration. Further, the authors noted a reduction in emotional lability among group members (Hatton & Valente, 1981).
During the same time period, Constantino (1981) conducted a pilot study to evaluate three types of interventions (a bereavement crisis group intervention, a social group intervention, and a control group) with 30 bereaved spouses (some of whom were suicide survivors). Results showed that self-reported levels of depressive symptoms increased in the control group and decreased in the two intervention groups. Several years later, Constantino (1988) used a larger sample (N = 17) to compare the effects of the three group interventions described above. Results suggest that subjects in the bereavement crisis intervention group reported significant improvements in depressive symptoms and increased social adjustment, when compared to the social adjustment intervention group. Subjects in the no treatment (control) group maintained a constant level of depressive symptoms with a decrease in perceived social adjustment (Constantino, 1988). Suicide survivors in the study verbalized that they felt “different” from other widows and the group did not meet their particular needs.
A study by Renaud (1995) suggests that participation in bereaved suicide survivor support groups impacts depressive symptoms and feelings of isolation. More recently, Callahan (2000) studied 210 survivors of suicide attending bereavement support groups and found that the loss of a family member by suicide is traumatic, and that seeing the body at the scene was the most significant predictor of high levels of distress. Emotional support from family and friends was the strongest protective factor (Callahan, 2000).
Results of the above studies suggest that psychotherapeutic group interventions for suicide-related bereavement have a positive effect on the participant’s mental health, although further investigative efforts must be invested to delineate which components of the group intervention actually lead to positive outcomes. For example, Trunnell and associates assert that psychotherapeutic group interventions for bereaved individuals should be designed to promote self-expression, constructive copying, and feelings of competency and personal control (Trunnell, Caserta, & White, 1992). Toward that end, Dunne (1992) offers guidelines for a psycho-educational postvention treatment of suicide survivors and Callahan (2000) concludes that a model that combines posttraumatic reactions and grief may be most appropriate in coping with the aftermath of suicide.
To contribute to the development of effective therapeutic group activities for suicide survivors, the present project was designed to analyze one of the central activities of a survivors of suicide support group; that is, the sharing of narratives of loss by the group participants. An approach that focuses on the description and interpretation of patterns of discourse (Maykut & Morehouse, 1996) was deemed most appropriate for the analysis of the participant’s stories. The following section provides a description of one suicide survivor bereavement support group, followed by the narratives of loss as told by some group participants. Through systematic analysis of the field notes, patterns in the narratives were inductively identified.
The suicide survivor bereavement support group discussed below is representative of an ongoing series of support groups offered at regular intervals throughout the year in southwestern Pennsylvania. The groups are open to adult members of the local community and the surrounding counties within a tri-state area, including West Virginia and Ohio. Participants drive as many as three hours to attend meetings. Typically the group consists of approximately 10 to 20 participants, 18 years of age or older, and any relationship to the deceased (i.e., parents, spouses, siblings, friends, coworkers, etc.). Each intervention consists of an 8-week psychoeducational support group that runs in two-hour sessions and is facilitated by an advanced practice psychiatric-mental health nurse and a social worker. Because of the nature of the group, the leaders tend to be more directive in their approach to the therapeutic process. Frequently, students from a variety of health care disciplines request permission to attend the group. If these observers commit to attend all eight sessions, the permission of the participants is requested. If the participants agree, observers are permitted to attend, introduce themselves, and are then integrated into the group.
The 8-week intervention model is set up in a number of sections. The sections are described in detail below, but generally, the program consists of five parts. Part one, the introduction, takes place during week one. The second part is two, three, or four weeks in duration, and provides each participant time to discuss the loss of their loved one. The third part consists of interactive presentations about suicidology, with updates on the most recent research in the field. The fourth section focuses on adaptive coping skills and strategies, while the final section (in week eight) is dedicated to termination of the group and discussion of resources for ongoing support.
The specific group presented in this case study is typical of other groups, consisting of seven members who attended all or part of six or more sessions, and three members who attended four or fewer sessions. Both the advanced practice nurse and the social worker facilitating the group had extensive experience in developing and leading such suicide survivor bereavement support groups. A graduate student in social work and a health communications instructor also attended the sessions described below. While their participation in therapeutic discussions was limited to answering questions posed directly to them by participants, they were included in group “rituals” (e.g., sharing of snacks, greetings, and leave-taking) which spontaneously developed.
Although the development of a particular form of narrative was not explicitly identified as a goal by the facilitators or participants, observation reveals that the therapeutic interventions of the survivor support group can be viewed as fostering the development of agentic narratives. Polkinghorne (1996) describes agentic narratives as depicting events in such a way as to suggest the narrator is in control, despite disruption by traumatic life events. On the other hand, victimic narratives suggest that events affecting the narrator’s life are controlled by outside forces (Polkinghorne, 1996). The reader is referred to an earlier article in this issue entitled, “Client Narratives: A Theoretical Perspective” in which theoretical explanations for client narratives are explored (e.g., plot structures of agentic and victimic narratives).
The remainder of this paper is devoted to an in-depth examination of participant narratives from the suicide bereavement group described above. The narratives were recovered from extensive field notes and memos taken (with participant permission) during the group sessions. Anecdotally, the sharing of personal narratives was cited by many participants as being the most powerful healing activity of the 8-week group intervention.
Both agentic and victimic narratives are clearly identified in accounts of loss each member shares within the suicide bereavement group. Agentic narratives were common among members who could be described as highly functioning as evidenced by continued employment, successful resumption of social roles, and expressions of self-satisfaction with the progression of their grieving process. The agentic theme is clearly traceable in the following example, volunteered by the speaker, Amy, at the beginning of the third session.
|How do you start? How they died, or how of they lived? [She shows a family picture]. He completed [suicide] when we were separated. He was depressed, really bad, so that like I was thinking, ‘Get a grip!’ He didn’t eat or sleep, and his behavior changed … [We were together several] years, and had two kids. We had a turbulent marriage, he was into alcohol and depression. With all his abuse, we had a real rough road; I tried to be there but I lost track of me.||
|Did your kids blame you?|
|I ask if they’re mad; they don’t seem to “get” the question. They were mad before, like, ‘Why can’t we live together?’ But now they don’t make the connection. His family says, ‘He was fine before you left.’ The kids, if they think it, they don’t say it. It’s calmer now … My big focus is the kids, of course. They’ve had good therapy, I really think [the nurse facilitator, who counseled the children] has been wonderful, and I’m always sort of “taking their temperature” about it, like talking to them about how they’re feeling and that. Maybe too much … I’m just concentrating on going to work, do[ing] the school stuff, just keeping it all together pretty much.||
Amy relates her account of the loss of her husband as an agentic narrative. Although such events as her husband’s depression are “outside” agents, her actions are attributable “to a happening inside,” a hallmark of an agentic narrative (Bruner, 1994, pp. 49–50). She reports an ongoing process of self-examination that forms the basis of action. For example, Amy’s evaluation of the marriage (“I lost track of me”) caused her to initiate a separation from her husband. She engages in an ongoing evaluation of her children’s adjustment process and the effectiveness of their psychological treatment. Acknowledging the hard work involved, she appears content with her progress (“keeping it all together”) to date.
Arguably, the narratives of all survivors of suicide are potentially victimic, insofar as individuals become survivors through the agency of others. In her narrative, Amy overcomes this tendency, differentiating her actions from those of her husband. She gives evidence of this in a later session, when she speaks of her feelings of guilt:
|I feel guilty sometimes about the kids—not like I took their father away, no—but I need to do more for them because they suffered … I feel guilty about being wrapped up in things that don’t matter as much, instead of providing them with things that matter.||
Self-critique, which is cited as characteristic of the agentic narrative (Bruner, 1994), may underlie Amy’s unflattering evaluation of her role performance as a mother. While she experiences guilt related to her children’s suffering, she differentiates between the suffering caused by another (their father’s suicide) and that caused by her own agency. While her agentic perspective leads her to what may be considered a harsh evaluation of her maternal performance, it also permits her to articulate a solution to the problem, namely reprioritizing her attention to give her children “things that matter.”
In contrast to those in the group who told agentic narratives, survivors who appeared to experience more difficulty adjusting to their loss frequently organized their stories around victimic plots. Such victimic narratives portray the narrator’s life as being controlled by the decedent or other loved ones. Victimic narratives are shaped by somebody else who has power to impose their will upon the narrator (Bruner, 1994). The domination of such outside forces is seen in the following narrative of Bill, a personable man whose life has been plagued by drug abuse and failed relationships since his parents died in a murder-suicide, which the unfortunate young man witnessed while still a child:
|My dad was just so incredibly mean to me. Abusive. I don’t know, it was like, what did I ever do to him? For any little thing, just anything, he’d take off his belt and whip me. I’d run away and hide, or scream at him … but I got into a lot of shit on purpose … One day, he took off his belt like he did, and, I just stood there and was like, ‘Okay, come on, let’s get this over with.’ That just shut him up for some reason, I don’t know. He left me alone that night … I kept thinking, ‘God, why doesn’t he just die, or go away and leave us alone?’||
Bill’s description of defiance (“let’s get this over with”) illustrates the passive nature of what Bruner (1994) terms “reactive agency” (pp. 41–42). Bill’s action can be seen not as one of self-determination, but of self-preservation, in which he attempts to influence his father, and professes no insight into instances when he appeared to have control (“That just shut him up for some reason. I don’t know”).
Bill’s account of his life is also marked by victimic themes in which others dictate the direction of his life. For example, he talks about being abandoned by his wife and brother who “broke off contact with me. I just think he didn’t want to get hurt. If you don’t open up to people, they can’t hurt you.” These statements reflect a victimic orientation because the narrator avoids negative possibilities more than actualizing positive possibilities (Polkinghorne, 1996).
Just as Amy maintains her agentic narrative by minimizing the very real control her husband exercised over her life through his suicide, Bill preserves his victimic narrative by disregarding the impact of his own actions upon the lives of others. For example, he downplays the contribution of his substance abuse to the dissolution of his marriage:
|Sometimes I get so lonely and depressed, and I don’t really know why. I’m starting to see that it relates to all the losses in my life. I’m getting a divorce for the holidays—that totally sucks. She’s having an affair.||
|Have you been gone long?|
|I haven’t been home in sixteen months.||
Bill makes no denials, excuses, or apologies concerning his behavior. Such apparent unwillingness to analyze his actions may be seen as irresponsible behavior. However, such behavior can also be seen as consistent with the logic of the victimic narrative insofar as Bill perceives his behavior as determined by factors beyond his control. Any judgment Bill makes of his actions would be superfluous. While Amy continually (and often critically) examines her behavior, Bill examines the disappointment and abuse he suffered through others’ actions and hopes that a more benevolent agent will enter his life (e.g., “I have a deep need to have someone in my life. Everyone’s been swept away. It’s my fault with the addiction. I want to be loved by someone”). That is to say, if his situation is to improve, it will be through the agency of another.
The stories told by Amy and Bill are characteristic of agentic and victimic narratives, and typical of the accounts given by other group members during the therapeutic group process. After the stories have been shared, subsequent portions of the support group serve to assist participants in recasting victimic narratives into more agentic ones. For example, following the sharing of stories, the group turns to a discussion of suicidology.
The psychoeducational section of the group process focuses on a neurobiological model that attributes the suicidal act to changes in brain chemistry, rather than the conscious desire on the part of the suicide victim. For participants with strongly agentic narratives, and their tendency for self-criticism, the neurobiological model provides an awareness that the survivor is not at fault for the suicide. Thus, an understanding of suicidology may diminish the potential for damaging self-recriminations by placing the agency with the disease process, not with the survivor or decedent. This is illustrated by such statements as, “it’s a terminal illness, like cancer” or “she wouldn’t have done it if she had a choice.” The neurobiological model also works to maximize the agentic narrative’s ability to provide the storyteller with a sense of the future (Polkinghorne, 1996). As noted by one participant, “I kept asking myself why? What did I do or not do?—But, that’s not the question. The question is how to forgive and live.”
The transition from victimic to agentic narratives is further fostered when the discussion focuses on the use of adaptive coping strategies. This topic is introduced during session six or seven. Grief is presented as a form of “work” which the survivor must undertake with commitment. One participant seemed to understand this notion as reflected in the following statement: “If anyone tells you they can take away your grief, run away as fast as you can!” Likewise, participants learn that although grieving is difficult, the acute pain experienced in grieving will resolve. For example, one participant notes, “You’ll get to a place where your loved one will be in your heart, not always in your thoughts.”
Over the course of the group intervention, participant narratives evolved from containing more victimic themes to being more agentic in nature. Interestingly, participants also occasionally edited victimic themes from the accounts of others. The shift toward an agentic perspective is evident in the following interaction, which took place in the sixth week:
|Can something like this stuff [surviving a suicide loss] bring on depression? I was depressed once, clinically depressed, and like I just couldn’t get out of bed. They put me on [antidepressants]. I wonder if this is going to happen again now.||
|Grief is a normal process. It has symptoms similar to depression—loss of appetite, insomnia; the feeling that you just can’t get out of bed. But in grief, you do get out of bed. You get up and do what you need to do. It may not be easy, but you do it.||
|Yeah, you’re here today. It wasn’t easy, but you did it. You were able to overcome, make the choice, and be here to work through your grief.||
|I guess, but it’s hard.|
While Paul’s suggestion of depression could be rooted in a victimic plot, leading to the assumption of a passive sick role, both the facilitator and other group members moved swiftly to place Paul’s narrative in an agentic stance. They did this by emphasizing the positive actions of “getting up” and “overcoming,” which Paul appeared somewhat reluctant to adopt (e.g., “I guess, but it’s hard”). This served as the catalyst for a group discussion exploring the similarities and differences between grief and depression.
Over the course of the group intervention, the values that constitute an agentic narrative (e.g., choice, purposeful action, and self-determination) became more prominent in the participants’ discourse. The participants whose narratives were victimic in nature were often observed in later session to “cosmetically enhance” their narratives. For example, they recast their lack of action as “free choice.” Joe, a survivor of a murder-suicide that claimed the lives of several children offers one illustration. The following interaction occurred during a discussion of the importance of anticipating and controlling grief during the holiday season by actively making plans to prepare for painful emotional reactions.
|My family wants me to come out to Oregon and spend [Christmas] with them. I don’t think so. That’s never what I did. I spent it with [the deceased loved ones’ family]. I’ll probably do what I did last year. I just sat round and watched the tube, then made the choice to get the pictures out and torture myself looking at them.||
|I’d say Oregon sounds better than that!||
|Besides, sitting around with the pictures wasn’t what you used to do anyway, so, like, it’s going to be new, whatever you do. Yeah, I’m for Oregon.|
|And if you do choose to look at the pictures, you might want to choose to make it a time-limited thing. Like say, only 15 minutes and then put them away, and just keep the memories in your heart.||
In this case, Joe depicts his viewing of the photographs as an agentic “choice,” a depiction perhaps rejected by other participants who instead endorse a visit with family. The facilitator does not challenge Joe’s description of his actions as “choice,” but rather, suggests the agentic action of consciously limiting his time with the pictures.
Participants restructured victimic narratives as an expression of both the willingness and ability to exercise meaningful control over their life and the grieving process. This shift can be seen in the narrative of one participant who lost her brother. Telling her story for the first time, she described her behavior following her brother’s death: “I just feel so heartsick sometimes, so lonely for him, I just go in the closet and hug his jacket and cry. I don’t know … ” Here, she sees her life as being controlled by her brother’s action; she does not evaluate her own behavior or assume responsibility for her actions.
After the sessions on suicidology and the grief process, her account of the jacket shifted substantially: “Sometimes, when I feel really low, I hug his jacket. I can smell him, and feel closer, like he’s still here. And he is, of course, because I know he’s in heaven. It makes me feel better.” In this account, the participant has shifted away from the victimic perspective. She now examines her own behavior as she tells the group how hugging the jacket seems to help her (“I can smell him, and feel closer”). This is a major change from her previous narrative, in which hugging the jacket was something beyond her control and understanding.
In the final session of the group, she refers to her habit of hugging the jacket from a more fully developed agentic perspective. This is illustrated nicely in the following sentiment: “All his clothes are ready for the veterans [to be given away], but not that jacket. I don’t hug it much anymore. I limit my time with it, and most of the time it’s just good to know it’s there. It’s with me, just to remind me. I’m going to keep it.” This final version of the narrative reflects several instances of introspective and evaluative behavior. The participant reports not only her initiative in preparing her brother’s clothing for disposal, but also her decision to keep the jacket, and is aware of her reasons for keeping it. Evaluated through the paradigm of agentic and victimic plots, the participant’s narrative reveals a person who is no longer the helpless victim of the grief caused by her brother’s action, and in so doing has gained control over what was formerly an intrusive behavior. The jacket has become a treasured souvenir of her brother’s life, and perhaps also of her own growth from victim to agent.
Participants in the suicide survivor support group frequently reported heightened well-being and a personal sense of community through sharing their narratives of loss with each other. Observations of how participants responded to and modified narratives in the course of this group intervention suggest that Polkinghorne’s (1996) paradigm has value as a therapeutic tool. These observations also suggest that development of agentic narratives may be fostered through group interaction and intervention. The therapeutic goals of this particular group were centered on helping participants develop a sense of closure around the death of the person, independent of the suicide, and to regain a sense of well-being. The suicide survivor group process underscores the importance of a useful history, and of the “telling of the story” that gives meaning to the event. This suggests that therapeutic interventions should include identifying methods to help survivors create therapeutically useful narratives. It is also important to direct future investigative efforts towards understanding whether the agentic narratives reflect adaptive coping among participants or if agentic narratives reflect participant well-being that is attributable to other causes.
Ann M. Mitchell, The University of Pittsburgh, School of Nursing, Pittsburgh, Pennsylvania, USA.
Deborah Dysart Gale, The University of Missouri Applied Language Institute, Kansas City, Missouri, USA and The University of Pittsburgh, Department of Communication, Pittsburgh, Pennsylvania, USA.
Linda Garand, Western Psychiatric Institute and Clinic, Pittsburgh, Pennsylvania, USA and The University of Pittsburgh, School of Nursing, Pittsburgh, Pennsylvania, USA.
Susan Wesner, Western Psychiatric Institute and Clinic, Pittsburgh, Pennsylvania, USA.