A Call for Research: The Need to Better Understand the Impact of Support Groups for Suicide Survivors

Abstract

Support groups for suicide survivors (those individuals bereaved following a suicide) are widely used, but little research evidence is available to determine their efficacy. This paper outlines the pressing public health need to conduct research and determine effective ways to identify and meet the needs of suicide survivors, particularly through survivor support groups. After describing the various approaches to survivor support groups, we explain the need for further research, despite the inherent challenges. Finally, we pose several questions for researchers to consider as they work with survivors to develop a research agenda that sheds more light on the experiences of survivors and the help provided by survivor support groups.

PUBLIC HEALTH PRIORITIES: SUICIDE AND ITS DESTRUCTIVE EFFECTS

In America, 30,000 deaths occur by suicide each year (Centers for Disease Control and Prevention, 2006). Suicide is clearly a significant public health issue (US Public Health Service, 2002). Each suicide produces at least six and as many as hundreds of “survivors,” those people who are left behind to grieve and make sense of the death (American Foundation for Suicide Prevention, 2004; Crosby & Sacks, 2002; Provini, Everett, & Pfeffer, 2000). While the word survivor tends to be used in The United States of America, in other parts of the world, “bereaved by suicide” is a more widely used term (Beautrais, 2004). Rough estimates, based on only six survivors for every suicide, indicate that at least one in every 64 Americans (1.5%) is a survivor of suicide (McIntosh, 2006). The estimate of six survivors produced by each suicide is probably a very low estimation in most circumstances. Results from a national telephone survey indicated that 1.1% of people stated they had lost an immediate family member or other relative to suicide in the previous year (Crosby & Sacks, 2002). It is quite likely that the proportion of the population affected by suicide is substantially greater. For example, seven percent of Americans surveyed in the same national telephone survey reported that they knew personally someone who died by suicide in the preceding year (Crosby & Sacks, 2002). While these percentages may reflect only exposure to suicide, instead of the type of loss that would be associated with the need for survivor groups or other postvention services, some of these nonfamily losses may be in people who consider themselves survivors. These estimates, reflecting percentages weighted as a result of the sampling methodology, indicates that in 1994 an estimated 13.2 million Americans knew about suicides in their social network in the preceding year, of whom 2.2 million experienced the suicide of an immediate family member or other relative. What this research does not address is how many people consider themselves directly affected by suicides in their families or social networks and how many people may know of or be affected by suicide across their lifetime. Unfortunately, there are no data on the number of people who are affected to the point of seeking out services or consider themselves survivors following the suicide of a family member or individual in their social network.

It is currently unknown how many immediate and extended family members, friends, coworkers or classmates could be considered survivors. However, it is clear that survivors of suicide may face difficulties similar to those bereaved by other types of traumatic death. They also must deal with the unique problems associated with a suicide death, including a prolonged and intense search for the reason for the suicide (Wagner & Calhoun, 1992), feelings of being rejected by the deceased (Van Dongen, 1993), a distorted sense of responsibility for the death and the ability to have prevented the suicide (Dunn & Morrish-Vidners, 1987), and feelings of being blamed for causing the problems that began the suicidal ideation of the deceased (Silverman, Range, & Overholser, 1995). Individuals grieving a suicide death also appear to have elevated levels of anger, family dysfunction, and feelings of social stigmatization (Jordan, 2001). Some evidence suggests that survivors of suicide are at risk for their own suicidal behavior, through both genetic and cognitive pathways. Suicide rates have been shown to be twice as high in families of suicide decedents as in families in which a suicide has not occurred (Runeson & Asberg, 2003).

In addition to the risk of developing psychiatric disorders, survivors of suicide may experience complicated grief. Suicide survivors, similar to survivors of other types of sudden traumatic deaths, may have an increased incidence of traumatic or complicated grief and posttraumatic stress disorder (PTSD; Jordan, 2001). More research is needed on the longitudinal course of bereavement following suicide to fully understand these patterns (Jordan & McMenamy, 2004). Complicated grief is a syndrome that shares features with both depression and PTSD. Complicated grief is defined as intrusive symptoms of yearning, longing for and searching for the deceased, as well as four or more persistent symptoms of trauma as a result of the death (Prigerson et al., 1999). These symptoms of trauma include: avoidance of reminders of the deceased, purposelessness, feelings of futility, difficulty imagining a life without the deceased, numbness, detachment, feeling stunned, dazed, or shocked, feeling that life is empty or meaningless, feeling like a part of oneself has died, disbelief, excessive death-related anger or bitterness, and identification symptoms or harmful behaviors resembling those suffered by the decedent (Prigerson et al., 1999). Complicated grief has been shown to occur in adolescents and young adults as a result of a peer’s suicide and in adults as a result of the suicide of a family member or partner. Among these adolescents and young adults, complicated grief was associated with a fivefold increased risk for suicidal ideation after controlling for depression (Melhem, et al., 2004) and in the adults, complicated grief was associated with a 9.68 times greater likelihood of suicidal ideation after controlling for depression (Mitchell, Kim, Prigerson, & Mortimer, 2005). In addition, complicated grief appears to be related to both the onset of depression and a prolonged course of depression and PTSD (Melhem et al., 2004). Suicide survivors with closer kinship relationships to the decedent have been shown to have higher levels of complicated grief (Mitchell, Kim, Prigerson, & Mortimer-Stephens., 2004). Among survivors of suicide, complicated grief has been shown to be associated with a substantially greater likelihood of suicidal ideation in the month after the death, even after controlling for depression (Mitchell, et al., 2005). Thus, suicide survivors who experience complicated grief are at elevated risk for suicidal ideation, and potentially a suicide attempt.

Given the numbers of people at risk and the seriousness of outcomes associated with being a suicide survivor, it is vitally important to understand their needs and how best to meet them. Unfortunately, research has not focused on the longitudinal course of bereavement following suicide. There is a need for prospective studies to determine how suicide impacts individuals in the years following the death (American Foundation for Suicide Prevention, 2004; Jordan & McMenamy, 2004). In their review of the sparse literature on interventions for suicide survivors, Jordan and McMenamy (2004, p. 345) state that “careful longitudinal research with a diverse, community-based sample of survivors would greatly increase our understanding of the challenges involved and the coping skills required after a suicide”; such research would also provide insight into the majority of survivors who do not seek organized or professional assistance following a suicide. In a workshop sponsored by The American Foundation for Suicide Prevention (AFSP) and the National Institutes of Health (NIH) to specifically determine a research agenda for suicide survivors, participants agreed upon the need to determine the most common treatment (if any) utilized by survivors (i.e., treatment as usual), how survivors access treatment, how effective it is (including which elements and in what dose; American Foundation for Suicide Prevention, 2004).

Most suicide survivors do not seek out formal or informal support or mental health treatment. Only about 25 percent of 144 next-of-kin survivors surveyed by phone reported receiving any help since the suicide, despite seventy-four percent indicating a desire for help (Provini, et al., 2000). In another study, half of Norwegian bereaved survivors felt a need for professional mental health treatment, but only one quarter actually sought out help (Dyregrov, 2002). Most individuals who receive help do so soon after the death, but suicide survivors also appear to have difficulty initiating a search for help on their own (McIntosh, 1993). In a recent study of 63 survivors recruited at survivors of suicide groups and events (85% had attended a survivors’ support group), 38% of participants reported moderate to high difficulty in finding support resources (McMenamy, Jordon, & Mitchell, 2008). Of those who attended survivors’ support groups, 94% found them to be moderately to highly helpful. All of the survivors in the study reported that talking one on one with another survivor was moderately to highly helpful. Initiating a search for mental health care or peer support may be challenging due to extreme grief or difficulty locating resources in a community. Traditional therapy may be helpful for survivors (e.g., de Groot, de Keijser, Kerkhof, Nolen, & Burger, 2007), and many survivors may prefer individual or family psychotherapy to support groups. However, psychotherapy often carries high costs, insurance coverage of mental health care is often limited, and formal mental health treatment may be stigmatizing. Although data on support groups specifically for suicide survivors are limited, support groups have been described as the most common form of intervention for other forms of bereavement due to their convenience, low or no cost, and perception of being less threatening than formal mental health treatments (Levy, Derby, & Martinkowski, 1993). For these reasons, we will focus primarily on survivor support groups in the remainder of this manuscript.

SURVIVOR SUPPORT

Some survivors may turn to advocacy, training, or other suicide prevention work. This may also serve as a source of support and connection either in combination with their group experience or instead of it. Survivors have used their grief to fuel a campaign to change the way that Americans and policy-makers think about suicide. These efforts have led to numerous legislative successes from the introduction of Congressional resolutions recognizing suicide as a serious problem in the 1990s to the passage of the Garrett Lee Smith Memorial Act in 2004, the first ever authorization and appropriation for youth suicide prevention. Some survivors report that creating political will and actually seeing change is a healing experience, but evidence of the effect of advocacy as a component in suicide bereavement is completely lacking. Other survivors use their pain and grief to work toward suicide prevention and training with the goal of having other families not have to experience the loss that they experienced.

Mutual support groups may be helpful in that they allow members to feel a sense of identification with other group members and feel like they and others are benefiting from sharing their experiences and listening to the experiences of newer attendees. Thus, “veteran” group members might describe ways they made it through difficult times and issues along the way. Research has found that individuals involved in general bereavement support groups for spousal death often have contact with other group members outside the context of the group and report feeling close to other group members despite moderate meeting frequency of the group itself (Caserta & Lund, 1996). This contact with other group members outside of group meetings might be among those group members with the highest levels of depression, loneliness and life stress, but their contact is not necessarily related to the intensity of their grief or coping abilities (Caserta & Lund, 1996).

Support groups for suicide survivors are among the most widely available type of support for survivors. The websites of the American Association of Suicidology (AAS; www.suicidology.org) and the American Foundation for Suicide Prevention (AFSP; www.afsp.org) host directories of support groups across the United States. While there are over 400 survivors support groups listed in these directories with at least one group in each state, it is unclear if groups are widely available for survivors seeking them out, especially in less populated areas.

Many view participation in a support group as an essential part of working through bereavement following a suicide. For example, in their new book Touched by Suicide, Myers and Fine’s top two suggestions for coping after suicide loss include: “seek out other survivors” and “find a support group in your community or a chat room on the internet where you can connect to others who are now residents in your strange new land” (Myers & Fine, 2006, pp. 12–13). The AFSP Web site (www.afsp.org), which includes a section aimed at survivors, states: “for so many survivors, a crucial part of their healing process is the support and sense of connection they feel through sharing their grief with other survivors. The most common way this sharing occurs is through survivor support groups. These groups provide a safe place where survivors can share their experiences and support each other” (Available via www.suicidology.org/displaycommon.cfm?an=1&subarticlenbr=55 or www.afsp.org/support group). In Touched by Suicide, Carla Fine describes how connecting with other survivors “assures me, once again, that I am not alone, and gives me the courage and language to reach out to others for support” (Myers & Fine, 2006, p. 180). The following section describes the various approaches to survivor support groups and the limited body of research that examined these types of group.

APPROACHES TO SURVIVOR SUPPORT GROUPS

Support groups are naturally appealing to many suicide survivors, as described in SOS: A Handbook for Survivors, “Support groups provide one of the most valuable resources for suicide survivors. Here, you can meet and talk with (or just listen to, if you prefer) people who are in your shoes. You can openly express your feelings and experiences with a group of caring individuals who will never judge you…” (Jackson, 2003, p. 31). A survey of 149 survivor groups in the US and Canada in the early 1990s provides most of the information that is known about survivor groups. The study found that on average, groups had been in existence for 8–9 years and provided services to less than ten people in monthly or twice-a-month meetings (Rubey & McIntosh, 1996).

Survivor support groups vary in their format and design. Key group characteristics include leadership, membership, format, length and timing, and access to group, each described below.

LEADERSHIP

Survivor groups can be led by a wide variety of individuals. In Touched by Suicide, it is emphasized that “one of the key factors that makes or breaks a support group is the facilitator” (Myers & Fine, 2006, p. 181). Groups are sometimes led by survivors themselves. Some survivors received training (such as that available from The Link Counseling Center [www.thelink.org/suicide_aftercare.htm] or AFSP (www.afsp.org/facilitatortraining), while others rely on their life experiences. Other groups are led by trained mental health professionals, such as social workers or psychologists. Finally, a combination of leaders, most commonly a professional and a survivor, is utilized by some groups. In Rubey and McIntosh’s (1996) survey, a third of groups were led by a trained facilitator, 21 percent were led by a mental health professional only, 27 percent were led by a combination of a trained facilitator and mental health professional, and 10 percent were led by a survivor leader who has limited or no specialized training. Overall, one quarter of the leaders identified themselves as suicide survivors (Rubey & McIntosh, 1996).

Survivor support groups are often led by experienced veteran survivors. Within the mental health field, there is an established tradition of peer-led (also known as consumer-led) groups. Consumer and family-member led services are quite common—serving almost 20 percent of mental health consumers (Wang, Berglund, & Kessler, 2000)—and often supplement traditional mental health services. There is limited research on the outcomes of mutual-support groups and self-help organizations for mental health consumers (Goldstrom et al., 2006). No research has examined the role peer groups play for survivors of suicide.

MEMBERSHIP

Some suicide survivors attend groups with people bereaved from a variety of types of death while others attend those specific to suicide survivors. Some groups are specifically designed for certain types of survivors. It is most common for children to have their own groups (Pfeffer, Jiang, Kakuma, Hwang, & Metsch, 2002); however, in larger communities and online, there are groups based on relationship to the decedent creating separate groups for adult children of suicide decedents, sibling survivors, and parents bereaved following a child’s suicide. There is no evidence on whether groups based on relationships are more or less helpful than those for one type of survivor. In addition, most communities do not have enough survivors active in groups to support multiple groups, or separate groups for survivors of different relationship-types (e.g., child, sibling, spouse) to the decedent.

While cultural and religious beliefs may influence bereavement and responses to suicide, there are no studies which examine the unique needs for support of various cultural or religious groups or which compare bereavement after suicide across cultures (Beautrais, 2004).

Group Format

A common format for suicide survivor support groups is described in the book No Time to Say Goodbye:

We sit in a circle, with each person giving a brief introduction: first name, who was lost, when it was, and how it happened. I then ask the people who are attending for the first time to begin, because they usually have an urgent need to talk. The rest of the group reaches out to them by describing their own experiences and how they are feeling. The new people realize they are not alone with their nightmare. By comparing their situations with others, they also begin to understand that they don’t have a monopoly on pain (Fine, 1997, p. 151).

In Rubey and McIntosh’s survey, 76 percent of groups were described as a “sharing of experiences,” while the remainder included a combination of lectures and sharing of experiences (Rubey & McIntosh, 1996). Suggesting that sharing one’s story is beneficial, Jackson writes, “In addition to receiving help, you’ll find tremendous benefit in the help your testimony will undoubtedly offer to others” (Jackson, 2003, p. 28). Anecdotal as well as clinical evidence supports that sharing is useful and there is no published evidence referring to survivors affected negatively by their mutual support group experiences. However, empirical research is near absent concerning critical issues of intervention effectiveness, cost and benefits, and even about its safety. For example, the field does not yet know if this type of sharing is beneficial or whether hearing and repeating traumatic stories may actually retraumatize survivors.

While many groups use a general, open-ended format, some groups have closed membership and cycle through a set-structure in eight or ten weeks. It is unclear if the open-ended format leads to individuals being retraumatized by hearing stories of violent deaths over and over without learning appropriate coping techniques. Myers and Fine acknowledge that “for some people, support groups may not be that helpful or comforting” (Myers & Fine, 2006, p. 183). We do not know who does or does not benefit from attendance at which types of groups. In Rubey and McIntosh’s (1996) survey, 85 percent of groups were open-ended with no fixed number of sessions, 11 percent involved a fixed number of sessions and the remainder included both formats. There is some evidence that the typical structure of support groups involving self-disclosure and sharing of feelings may not be helpful, and might actually be harmful, to individuals with a more avoidant style of coping (Jordan & McMenamy, 2004). Males may have more difficulties than females with traditional support groups consisting of an open structure and sharing of feelings (J. Jordan, personal communication, 2008). Nothing is known about how the effectiveness of the group is affected by group size, duration of attendance, admission practices (rolling versus closed), theoretical orientations (e.g., family systems), context (e.g., within the structure of a faith community), or setting (such as a home, church, or a professional facility).

Other group formats have been utilized and show preliminary evidence that they may be effective. A theory-based group program for parents bereaved by their children’s sudden death (including suicide, homicide, or accidental death) which combines psychoeducation, skill-building, and emotion-focused supportive discussion was associated with improved psychological functioning, reduced PTSD and improved physical health when compared to a nontreatment control group (Murphy, et al., 1998). Preliminary study also suggests that another helpful model is an 8-week group for all adult survivors of suicide that includes psychoeducation, adaptive skills, and narratives about the death developed by each group member (Mitchell & Kim, 2003). A study of a suicide survivor group for children which included a substantial psychoeducation portion for the children (as well as their survivor parents) appeared to show improvements in the parents’ depressive symptoms compared to a no-treatment control (Pfeffer et al., 2002). In an entirely different format, a one-session family-focused debriefing intervention showed trends towards positive outcomes in terms of grief and perceived stress a month (Mitchell, Evanczuk, & Lucke, 1999) and three months (Mitchell & Kim, 2003) after the intervention.

Length and timing of group interventions for survivors are also important to consider. As stated above, some groups are open-ended and some survivors attend these for years, while other groups end after a certain length of time. While it is unknown which kind of intervention is the most helpful, most bereavement group interventions are attended for such a short time that they seem to be of “insufficient strength and duration to make impact” in the life of the bereaved (Jordan & McMenamy, 2004, p. 344).

Finally, online groups have become a popular option, especially for those survivors who live in an area without a formal support group or who may not want to disclose their identity as a survivor. Online groups can be for a specific population (e.g., those bereaved by the suicide of a spouse, child, or sibling), and a directory can be found on SPAN USA’s website (www.spanusa.org/onlinesupportgroups). Several survivor organizations offer such groups, including Survivors of Loved Ones’ Suicides (SOLOS; www.1000deaths.com), Grief Recovery Online founded by Widows and Widowers (GROWW; www.groww.org), GriefNet.org (www.griefnet.org), and Yellow Ribbon Suicide Prevention Program (www.TeenHelp.org).

A 2005 report commissioned by SPAN USA indicated online groups provide such services as regularly scheduled facilitated chats, e-mail discussion lists, and message boards (SPAN USA, 2005). The appeal of online support is evident: one e-mail discussion list alone receives between 1,000 to 2,000 e-mails a month, and the moderator of one group felt it is the “intimate, anonymous nature of the computer which allows a normally reserved, shy individual, who may also be feeling ashamed and guilty over their loss of a loved one to suicide, to share his or her deepest feelings” (SPAN USA, 2005). However, there is no published research on online survivor groups, including questions about the nature and composition of the group and its leadership, the length of time and manner in which individuals participate in group, indications and contraindications, or indeed if the groups have benefit or cause harm with regard to participants’ psychosocial and psychological functioning.

TIMING

There is no research, and seems to be no consensus, about the optimal time for survivors to join a group after their loss. The AFSP website states, “some survivors attend a support group almost immediately, some wait for years; others attend for a year or two and then go only occasionally—on anniversaries, holidays, or particularly difficult days.” Until recently, traditional attempts to reach survivors have been passive in their approach to recruiting survivors, waiting for the survivor to seek out services. The Active Postvention Model (Campbell, Cataldie, McIntosh, & Miller, 2004) aims to reach out to survivors to help them access resources including survivor support groups as soon as possible following the death.

ACCESS TO GROUP/POSTVENTION

There is some evidence that most referrals for survivors of suicide groups in the United States come from physicians or nurses, professionals who typically share referral information with survivors when the death occurs at a hospital (Rubey & McIntosh, 1996). Many suicide deaths occur outside of a facility and are pronounced at the scene; therefore, a hospital is never involved and cannot serve as a primary referral resource for survivors. Postvention, defined as “interventions after a suicide,” is “aimed at reducing the impact of suicide on surviving friends and relatives” (US Public Health Service, 2002, p. 41) by assisting survivors in finding professional and peer support. Even when resources are available in communities, the length of time between the death and the survivor seeking help is often very long, partially due to a lack of knowledge of the resources by the survivors and by healthcare workers and other gatekeepers (Campbell et al., 2004; Cerel & Campbell, 2008).

Local Outreach to Survivors of Suicide (LOSS) teams are one example of active postvention for survivors. A team of mental health professionals with extensive training in assisting suicide survivors, volunteer crisis center staff, and volunteer survivors respond to the scene of a suicide in addition to the traditional first responders (e.g., police, emergency medical personnel, coroner; Campbell et al., 2004; Campbell, 1997). Under this model, outreach to survivors begins as close to the time of death (or notification) as possible. The team lets survivors of suicide know that resources exist, provides support services and referrals to all those identified as potential survivors of suicide, comforts survivors at the scene, explains the protocols used to investigate the scene, and answers the many questions that arise when multiple responders are at the scene. Cerel and Campbell (2008) have reported that among survivors who seek treatment following a suicide, those who received active postvention seek services significantly sooner, and appear to be more likely to attend support group meetings and to attend more often than those who received no active postvention. LOSS-type programs have been implemented in communities across America and in Australia and Singapore, but no systematic prospective evaluation of these programs has yet taken place. What remains to be seen is if active postvention is related to fewer symptoms of trauma, depression, and complicated grief than for those who receive traditional passive postvention and whether seeking mental health services sooner is related to better long-term outcomes for these survivors.

NEED FOR EVALUATION

Survivor support groups are a commonly used resource for those grieving a suicide death. Given the wide variability among groups, there is a vital need to evaluate survivor support groups to determine approaches that are most helpful to survivors. In addition, it is essential to determine whether some approaches to groups may have no effect or may cause harm.

A tension exists between the needs of survivors—who are currently in pain and looking for ways to find immediate help; and researchers—who need precise definitions to systematically study phenomenon. To resolve some of this tension, research should take place at several levels concurrently in order to clarify best practices for those in immediate need while methods development and definitional research takes place. Research should help specify which approaches are most helpful for various types of survivors. To appropriately study survivor support groups, researchers must involve survivors of suicide in the design and implementation of their research.

Evaluation of survivor groups should include comparison groups whenever possible, and should be constructed so that differential responses associated with personal differences such as gender, race, ethnicity, religious beliefs, culture, personality, and survivor relationship types can be examined. Research evaluating survivor groups should acknowledge that sampling bias is likely to be present as individuals in mutual support groups may not be representative of the population of survivors, many of whom do not attend groups. Finally, because previous research on bereavement support groups has found that most groups are of “insufficient strength and duration to make impact,” it will be important for future research to determine how much time in group is an appropriate dose of treatment (Jordan & McMenamy, 2004, p. 344).

RESEARCH CHALLENGES

In the Institute of Medicine Report, Reducing Suicide: A National Imperative, there is no mention of the needs of suicide survivors or of the existence of survivor support groups (Goldmith, Pellmar, Kleinman, & Bunney, 2002). In the National Strategy for Suicide Prevention (US Public Health Service, 2002), there is little mention of the needs of survivors. The only objective specifically about survivor support programs is “by 2005, define national guidelines for effective comprehensive support programs for suicide survivors. Increase the proportion of counties (or comparable jurisdictions) in which the guidelines are implemented” (US Public Health Service, 2002, pp. 104–105). This objective has not yet been accomplished. This lack of attention to the needs and roles of survivors, including the use of survivor support groups, must be addressed as researchers, clinicians and policy makers move forward with suicide prevention activities.

Another major challenge to studying survivor support groups is the lack of definitional clarity about who is a suicide survivor and what constitutes a survivor support group. Developing this definition is hampered by the stigma of suicide which leads many survivors to not publicly disclose their experiences and suffer in relative silence. Nothing is known about these survivors who do not disclose their survivorship because of fear of negative community reactions. Some survivors may want to carry on with their lives and not acknowledge their grief; others want to quickly find treatment which can help, but not take part in research which may be seen as exploitative or cumbersome.

Finally, research on support groups for suicide survivors presents important ethical challenges, particularly conducting research with individuals who are extremely vulnerable using untested treatments with unknown efficacy and potential “side effects.” The following discussion on major research questions includes a call for controlled trials which addresses these ethical issues directly.

MAJOR RESEARCH QUESTIONS

Given this background, the need for new research on survivor support groups fall into four main categories: (1) methods development; (2) epidemiological studies; (3) naturalistic studies; and (4) controlled trials.

Methods Development

For sound research to take place, researchers need to define issues such as “who is a survivor?” Researchers also need to be informed by strong theoretical models which lead to research questions and help determine which parameters are relevant for measurement. Finally, there is a need for sound metrics to be developed for precise measurement of theoretically important constructs. For example, research could begin with standardization of the term “survivor,” determine what parameters define the term and agree on how to validly measure the construct.

As this is a field in which research is nascent, there is a place for qualitative research, mixed methods and participatory research in which survivors’ stories and beliefs can help to shape definitions and future quantitative research. In addition, the creation of registries of survivors willing to take place in research might be considered as a useful means of gathering appropriate sample sizes for future research. While this approach will lead to biased samples, these samples would still be a way to start the process of understanding survivors without having to recruit new samples for each research effort.

Epidemiological Studies

Epidemiological research should focus on determining the number of people who are survivors of suicide and the breadth of the survivor support group network. In order to better understand how support groups aid suicide survivors, and which groups work for which survivors, research must better define the extent of the survivor population. This epidemiological research should include surveys of members of the general public to determine how many people have been directly affected by suicide in their family, social networks and communities not only in the past year but over the course of their lives and which types of survivors attend support groups. This research also needs to determine how words such as Survivor do or do not define people who report losses due to suicide. For example, some clinicians who have lost patients to suicide have sought out support groups and identify with the role of survivor while others do not. In addition, a better understanding of different categories of survivors is needed to answer questions such as what differences exist based on kinship relationship, emotional closeness to the decedent or exposure to the death itself. Understanding which part of the survivor population attends support groups will help clarify the role of groups in the course of bereavement following suicide.

Epidemiological research also involves studying the types of survivor groups currently available and information about who does (and does not) attend. Replicating and extending the Rubey and McIntosh survey of survivor groups would be particularly helpful to determine how trends in group leadership, content and composition have changed in the last ten years. A new study could also include more precise questions about the characteristics of group leaders, including variables such as their academic discipline, age, gender, prior experience with suicide, training in group therapy skills, or level of empathy. An updated study of group leaders could be conducted easily via the Internet to encourage more responses or through more in-depth telephone interviews.

Naturalistic Studies

Several types of naturalistic studies are needed to determine the course of suicide bereavement and the role support groups play for survivors. Questions include:

• What is the natural course of bereavement for survivors? Research is needed to examine which variables (e.g., demographics, exposure to the suicide or the scene, interaction with first-responders, social support, interventions) are related to good or poor outcomes in terms of psychosocial, psychological, family, occupational, and health outcomes. This research needs to take into account cultural factors and compare the typical course of bereavement across cultures. Research is also needed to describe the broad spectrum of grief following suicide including non-pathological grieving associated with suicide. This research should include predictors of positive outcomes of bereavement (e.g., past coping and adaptation, social supports, etc.).

• Does participation in support groups play a role in the course of bereavement? Research also must address how support group membership is related to longitudinal outcomes of psychiatric symptoms, complicated bereavement and overall functioning. The first step is to conduct naturalistic studies of survivors who participate in existing groups. Survivors active in groups could be surveyed to determine which group elements they perceive to be the most helpful. Such studies will also help identify the normal course of “treatment” for most suicide survivors and determine how survivors access groups. This research will answer whether survivors see survivor groups as their primary form of treatment or as supplemental to formal mental health services, and which survivors find participation in groups sufficient for meeting their needs.

• Which types of support groups are perceived to be more helpful to survivors? A large-scale study of several types of groups can compare the perceived helpfulness and effectiveness of different types of groups. Researchers can compare survivor outcomes associated with the different group characteristics (e.g., leadership, membership, format, length and timing, and access). This research can help define for whom support groups may be most beneficial and when. It can also guide future research on tailoring groups to people at different levels of grief and at different stages in the process. This research can also look at group therapy in the context of other mental health services to determine when individual psychotherapy or medication might be indicated for survivors as their sole treatment or as adjunctive to support groups.

• Are there group characteristics associated with poor outcomes? One important question is whether the survivors sharing their stories can itself be traumatic or slow recovery?

• What is the relationship between survivor support groups and advocacy/working for suicide prevention? The role of advocacy/suicide prevention work in decreasing deleterious outcomes for survivors is an important one which is worthy of study. It may be that support groups lead some survivors to the advocacy role/suicide prevention while for others advocacy in itself is more helpful for them than group support.

Controlled Trials

After a time, gains in research described above will support the development of controlled trials. For example, those theory-based groups which seem to be more typical therapy groups and which have been shown in preliminary studies to be effective might be ready for controlled trials sooner than the typical open-ended survivor support group. In these studies, survivors can be randomly assigned to groups with various characteristics and a variety of treatment modalities and followed over the course of their exposure to “treatment.” It is in this stage of the research that we can learn which approaches to leadership, membership, format, timing, and access lead to better outcomes for which types of survivors. Ethical considerations will certainly be complex in this stage of the work; however, as more is learned through the earlier stages of research, they will likely become more manageable.

CONCLUSION

Suicide survivors may be at increased risk for PTSD, complicated grief and suicidal ideation. There is a tremendous need for research to understand the needs of suicide survivors and the benefits they may gain from participation in support groups. Support groups are commonly sought by survivors of suicide and become an understanding community that can help ease the pain of their grief. These support groups vary greatly in their leadership, membership format, timing/ length, and access. One important finding of this research is that some theory-based groups, which are more similar to traditional therapy groups, seem to show preliminary efficacy and merit future research. Yet, little is known about the effectiveness any of these groups in meeting any of their desired outcomes.

As support groups may have deleterious effects for survivors, it is also important to determine which characteristics of group may contribute to or cause harm. Working with survivors, researchers are called to craft a thoughtful research agenda that includes methods development, epidemiological research, naturalistic studies of existing groups, and ultimately controlled trials of promising treatments.