Suicide Survivors: The Aftermath of Suicide and Suicidal Behavior

John L McIntosh. Handbook of Death and Dying. Editor: Clifton D Bryant. Volume 1. Thousand Oaks, CA: Sage Reference, 2003.

Historically, one of the most neglected areas of suicidology (i.e., the scientific study of suicide) has been the issue of the aftermath of suicide and suicidal behavior. As this discussion will show, this disregard has lessened in recent years, particularly for the topic of “survivors of suicide.” Still largely ignored is the aftermath for those who have made nonfatal attempts on their own lives and for their significant others. In this chapter, I provide a review of the literatures on these topics and suggest some directions for future research on this important component in the full range of suicidal behavior.

In the suicide literature, Albert Cain edited a pioneering work in 1972 titled Survivors of Suicide. The chapters in this volume outlined the issues of grieving associated with the loss of a loved one to suicide. As I will show below, Cain and the other contributors to Survivors of Suicide discussed the psychological and social factors associated with the aftermath of deaths by suicide. Up to that time, however, there had been almost no research investigations of this topic, so the contributors presented much conjecture and extrapolation from theory and other information. More than any other publication, this book established the term suicide survivor in the lexicon of suicidology. In his foreword to the book, Edwin Shneidman, the founder of the American Association of Suicidology and cofounder of the Los Angeles Suicide Prevention Center, referred to the “survivors of suicide, [as] … the largest mental health casualty area related to suicide” (p. ix). The importance of this topic was thus highlighted among the many dimensions of suicidal behavior by perhaps the most eminent figure in the field. Shneidman had previously coined the term postvention to refer to work with those who remain behind after a death by suicide. Within the other dimensions of suicide prevention, including prevention and intervention, postvention involves any issues of the aftermath of suicide deaths or other suicidal behaviors.

Despite Shneidman’s strong words about suicide survivors and the appearance of Cain’s book, only a small number of researchers, clinicians, and other suicidologists studied or wrote about this topic for some years after Survivors of Suicide was published. In this chapter, I will show that although they have continued to attract less attention than warranted, the research, clinical, and other literatures in this area have grown considerably in the past 15 years. These significant recent advances notwithstanding, our knowledge of survivors in many ways remains only a few steps beyond the level of understanding that existed 30 years ago when Cain’s seminal effort appeared.

It is important at this initial point to continue a bit more on the topic of the words employed in this chapter. In one of the first full volumes devoted to the topic following Cain’s book, a book edited by Dunne, McIntosh, and Dunne-Maxim (1987), I defined the issue of “survivors” by stating that “the death is the result of suicide, and the lives [being considered] are those of the family and friends who remain after a person commits suicide” (McIntosh 1987a:xvii). Thus survivors of suicide are those who have lost a significant other to a death by suicide. The focus is on the aftermath of the suicide death—that is, on the lives of these individuals and the resulting psychosocial effects of their loved one’s suicide. Although the term suicide survivor has been employed with some clarity, the terminology is imperfect and can create confusion. This confusion arises primarily among persons who are not members of the suicidology community. The suicide literature includes other confusing, awkward, and even offensive terms. For example, successful suicides are those whose acts end fatally, whereas those who remain alive are unsuccessful, as they failed at their attempts. In other words, those who have made suicide attempts that did not have fatal outcomes are often seen as having “survived” their acts. Thus the term survivors of suicide is not ideal. A few other expressions have emerged in reference to those who have lost loved ones to suicide, but none is used as widely as survivors. In Australia, for example, survivors of suicide are often referred to as the bereaved of suicide, a more precise wording that avoids the confusion that occurs elsewhere with the term survivor. However, the reality is that the term survivors of suicide is now so widely utilized among the suicidology community that, at least to members of that community, the term does not generally produce misunderstanding.

Epidemiological Aspects of Suicide’s Aftermath

A major problem in the suicide survivor literature is the lack of epidemiological investigations. Such studies are important to quantify the problem, because numbers often determine whether an issue receives attention and, particularly, needed resources, including funds for treatment, research, and prevention efforts. In order to examine the extent to which suicide survivorship is a mental health issue deserving of focused attention and support, researchers must initially determine the number of individuals involved. Eventually this will also involve calculating the “cost” of suicidal behavior and its effects on society and individuals.

Although suicidologists have identified the need for such research for some time, no national epidemiological study has been conducted to clarify the probable number of individuals who have been affected by the suicides of loved ones. In this void, I have advanced estimates of the number of survivors in the U.S. population as a basic and vitally needed figure (McIntosh 1989). Utilizing these conservative figures, and updating them to the most currently available data, we can estimate that 1 of every 62 Americans is a survivor of suicide, a conservative total of 4.4 million in 1999 (McIntosh 2001). If there are 6 survivors on the average for each suicide (Shneidman 1969), the more than 29,000 annual suicides produce an estimated 175,000 survivors each year—a pace of 6 survivors every 18 minutes (on average, there is a suicide every 18 minutes), more than 480 per day. On the other hand, based on one small study, Callahan (1989) has suggested a rate of suicide survivorship of 5.5% (and perhaps even higher), which would imply that there were more than 15 million survivors in 1999. Thus Callahan and others have suggested an even higher ratio than Shneidman’s 6:1, but without a high-quality epidemiological investigation, the estimate will have to suffice (“Interaction” 1996) as a conservative indication of the number of suicide survivors.

It is, perhaps, understandable that even basic information about suicide survivors would be missing given the long-term neglect of this topic described above. Surprisingly, the extent to which the aftermath of suicide attempts is a problem has been entirely ignored, despite the constant discussion and study of suicide attempters throughout the history of suicidology. Although no official statistics are compiled for suicide attempts, a large literature describes the many investigations that have been conducted concerning those who make nonfatal suicide acts, including the epidemiology of suicide attempts. The U.S. Surgeon General’s National Strategy for Suicide Prevention presents estimates that there are between 8 and 25 suicide attempts for every completed suicide in the United States and that each year 650,000 persons are treated in emergency care settings following suicide attempts (U.S. Department of Health and Human Services 2001). Although some of these individuals represent multiple attempts, these estimates obviously suggest that the number of attempters in the population is far larger than the number of survivors. Larger still are the numbers of persons whose lives are affected by the suicide attempts of their family members or friends. If each suicide affects at least 6 others, it seems logical that this figure would be a conservative one for the number affected by suicide attempts. The ratio is probably even more conservative than for suicide survivors, because suicide attempters are younger as a group than those who die by suicide, and as a result the number of individuals affected is likely to be larger. However, using the figure of 6 individuals from each attempt, that would mean that the lives of more than 4 million Americans are affected by suicide attempts each year. If these annual figures were cumulated over time to account for a “lifetime” risk, the number would be substantial.
Survivors of Suicide

At least three distinct subtopics exist with respect to the bereaved of suicide. A growing research literature that attempts to delineate the characteristic aspects of suicide bereavement has emerged over the past 30 years, along with a body of personal accounts by individual survivors of suicide. Finally, descriptions of clinical and other therapeutic interventions that assist survivors in their efforts to heal following the loss of significant others have also appeared. I discuss each of these topics briefly below to portray a shared primary goal of researchers, clinicians, support group leaders and facilitators, and survivors themselves: to understand suicide bereavement, including its commonalities and individual differences.

Research on Survivors of Suicide

Just as epidemiological information is needed, researchers must conduct methodologically sound research investigations if they are to obtain the community support and allocation of resources required to improve our understanding of suicide survivorship and assist in the healing process. As several scholars have noted in detail, early research investigations on survivors of suicide were marred by a number of methodological problems that limited the usefulness of the findings and their ability to portray suicide survivorship fully (for critiques, see McIntosh 1987b, 1999; Calhoun, Selby, and Selby 1982; Henley 1984). However, these studies served the important role of initially describing various aspects of suicide survivor grief.

Most glaring among the limitations and problems of these early studies was their inability to show how suicide grief is different from bereavement associated with other modes of death. The basic reason for this limitation was that the studies lacked comparison groups—that is, the researchers included only suicide survivors in their samples. At the time of Calhoun et al.’s (1982) review, not a single study had included a comparison or control group. Differences in grieving require that researchers include comparison groups, applying to them the same methods, procedures, and measures as they apply to suicide survivors. Despite this problem (and others), these exploratory, descriptive, non-comparison group studies advanced the knowledge base about the experience of surviving suicide. They particularly helped to demonstrate the variability in experiences as well as probable common aspects of suicide survivorship. I describe some of the findings of these studies below in the context of the larger research literature.

Early Non-Comparison Group Literature

A number of scholars have published reviews of the body of research on survivors of suicide (Calhoun et al. 1982; Clark and Goldney 2000; Cleiren and Diekstra 1995; Ellenbogen and Gratton 2001; Foglia 1984; Hauser 1987; Henley 1984; Hiegel and Hipple 1990; Jobes et al. 2000; Jordan 2001; McIntosh 1987c, 1993, 1999; Ness and Pfeffer 1990; Rudestam 1992; van der Wal 1989). Apparently, the first comprehensive compilation and review of the existing literature appeared in Cain’s (1972a) introduction to his classic edited book Survivors of Suicide (see McIntosh 1999). In his “clustered and capsuled” portrayal, Cain noted nine reactions that are all included in what Dunne (1987:143) would later call the “survivor syndrome” (see also Dunne and Wilbur 1993). Reflecting some of the compiled psychodynamic considerations, Cain described these reactions as follows: reality distortion, tortured object-relations, guilt, disturbed self-concept, impotent rage, search for meaning, identification with the suicide, depression and self-destructiveness, and incomplete mourning (pp. 13-14). Cain also noted in the scant literature behaviors such as anniversary reactions associated with the death, preoccupation with the phenomenon of suicide and involvement with prevention efforts, and feelings of shame, stigma, and abandonment.

Calhoun et al.’s (1982) review 10 years later added few issues to Cain’s list, although the number of empirical studies had increased and the findings of those studies supported the set of reactions noted earlier. The review and subsequent research yielded some additions to the survivor syndrome. The lack of comparison groups precluded Calhoun et al. from making any definitive statements regarding how suicide survivors and their survivor syndrome might differ from or be similar to the grief and bereavement observed among survivors of other modes of death. However, Calhoun et al. expanded Cain’s list of reactions to include several categories and a number of new reactions. They organized the findings into affective (i.e., emotional), cognitive, behavioral, physical, and family interaction reactions. In most cases, a number of researchers had reported these reactions among suicide survivors. These additional reactions included relief, shock and disbelief, health-related problems as well as more physician visits and even higher mortality rates, and possible negative effects on the family system.

One of the most interesting elements of Calhoun et al.’s review is the authors’ best guess as to a list of reactions that might eventually be determined to be unique aspects of suicide survivorship. Recognizing the tremendous methodological limitations in the research, Calhoun et al. noted that they believed enough consistency was present to support three cautious generalizations. Specifically, they said, suicide survivorship may be unique in survivors’ “search for an understanding of the death,” greater feelings of guilt, and “the lower levels of social support” they were likely to receive (p. 417) compared with other bereaved persons.

Evidence was lacking to show that this list of reactions for a possible suicide survivor syndrome represents unique aspects of suicide survivor grief. The reactions listed may well be common to grief following suicide, but at the time of their compilation, research verification regarding their uniqueness to suicide bereavement did not exist. Calhoun et al. could only conclude that how much bereavement associated with suicide differs from bereavement associated with other modes of death was unclear, and that the uniqueness of suicide survivorship may be limited to a small subset of reactions.

In my own work, I have suggested that these findings might be interpreted differently. Perhaps a larger set of reactions isunique to suicide’s aftermath, not simply in their appearance, but in the quality of their appearance. In other words,

a reaction that is labeled the same by suicide survivors and those from other causes who are interviewed may indeed occur for both, but there may be facets and issues associated with that reaction that are unique to bereavement from one cause as compared to another. For example, suicide survivors and accident survivors may both feel some degree of guilt for actions they performed or those they might not have performed which they associate with some degree of responsibility for the death of their loved one. However, the amount and subjective aspects of guilt may differ generally for suicide as opposed to accident survivors. (McIntosh 1999:164)

At the time of Calhoun et al.’s review, without comparison groups, research regarding this interpretation did not exist, and, in fact, such an approach is virtually nonexistent even in the current comparative research literature. Among many critical research needs is a focus on qualitative as well as quantitative aspects of specific bereavement reactions.

Comparison Group Literature

Since Calhoun et al. (1982) published their review of exploratory investigations, researchers have conducted a growing body of studies with stronger methodological characteristics. A decade after Calhoun et al.’s review, which was devoid of comparison group studies, I reviewed 14 published studies that included such groups (McIntosh 1993). By 1999, I had found and reviewed an additional 16 investigations that had appeared in the published literature. These 30 studies (and several more since 1999) all included various survivors of other modes of death in comparison groups, were methodologically more sound than previous studies, and represented an impressive array of relationship or kinship relations between the deceased and the survivors. Although methodologically superior to the early research with respect to the addition of control/comparison groups, research investigations that directly compare survivors of various causes of death often suffer from their own methodological flaws (for a detailed consideration, see McIntosh 1987b; see also Cain 1972b; Calhoun et al. 1982; Henslin, 1971). These problems include the use of selective samples (e.g., college students or support group attendees), the use of retrospective designs (with researchers often asking survivors to recall feelings from many years earlier), the inclusion of small numbers of individuals, the use of nonstandardized instruments of unknown reliability and validity, the omission of measures of emotional closeness to the deceased, and the omission of many important relationship categories of survivors. Although methodologically still limited, such direct comparison group studies are able to address possible bereavement differences and similarities between survivors of suicide and survivors of loved ones lost through accidental and natural deaths.

In previous work I have summarized the overall body of research (McIntosh 1993, 1999) and concluded that all the investigations examined have had methodological problems. Despite these issues, collectively, their results suggest the following six points: (a) Evidence supports findings of a generally nonpathological bereavement reaction to suicide; (b) more similarities than differences are observed between suicide survivors and those bereaved by other modes of death, particularly when suicide survivors are compared with accidental death survivors; (c) there are possibly a small number of grief reactions or aspects of grieving that may differ or are unique for suicide survivors (and these unique reactions, along with the larger number of others, may constitute a nonpathological but definable “survivor syndrome”), but the precise differences and unique characteristics are not yet fully apparent; (d) the course of suicide survivorship may differ from that of other survivors over time, but (e) by some time after the second year, differences in grief seem minimal or indistinguishable across survivor groups; and (f) the kinship relation of the survivor to the suicide as well as the precise closeness and quality of the relationship and the time that has passed since the suicide seem to be important factors in bereavement.

In addition to these general summary findings, as might be expected, some aspects of grief have been found to differ for suicide survivors compared with survivors of deaths by other modes, whereas other reactions have not been differentially observed between these groups. Unfortunately, these results have not necessarily been consistent across studies of survivors of the same kinship relations or across kinship relations. As this would indicate, the research literature contains many inconsistencies and contradictory findings. It should also be noted that among reviewers of the research literature, some disagree with my interpretations and those of others that the findings do not clearly show unique differences between suicide bereavement and that following deaths by other causes (e.g., Ellenbogen and Gratton 2001; Jordan 2001). These writers believe that the existing evidence is compelling enough to support assertions of differences. For instance, Jordan (2001) finds the qualitative evidence strong enough to support three “themes” with respect to suicide bereavement. The first two are among those given by Calhoun et al. (1982) in their cautious generalizations, whereas the third is not. These three themes of suicide grief are (a) the greater struggle to find meaning in the loss of the loved one; (b) greater feelings of guilt, shame, responsibility, and blame; and (c) greater feelings of rejection, abandonment, and anger toward the deceased. Jordan also reviews the findings on social networks and argues that the bereaved of suicide receive less social support (the third of Calhoun et al.’s generalizations) and have more disruptions of social interactions than do survivors of other deaths. Ellenbogen and Gratton (2001), on the other hand, believe that when studies examine the differences between various kinds of bereaved groups, they ignore the variety of individual differences among suicide survivors. In other words, these authors encourage researchers to attempt to determine whether there are definable patterns of bereavement within the population of suicide survivors. The existence of such discernible and distinct subpopulations might explain the inconsistencies and contradictions in the research literature.

Any search for a single suicide survivor syndrome, whether in a clinical or an experiential bereavement context, will yield only partial consistency among the unique and shared reactions observed. In addition to, and in combination with, the emotional closeness of the bereaved to the deceased, one of the reasons for the variability in bereavement reactions is the variability in relationships or kinship relations of the individuals. The majority of comparison group investigations, therefore, have involved survivors of a single relationship category to the individuals they survive (e.g., parents only, widows only), and rightfully so. Although there have been exceptions (i.e., studies that have combined survivors of all relationships into a single group), studies throughout the history of research on this topic have largely examined groups with a single form of kinship relation. Each type of relationship likely involves somewhat distinct grief and bereavement aspects. Specific relationship studies seem most likely to reveal findings that will enhance our knowledge and understanding of survivor issues. Although many of the reactions noted in the early research are probable in all survivors, the kinship relation may alter not only the reactions but also the specific set of reactions experienced. It is also important to remember that the entire set of reactions noted here will occur in few individual survivors, and all bereaved will not necessarily experience these reactions to the same degree. As I have reviewed the research on relationships in detail in earlier publications (McIntosh 1987b, 1993), I will address only the most prominent issues here (for specific references, see McIntosh 1999).

Parent survivors (of a child’s suicide). Parents play special roles in the lives of their children, and they and society have specific expectations of the responsibilities of parents toward their children. These expectations and responsibilities are important factors in the experience of parents of children who die by their own hands. Although present in studies of nearly all relationships, guilt and its related dimensions (such as shame and stigma) are perhaps the most prominent aspects of grief found among parents who have lost a child to suicide. The literature on attitudes toward parents of suicides underscores the belief that these parents are blamed and held responsible for their children’s deaths and that they are liked less than parents whose children died by other means. Although not all of the few studies of parents have found consistent results for these reactions when parent suicide survivors are compared with parents of children lost through other modes of death, present findings often support this difference. Similarly, the issue of lack of support from others is prevalent in the accounts of parents, but research findings with comparative groups are equivocal on this aspect in some studies; in still other studies it has not been included in those grief aspects examined. Finally, as is the case with all modalities of death, both positive and negative outcomes in family adaptation occur, although parents surviving a suicide may more often experience negative adaptation, both in the family and in the relationship with the other parent.

Spouse survivors. As the marital relationship is often the most important relationship in a person’s life, the death of a spouse has potentially major social, psychological, financial, and even health ramifications for the surviving spouse, and particularly so when the death is by suicide. Nearly all studies of spouse survivors are of widows—that is, wives whose husbands died by suicide. This undoubtedly reflects the much higher levels of suicide by men and perhaps also the fact that women are more likely than men to agree to participate in research. Both early and more recent comparative studies have frequently found more similarities than differences for spouse survivors of suicide and accidental deaths. Specific study results have indicated differences between spouses who survive suicide and those who survive other modes of death, but there has not been much consistency in the particular aspects of grief for which differences have been observed. That is, some studies have found differences that have not been observed in other investigations. This includes aspects such as guilt, shame, stigma, and social support following the death. One important factor in the bereavement outcome may be the nature and quality of the marital relationship at the time of the spouse’s suicide.

Child survivors (of a parent’s suicide). The issue of the age of the survivor (as well as the suicide) has frequently been noted in survivor studies, but this issue is particularly salient when children survive the suicide of a parent (obviously, it is also important when the suicide was a sibling or other family member). It should be noted that the vast majority of studies of child survivors have focused on children who are mental health clients, and this aspect may confound to some degree the findings reported. Particularly, the vulnerability and high risk of child survivors for mental health symptoms is commonly observed. Among the other aspects of grief that emerge in descriptions of child survivors of parental suicide are the guilt that children feel, the identification of the child with the deceased parent, and the distortions and often misleading information provided to children by other family members about the death. It is worth noting here that studies of adult children who have survived the suicide of a parent have thus far not been conducted (or at least published), although the issues in such cases are detailed in several personal accounts appearing in Stimming and Stimming’s (1999) edited volume on child survivors. Stimming and Stimming summarize some possible common themes among the accounts they collected, including identification with the parent, feelings of abandonment, the children’s anxiety over their own risk of future suicide, lack of opportunity to know other adult child survivors with whom they can share their experiences, and effects on their own family relationships (with spouses and children) and work relationships. Although this set of commonalities is not unique to adult child survivors, the dimensions and characteristics of the reactions might well reveal special aspects on closer investigation.

Sibling survivors. When children die, not only do their parents become survivors, but in most cases deceased children had siblings who also become survivors. Children in this group, sometimes called “forgotten mourners,” have rarely been studied, and the subject of sibling survivors seems to receive little attention. The few studies of siblings that have appeared have, like those of child survivors of parent suicide, focused on psychopathology. When sibling survivors in these studies have been compared with controls, they have shown mixed results concerning the vulnerability of sibling survivors to psychiatric illness and symptoms following the suicide. One difference that has been reported, however, is a longer period of strong grief symptoms among those who survived the suicide of an adolescent sibling. Some siblings have reported experiencing rapid emotional and social maturity following the suicide. Issues of identification and guilt are also obvious possible aspects of responses among sibling survivors. No research has yet examined bereavement effects among siblings of various ages, particularly where the suicides involve adult siblings. As is the case for other kinship relations, many more studies of sibling survivors are required before definitive conclusions can emerge.

Therapist survivors (of a client’s suicide). Individuals with mental health problems, particularly depression, are at elevated risk for suicide. Obviously, many persons with mental health problems, especially serious ones, are actively involved in therapy. Thus the probability that therapists at some time in their careers will eventually experience the suicide of a client is fairly high. The topic of therapists as survivors of client suicides has been largely ignored in the suicide literature (although this is changing; see, e.g., information on the efforts of the American Association of Suicidology’s Therapist Survivors Task Force on the association’s Web site at http://www.suicidology.org). Existing information suggests that the suicide of a client often has a significant effect on the therapist, both personally as well as professionally. Among often-mentioned issues are concerns about lawsuits, doubts about professional competence, depression, and the usefulness of talking to other therapists who have also experienced client suicide. As is a common theme, therapist survivors have thus far not been adequately studied for us to understand how they are affected by client suicides.

Suicide Survivors Literature: Personal Accounts

Researchers combine the responses of many study participants to present their findings in brief summaries or small numbers of generalizations. In the process of showing “typical” or “average” reactions, however, researchers often make it appear that all survivors (overall or of a particular relationship to the deceased) exhibit similar grief reactions and bereavement processes. Thus research findings give us a general impression of the responses of groups of survivors, but they cannot provide us with the rich detail, complexity of information, and variety of individual survivor experiences. This is best and most poignantly accomplished when survivors are allowed to tell their own stories. Personal accounts also permit listeners or readers to show empathy and help to normalize reactions and assist survivors in understanding that they are not alone in their grief. My latest review of the literature includes a table of personal accounts that had appeared up to that time (McIntosh 1999:160), and others have appeared subsequently (e.g., Linn-Gust 2001, sibling; Stimming and Stimming 1999, multiple adult child survivors). These personal accounts come from suicide survivors in a wide array of kinship relations. When one reads personal accounts of the aftermath of suicide, one most often finds that they echo the generalities presented in this chapter, but the authors relate particular circumstances, events, and difficulties that are not represented in characterizations based solely on group results.

Finding the “Gift” Following a Suicide

Although no research studies have reported any findings on this issue and its incidence among survivors, those who have contributed to the personal experience literature have often mentioned finding a “gift” after a suicide. Many survivors express their wish to find meaning in the death of their loved ones. This is a subissue of understanding the loss that has often been noted by researchers and writers on this topic. Iris Bolton (1983), a pioneer in the area of suicide survivorship, was told by a psychiatrist and friend shortly after Bolton’s son’s suicide that there “is a gift for you in your son’s death” (p. 16). Although this statement outraged and perplexed Bolton at the time, she subsequently found the gift and has shared it with countless others over the years. One aspect of the “gift” emerged in her book My Son … My Son …, in which she shares her own experiences and feelings about her son’s death and the effects it had on her. This book has been of tremendous help to many survivors, particularly parent survivors. Bolton also completed an advanced degree focusing on the topic of suicide, and she now shares her experience and knowledge internationally in presentations. Her practical and invaluable suggestions for survivors have been widely disseminated (see Dunne et al. 1987:289-90). Other survivors have likewise discovered “gifts” in the suicide deaths of their loved ones. For example, after their daughter’s suicide, Jerry and Elsie Weyrauch founded the Suicide Prevention Action Network (SPAN), an extremely active national advocacy group that works to promote legislation to provide funding for and research on the topic of suicide and suicide survivors (see the organization’s Web site at http://spanusa.org). Following their son’s death, Dale and Dar Emme founded a youth suicide prevention program designed to empower and get help for those who are troubled or suicidal; this international organization is called the Yellow Ribbon Suicide Prevention Program of the Light for Life Foundation (see its Web site at http://www.yellowribbon.org).These are but a few examples of the ways in which some survivors of suicide have worked to benefit others who have shared their experience and at the same time healed themselves and found meaning in their loss. This aspect of the bereavement reactions of some survivors deserves more formal study.

Postvention: Therapy and Support for Suicide Survivors

As is true for many aspects of suicide survivorship, no definitive data exist regarding the numbers, proportions, or characteristics of survivors of suicide who seek or require postventive care. Over time, a variety of resources have been developed and have increasingly become available for survivors who seek assistance in dealing with their loss (for discussions of these resources, see chapters in Dunne et al. 1987). Although general bereavement resources such as support groups are available in many communities, and some suicide survivors will benefit from them, one clear theme of the personal experience literature is that suicide survivors benefit most from group support settings that are homogeneous with respect to the mode of death (i.e., that include only suicide survivors). Although there is no empirical support for this claim (it has not been investigated directly), some survivors have described their discomfort and feelings of being different, of not fitting in, as members of survivor groups in which the modes of death were mixed. Such feelings likely arise from the feelings of guilt, shame, and stigma so often observed among suicide survivors. Following their loss, survivors of suicide sometimes seek or are referred to traditional individual or group therapy services in their communities that specifically address suicide bereavement. In other cases, survivors seek, and often find, self-help and support groups, some of which are led by mental health professionals and others of which are facilitated exclusively by individuals who are themselves suicide survivors (a list of such resources in local communities throughout North America may be found on the American Association of Suicidology’s Web site, at http://www.suicidology.org). Although limited information exists about these various resources (see, e.g., Rubey and McIntosh 1996), anecdotal evidence testifies to the tremendous assistance survivors receive from them in their healing process. In a promising recent evaluation study that incorporated several desirable research characteristics, Pfeffer et al. (2002) provided a group intervention for children who were survivors of relatives’ suicides to one group and not to another. The researchers found that the group intervention (which focused on the children’s reactions to their relatives’ deaths as well as on strengthening the children’s coping skills) lessened the distress of children who had lost a parent or sibling to suicide compared with those who did not receive the intervention. More such research-based evaluations and outcome evidence are sorely needed.

Survivors of Suicide: Research, Therapy/Support, and Educational Needs

Thus far in this chapter I have summarized our current knowledge of the varied facets of suicide survivorship. As can be seen, answers to many questions are beginning to emerge, but clearly much information remains unknown or uncertain. The need is great for further research and clinical attention to suicide survivors to expand our understanding of and ability to intervene effectively with suicide survivors. The knowledge base on survivors has expanded significantly since Cain’s (1972b) pioneering effort, but Cain (1972a) noted many of the issues and needs indicated here in his introduction to Survivors of Suicide. The limitations that Cain noted 30 years ago persist to the present (for review and discussion of these limitations, see McIntosh 1987b).

Basic Information Needs

Definition and precision of the term “suicide survivor.” In work published more than 15 years ago, I stated that among the most basic issues with the widest implications for this field is agreement on nomenclature surrounding suicide survivors (McIntosh 1987b). In the most general sense, the meaning of the term suicide survivor can now be considered resolved (as discussed above). However, scholars must learn to apply the concept of “suicide survivor” with far greater precision and specification, to delineate clearly the populations investigated and served. The general definition of suicide survivors noted earlier includes the affected family and friends of those who die by suicide. As I have pointed out elsewhere, “This broad definition may include many family relationships (immediate and extended family) and nonfamily relationships (friends, therapists, coworkers, fellow students, etc.), from those emotionally close to those remote from the suicide, and individuals greatly and hardly affected by the death” (McIntosh 1987b:264). And as I have observed in relation to the personal accounts noted above, each relationship category represents a group and individuals within the group who have vastly different survivor experiences. At a minimum, researchers and clinicians must indicate precisely what subpopulations they have studied or assisted to indicate the possible limits of generalizability of their findings (e.g., findings for parent survivors may not apply well to children who survive a parent’s suicide). In addition, it is important to determine how far the concept of “affected by the death” extends. For instance, how much (qualitatively and/or quantitatively) does a person need to be affected by someone else’s suicide to be considered a “survivor”?

Epidemiological information. As noted earlier, a vital need exists for epidemiological and demographic investigations to determine the number and characteristics of survivors in the general population. That is, Who are those we call suicide survivors? and How many survivors are there? More specifically, what are the distributions and numbers of suicide survivors by age, sex, race, relationship to the deceased, time since the death? If the bereaved individuals who make up this special group are to receive the research, services, and funding they deserve, researchers must compile the data needed to arrive at fundamentally sound estimates of the numerical dimensions of the population at risk. Methodologically sound investigations of the population are needed to provide this information.

Family history information. The clinical literature includes many studies regarding the family history of suicidal behavior in populations such as depressives or schizophrenics. Similarly, investigations of nonclinical populations of survivors are needed. Such research could help clarify the risk of suicide among survivors of suicide and the incidence of suicide among the generations of a family (and may potentially lead to a better understanding of social and biological factors associated with suicide).

Clarification of Bereavement Reactions and Experiences of Suicide Survivors

Determination of suicide survivor syndrome characteristics. The features of suicide survivors’ grief reactions must be determined through carefully conducted empirical research and clinical observation. From a biopsychosocial perspective, information is needed about the broad range of psychological and behavioral aspects as well as social, physical, and health changes. This information may identify survivor subgroups with differential needs with respect to postvention.

Comparison to bereavement by other modes of death. Nearly all contemporary research designs include comparison groups and should continue to do so. Descriptions of how suicide survivors’ grief and bereavement differ from and are similar to those for other sudden (accidental, natural, and homicide) and nonsudden deaths can be derived only from such designs. Once again, the results of such investigations may guide interventions with suicide survivors, including traditional therapy and mutual support groups.

Study of the variety of relationships. Not only do we need to understand how suicide survivorship differs from and is similar to that from other causes of death, we need better knowledge of how reactions to suicide are affected by the relationship of the survivor to the suicide. Although some researchers have begun to explore certain basic relationships, as reviewed briefly above, investigations into a number of potentially important relationships are also needed. Such relationships might include the following: “parents of child or adolescent suicides; parents of adult suicides; adolescents and children as survivors of parental, sibling, or friend suicide; widows and widowers; grandparents; unmarried lovers, including individuals in gay and lesbian relationships …; closest friends and confidants compared to acquaintances and casual friends; therapists; those who discovered the suicide” (McIntosh 1987b:267). Each of these relationships has the potential to produce relatively unique aspects in the grieving experience.

Focus on effective coping strategies. At least as important as identifying bereavement reactions of suicide survivors is the determination of successful coping mechanisms that survivors employ to deal with their grief and loss. Such information may inform better intervention and help for the bereaved.

Role of social and familial networks. Another important but largely unstudied aspect of coping with suicide is the social support network of the survivor (see also Jordan 2001). It remains to be determined how such networks react under various survivor circumstances and how a network’s extensiveness may affect a survivor’s reactions, their intensity and duration, and recovery. As noted earlier, one of Calhoun et al.’s (1982) generalizations was that suicide survivors receive less social support than do other survivors. Another facet of this issue is how the family system is affected by a suicide death with respect to contacts, closeness, and communication. Special, but commonly occurring, social settings in which suicides occur also need to be investigated with respect to systemic effects as well as individual reactions. Families are one such system, and in addition to families, schools and universities, work settings, therapy groups, and other systems are among those that should be investigated. Thus far, research regarding social and support networks, familial and nonfamilial, among suicide survivors has been inadequate.

General Research and Methodological Issues

Research with survivors of deaths, and particularly suicide deaths, involves a number of primarily methodological issues. Recent individual studies have been more methodologically sound than earlier studies, but there are still many needs and issues to be addressed in future research. Among these are the following:

  1. Research should be based in theory (e.g., general bereavement models and concepts, such as complicated grief, disenfranchised grief, or traumatic grief; coping; trauma and PTSD; stress and life change; attachment).
  2. Researchers should attempt to replicate important findings with different populations and settings.
  3. Researchers should utilize better sampling methods and include larger numbers of survivors in their study samples.
  4. Researchers should develop and use better measures of general and specific aspects of bereavement (standardized instruments with desirable psychometric properties of reliability and validity determined).
  5. Researchers should undertake systematic study of the entire range of kinship relations (both familial and others, including therapists), using, whenever possible, similar methods and measures.
  6. Studies should include measures of the emotional closeness or even the strength of the bond or attachment between the survivor and the suicide (i.e., not all individuals, even within specific kinship relations, are as close to or as affected by the loss; see, e.g., Reed, 1993).
  7. Researchers should use longitudinal (“follow-up,” rather than the more common retrospective) research methods to determine the time course of bereavement, including potentially critical periods in the bereavement process. There is a particularly great need for studies that begin early in the bereavement process (respecting and recognizing this sensitive and difficult time) and proceed over an extended period of months or even years.
  8. In addition to bereaved comparison groups among other modes of death, studies should include nonbereaved control groups whenever possible, as well as participants who do not belong to the often-studied and readily available clinical, college student, and support group populations (i.e., findings based on these groups may or may not generalize to all survivors).
  9. Whenever possible, research designs should include control features for potentially confounding factors that may contribute to suicide survivor reactions and hamper clear interpretation of the primary factors under investigation (e.g., emotional closeness or attachment of the survivor to the deceased, time elapsed since the death, relationship of the survivor to the deceased, the sex and/or age of the survivor, and the sex and/or age of the deceased) (with respect to closeness, see, e.g., Reed 1993).
  10. Studies should attempt to determine the impacts of a broad range of general factors in addition to kinship relations, including, for example, “the age of the deceased; the age of the survivor at the time of the death; sex, racial/ethnic, socioeconomic differences; family vs. friend differences; cultural differences (including the effects as a result of grief and mourning practices, religion, rituals, attitudes toward suicide, etc.); support group attendees vs. nonattendees vs. therapy client differences; those who discover the body of the suicide; and differences in the survivor experience in the case of the various methods employed in the suicide (i.e., violent vs. nonviolent, such as firearms and hanging vs. drugs)” (McIntosh 1987b:271).
  11. Researchers should extend and continue studies of attitudes toward suicide survivors to determine what factors affect perceptions of and attitudes toward survivors. In a related but new direction for this research, Peterson, Luoma, and Dunne (2002) investigated survivors’ perceptions of the clinician who treated their loved one who died by suicide. The results of such studies have implications for treating the suicidal as well as for issues after a suicide. A particularly interesting set of findings relates to differences between family members who did or did not file legal actions against the treating therapist.

Postvention: Clinical, Therapeutic, and Support Issues

A number of issues remain to be resolved with respect to therapy and support for survivors of suicide.

The terminology and description of postvention. Just as terminology issues exist in the larger context of suicide survivors, terms used in postvention circumstances are also problematic. Several terms are currently employed to refer to group settings intended to aid suicide survivors. The use of consistent terminology from the larger bereavement literature would benefit communication and clarification of the operation of such suicide support groups. For instance, in the larger literature, groups that involve professionals are called mutual help support groups, whereas those conducted and run by survivors are referred to as mutual help groups. Clear designation of these mutual aid groups and their various forms is important, as is differentiation of these groups from traditional therapeutic methods. That is, mutual aid is not “therapy” in the traditional sense of the term as used by clinicians. However, it is clearly the case that mutual aid groups can have therapeutic effects (see, e.g., Dunne 1987). More formal professional communication of the variety of approaches being employed with suicide survivors is needed (e.g., therapy, support groups, self-help groups). Dissemination of the specific details of these approaches—in traditional therapy and support settings, in publications, and in presentations at professional meetings—is desirable; these details should include information about features and approaches that are beneficial as well as those that do not work well. This would facilitate the development of model approaches for survivors that may advance society’s ability to meet suicide survivors’ needs through the provision of services and support.

Evaluation and efficacy efforts. Perhaps the most crucial aspect of postvention approaches and their proliferation is the need for evaluation. Only a small number of formal efficacy, evaluation, or outcome studies have been conducted for suicide survivor groups and postvention approaches (for a review, see McIntosh 1999:171-72; also, for a recent study, see Constantino, Sekula, and Rubinstein 2001), but models for such research appear in the general literature on mutual aid groups (for some examples, see McIntosh 1987b:275). Both mutual aid groups and traditional therapy approaches for suicide survivors would benefit greatly from a large growth in such research as a formal aspect of providing services. Immediate as well as long-term follow-up measures are needed to demonstrate the effects of assistance to survivors. Crucial questions exist in postventive care that can be answered only through formal evaluation. For instance: What approaches are most effective and helpful, and for which subgroups of survivors? Under what circumstances would therapy as opposed to support be indicated? When in the grieving process is therapy or support most helpful? How soon after the loss should a survivor enter into postvention care? Would all survivors benefit from therapy or support, or are both unnecessary for some survivors? Is the involvement of professionals essential in mutual help settings? If so, in what capacity should professionals be involved? Should professionals or facilitators who work with support groups also be survivors themselves? Without evidence from evaluation studies, postventive efforts cannot advance, nor can they gain acceptance within the larger mental health provider system.

Needs Related to Personal/Qualitative Accounts of Suicide Survivors

Variety in data sources. Although relatively controlled, traditional research investigations are an ideal embraced by the professional community, the use of multiple methodologies and sources of information is desirable. This is particularly true given that current levels of knowledge regarding suicide survivors are not well advanced. Varied sources may provide a wealth of information about the survivor experience and identify issues in need of closer and more controlled scrutiny. In the introduction to his seminal volume, Cain (1972a:24) suggested the use of such diverse data sources as autobiographies, poetry, court cases, and coroner’s anecdotes. Similarly, continued expansion of the personal accounts of suicide survivors, as described earlier, is needed as well. Conspicuously missing from the qualitative/personal experiential literature are clinical case studies of suicide survivors. Particularly desirable in this regard would be detailed case information on survivors who have experienced difficult or problematic grief. To date, the experiential literature (i.e., personal accounts) has almost exclusively been written by or about individuals who, although they have experienced bereavement process difficulties, rarely if ever could be considered “pathological” in their grief. Given that most suicide survivors seem to belong to this latter category, their personal accounts remain crucial to the body of knowledge. However, the clinical community requires cases across the entire range of experience to inform therapy and supportive interventions.

Descriptions of personal experiences in postventive settings. Few survivors have published accounts of their therapy/support group experiences and how these approaches benefited them. Eleanora Betsy Ross’s (1997) depiction of the Ray of Hope program she developed in the late 1970s following her husband’s suicide remains one of the few highly detailed program descriptions. It would be valuable to have survivors’ detailed accounts of their experiences in therapy and/or support groups, particularly to see how they would portray “the progression, changes, frustrations, lessons learned, and ‘stages’ of healing that likely resulted over time” (McIntosh 1999:159, 161).

Training and Education Needs

Education and training are logical areas to include when confronting a problem, and they seem appropriate prescription components in this context as well. However, virtually no empirical evidence has been collected that clearly demonstrates the effectiveness of training or education programs with respect to postvention. Such evaluative efforts are themselves essential in the design and implementation of educational programs. With evaluation as a crucial component, therefore, the following are some areas in which special efforts for training and education seem warranted.

Caregiver training. Caregivers and members of gatekeeper groups who frequently come into contact with suicide survivors (e.g., law enforcement personnel, emergency and medical personnel, coroners and medical examiners, funeral directors, clergy, teachers, mental health professionals) need training to sensitize them to the issues of suicide and survivors. Ideally, these individuals would receive such training as part of their professional curricula. In some cases this training would involve teams of professionals to help groups of survivors (e.g., students in a school in which a suicide has occurred). In some instances, caregiver training is also a recognition that persons in certain occupation groups (e.g., police officers, emergency personnel) are likely to become involved with suicidal individuals and so become survivors themselves. Thus training should include discussions of coping with one’s own survivorship, and postvention services should be readily available for these professionals.

Public education efforts. Although caregivers may encounter suicide survivors more frequently than do most members of the general public, education of the population at large also seems warranted. Such education should include the identification of available community resources as well as general information regarding mental health services and suicide.

The Aftermath of Suicide Attempts

Unlike the extensive (by comparison) literature that exists on the aftermath of death by suicide for surviving significant others, almost nothing has been written regarding the effects of attempted suicide on the lives it touches. The literature on unsuccessful suicide attempts (or “parasuicides”) is long-standing and extensive (e.g., Kreitmann 1977; Maris, Berman, and Silverman 2000:chap. 12). However, this large body of research has focused largely on therapy and intervention for the attempter following the nonfatal attempt as well as on the prevention and risk of future suicidal acts and death by suicide (e.g., Arensman et al. 2001; Stewart, Manion, and Davidson 2002; Suokas et al. 2001). Almost completely missing from consideration and study have been the variety of effects of an attempt on the life (and feelings, cognitions, and social world) of the attempter and his or her significant others. A small number of studies have examined the subjective experiences of parents of adolescent attempters (Wagner et al. 2000) and greater problems among friends of adolescent attempters (e.g., Hazell and Lewin 1993; Ho et al. 2000). Other than these few investigations, however, there is no body of research similar to the one that exists concerning suicide survivors on this topic, despite the substantial numbers of individuals involved with suicide attempt behaviors annually and cumulatively (as noted earlier). The pain, stress, emotional and social turmoil, shame and stigma, and other logically common aspects of life following a nonfatal suicide attempt may often be unaddressed or poorly addressed.

Unlike the now visible and growing personal accounts published by some suicide survivors, few discussions of such issues are available (an exception is Heckler 1994) for inquiring attempters or their loved ones seeking normalization and guidance in their struggle with an attempt’s sequelae. Efforts similar to those that assist suicide survivors, both professionally and therapeutically as well as with respect to self-help approaches, might well benefit attempters and their significant others as well. As is true for suicide survivors, given the large numbers of people affected by suicide attempts, this topic deserves concerted research efforts to determine the scope of the problem and what issues and factors are most important following suicide attempts. In these cases, however, we not only have “survivor-victims” in the loved ones for whom assistance in healing is needed, we also have the individuals whose behavioral acts created the situation. These attempters are, obviously, alive. Information about their subjective experiences and efforts to deal with their pain and “return to life” (Heckler 1994) is potentially valuable to other attempters as well. The significant others in the lives of suicide attempters would also benefit from information on the practical and subjective experiences of others who have dealt with and recovered from such attempts.

Conclusion

For many years, researchers ignored the topic of suicide’s aftermath. However, the literature on suicide survivors and our understanding of the experience of suicide survivorship is now growing annually. Still almost entirely ignored is the issue of the effects on attempters’ loved ones following nonfatal suicide attempts. More and better designed studies have begun to be conducted that will help us to quantify and place suicide grief in the larger context of bereavement from all modes of death. This research has been supplemented by highly detailed individual accounts that personalize suicide grief and portray survivors’ variety of experiences in ways that are not possible in group research finding reports. There is certainly a need for substantial further expansion of this body of knowledge in the areas of research, education, therapy/support, and postvention. The existing literature has raised awareness of this important aspect of suicidal behavior and has led to the establishment of therapy and support interventions. Long overdue, greater attention to the issues and individuals affected in the aftermath of suicide and suicidal behavior will further increase our understanding and, more important, assist survivors in their efforts to confront their loss, or the ramifications of a suicide attempt, and heal.