INTERNET-BASED TREATMENT FOR COMPLICATED GRIEF: CONCEPTS AND CASE STUDY BIRGIT WAGNER Department of Clinical Psychology and Psychotherapy, University of Trier, Trier, Germany CHRISTINE KNAEVELSRUD Centre for Victims of Torture, Berlin, Germany ANDREAS MAERCKER Department of Clinical Psychology and Psychotherapy, University of Trier, Trier, Germany The authors review current concepts of complicated grief and propose a new Internet-based cognitive-behavioral treatment program for complicated grief. They show how complicated grief is distinguished from disorders such as posttraumatic stress disorder, depression, and anxiety in the literature and explore the effective- ness of existing bereavement interventions. According to meta-analyses, conven- tional bereavement interventions show no overall benefit and may even have harmful effects if initiated too soon. For people with complicated grief symptoms or those affected by a death in traumatic circumstances, however, interventions have proved to reduce symptoms. Against this background, the authors introduce an Internet-based cognitive-behavioral treatment program for complicated grief that integrates established methods of psychotherapy with new technology. The intervention was conducted via e-mail and aimed exclusively at clients with com- plicated grief. The procedure and course of this new approach are illustrated in a case study. The treatment successfully reduced symptoms. Results and further implications of an Internet-based writing protocol are discussed. Grief is a natural, nonpathological phenomenon. It involves a continuing and changing process of adjustment to life without the deceased. Most bereaved people do not show pathological symptoms only a minority suffer from complicated grief (also called ������pathological������ or ������traumatic������ grief ). As such, grief usually Received 23 December 2004 accepted 18 January 2005. Address correspondence to Birgit Wagner, University of Trier, Department of Clinical Psychology and Psychotherapy, 54286 Trier, Germany. E-mail: wagnerb@uni-trier.de 409 Journal of Loss and Trauma, 10:409���432, 2005 Copyright # Taylor & Francis Inc. ISSN: 1532-5024 print/1532-5032 online DOI: 10.1080/15325020590956828

does not require counseling or specific bereavement treatment (Stroebe, Hansson, Stroebe, & Schut, 2001). In fact, results show that interventions initiated too soon after the loss can be harmful (Brom, Kleber, & Defares, 1989 Mawson, Marks, Ramm, & Stern, 1981 Murphy et al., 1998 Sireling, Cohen, & Marks, 1988). Therefore, several authors (Murphy et al., 1998 Neimeyer, 2000 Sireling et al., 1988) have concluded that interventions should only be provided for specific subgroups of bereaved individuals: (a) preventively for bereaved people lacking social support, (b) if the death occurred in traumatic circumstances, and (c) for those already showing a high level of symptoms, that is, persons suffering from complicated grief. Nevertheless, there are still many interventions aimed at people who are experiencing a normal grieving process rather than suffering from complicated grief. This is partly due to a lack of information about the effects of bereavement interventions, but also because there is no clear concept of complicated grief or clear distinction from the normal grief process. The Diagnostic and Stat- istical Manual of Mental Disorders (4th edition DSM-IV American Psychiatric Association [APA], 1994) and the International Classi- fication of Diseases (10th revision ICD-10 World Health Organiza- tion, 1992) still do not recognize complicated forms of prolonged and distorted mourning as a distinct diagnostic category. Complicated grief can be defined as a deviation from the (cul- tural) norm in the duration or intensity of the symptoms of grief (Stroebe et al., 2001). Several reactions of complicated grief are described in the literature, including feelings of guilt, self-blame, anxiety and depression-related symptoms, posttraumatic stress dis- orders (Schut, de Keijser, van den Bout, & Dijkhuis, 1991 Zisook, Schneider, & Shuchter, 1990 Znoj & Maercker, 2004), and physical health problems (Rogers & Reich, 1988 Zisook, Shuchter, Sledge, Paulus, & Judd, 1994). Bereaved individuals often suffer from biased thinking (Fleming & Robinson, 2001), especially when analyzing the circumstances of and reasons for the death (Sherman & McConnell, 1995). In the wake of a traumatic death, many bereaved individuals experience interpersonal and intrapersonal social difficulties (Dyregrov, 2004). Particularly if the loss is stigma- tized (e.g., suicide, child loss, murder), many bereaved people develop strong feelings of guilt and shame. The effects on social interaction are particularly marked in this subgroup���not only do 410 Wagner et al.

the bereaved individuals tend to retreat from social interaction (Lang, Gottlieb, & Amsel, 1996), contact with them is also often avoided, leading to further isolation and loneliness. Bereaved individuals often feel stigmatized or stigmatize themselves (Dunn & Morrish-Vidner, 1987 Jordan, 2001). One approach to diagnosing complicated grief is based on Horowitz���s stress response theory (Horowitz, 1993, 1997, 2001 Horowitz et al., 1997). Complicated grief is characterized by symp- toms of intrusion and avoidance. Sufferers fail to adapt completely to their loss and to the new situation. The processes observed are similar to those occurring in posttraumatic stress disorder (PTSD), acute stress disorder, and���as proposed by Maercker, Einsle, and Kollner �� (2004)���adjustment disorder. The Internet- based cognitive-behavioral treatment presented below is based on stress response theory. Distinctiveness From Other Disorders Complicated grief shows high comorbidity with major depression (Zisook et al., 1994) and PTSD (Melhem et al., 2001). In the past decade, however, consensus has emerged that although overlaps with these disorders do exist, complicated grief is a distinct com- plaint (Prigerson et al., 1995, 1996). Research has shown that major depressive disorder takes a different clinical course than compli- cated grief (Pasternak et al., 1991 Prigerson et al., 1997). Com- pared to individuals suffering from complicated grief, depressive patients have different EEG sleep profiles (McDermott et al., 1997) and distinct neuroendrocine responses (Jacobs, 1987). An event can be considered traumatic if it involves experienc- ing, witnessing, or confronting actual or threatened death, injury, or threat to the physical integrity of oneself or other people (DSM-IV text revision APA, 2000). Consequently, the death of a close person falls into the category of potentially traumatic events. With reference to these diagnostic criteria, complicated grief often has been subsumed under PTSD. However, it has become increas- ingly clear that complicated grief has its own set of symptoms that distinguish it from PTSD (Boelen, van den Bout, & de Keijser, 2003 Prigerson et al., 1995). Another major difference between PTSD and complicated grief is that the trauma in complicated grief normally seems to result from the pain of separation from the Internet-Based Treatment for Complicated Grief 411

deceased rather than from exposure to a horrific experience. Individuals suffering from complicated grief appear not to avoid reminders of threat, as individuals with PTSD are often observed to do. Instead, they tend to avoid reminders of the absence of the deceased through denial and dissociation (Prigerson & Jacobs, 2001). This distinct phenomenology of depression, PTSD, and complicated grief indicates that complicated grief can be seen as a clinical entity in its own right. Lichtenthal, Cruess, and Prigerson���s (2004) extensive review demonstrates that complicated grief consti- tutes a distinct psychopathological diagnostic entity and corroborates these findings. Two Recent Diagnostic Approaches In recent years, two notable research groups have proposed diagnostic criteria for distorted and prolonged grieving with the following terminology: traumatic grief disorder (Prigerson et al., 1999) and complicated grief disorder (Horowitz et al., 1997). Prigerson and Jacobs (2001) listed 10 symptoms of ������traumatic grief.������ These symptoms reflect the bereaved person���s feelings about the death of a significant other and can be classified into two categories: (a) symptoms of separation distress (e.g., preoccu- pation with thoughts of the deceased, longing and searching for the deceased, loneliness after the loss) and (b) symptoms of trau- matic distress (e.g., disbelief about the death, anger, feelings of shock). The symptoms should last at least 2 months for a diagnosis of traumatic grief disorder. Prigerson and Jacobs (2001) substituted the original term ������complicated grief������ with ������traumatic grief,������ considering the latter to be a more precise description of the two dimensions of the syndrome (i.e., trauma and separation stress). Moreover, the new term was assumed to sound less negative or stigmatizing than ������complicated grief������ (Jacobs, Mazure, & Prigerson, 2000). How- ever, the term traumatic grief might give the misleading impression that the death occurred in traumatic circumstances. In fact, this is not necessarily the case���the authors��� terminology describes the nature of the disorder itself rather than the objective circumstances of the bereavement (Jacobs et al., 2000). Parallel to Prigerson and colleagues, Horowitz et al. (Horowitz, 1993, 2001 Horowitz et al., 1997) have also published diagnostic 412 Wagner et al.

criteria for complicated grief disorder. As mentioned above, these criteria are based on stress response theory. According to Horowitz, complicated grief disorder has a generic relationship to PTSD, as both result from exposure to a stressful event. As such, Horowitz (1997) emphasized the intrusive and avoidance symptoms as well as the maladaptive behavior occurring in complicated grief. Traumatized persons oscillate between intrusive memories of the traumatic event and avoidance. Avoidance may prevent habituation to painful memories (Foa & Kozak, 1986) and interfere with the development of new perspectives without the deceased (Boelen et al., 2003). These processes are thought to be similar for individuals suffering from complicated grief. In their study with bereaved participants, Langner and Maercker (in press) confirmed that the stress response model also applies to complicated grief. Despite their different emphases on aspects such as avoidance, sleep disturbances, and duration (see Table 1), there is remarkable similarity in the symptoms proposed independently by the two research groups. Hopefully, agreement will soon be reached on the precise set of symptoms defining a grief disorder. Treatment Approaches to Grief Reactions or Complicated Grief Disorder Increasing numbers of grief treatments with various theoretical backgrounds have been developed over recent years. Most inter- ventions tend to use a combination of social support, stress- reduction, and psychodynamic approaches (Kato & Mann, 1999) combined with social activities and guided mourning. Moreover, several studies (Brom et al., 1989 Mawson et al., 1981 Murphy et al., 1998 Sireling et al., 1988) emphasize the importance of dis- tinguishing between normal and complicated grief processes, given that inappropriate therapeutic interventions seem to disturb the natural emotional processing and indigenous social support systems of the bereaved. Empirical Evaluations A number of meta-analyses and review studies (Forte, Hill, Pazder, & Feudtner, 2004 Kato & Mann, 1999 Litterer Allumbaugh & Hoyt, 1999 Neimeyer, 2000) describe the results of interventions Internet-Based Treatment for Complicated Grief 413