سوگ و داغداری پیچیده در پناهندگان بوسنیایی: ارتباط مبتلا به اختلال استرس پس از سانحه و افسردگی

Abstract Complicated grief is likely to be common among refugee populations exposed to war trauma. However, there have been few studies investigating the traumatic antecedents and correlates of complicated grief in refugees, and the relationship of that symptom pattern with other common disorders such as posttraumatic stress disorder (PTSD) and depression. We studied Bosnian refugees recruited from a community center in Sydney, Australia, with the sample being supplemented by a snowball method (N = 126; response rate, 86%). Measures included a trauma inventory, the Clinician Administered PTSD Scale (CAPS), the depression module of the Structured Clinical Interview (SCID), and the Core Bereavement Items (CBI). A dimension of traumatic loss derived from the trauma inventory was a specific predictor of complicated grief, with exposure to human rights violations being associated with images of the traumatic events surrounding the lost person. There was no link between PTSD and grief other than for a low-order association with the PTSD intrusion dimension. In contrast, depression was strongly associated with grief and its subscales. Only the subgroup with comorbid grief and depression reported higher levels of traumatic loss. The results suggest that complicated grief in refugees can become persistent and associated with depression. While PTSD and grief share common symptoms of intrusion, the two symptom domains are sufficiently distinct to warrant independent assessment of grief in refugee populations. COMPLICATED GRIEF is likely to be common among refugee populations exposed to war trauma. However, there have been few studies investigating the traumatic antecedents and correlates of complicated grief in refugees, and the relationship of that symptom pattern with other common disorders such as posttraumatic stress disorder (PTSD) and depression. We studied Bosnian refugees recruited from a community center in Sydney, Australia, with the sample being supplemented by a snowball method (N = 126; response rate, 86%). Measures included a trauma inventory, the Clinician Administered PTSD Scale (CAPS),1 the depression module of the Structured Clinical Interview (SCID),2 and the Core Bereavement Items (CBI).3 A dimension of traumatic loss derived from the trauma inventory was a specific predictor of complicated grief, with exposure to human rights violations being associated with images of the traumatic events surrounding the lost person. There was no link between PTSD and grief other than for a low-order association with the PTSD intrusion dimension. In contrast, depression was strongly associated with grief and its subscales. Only the subgroup with comorbid grief and depression reported higher levels of traumatic loss. The results suggest that complicated grief in refugees can become persistent and associated with depression. While PTSD and grief share common symptoms of intrusion, the two symptom domains are sufficiently distinct to warrant independent assessment of grief in refugee populations. The relationship between life-threatening trauma and psychiatric morbidity in refugees has been extensively studied in recent times.4, 5 and 6 In comparison, the role of traumatic loss in generating grief and other psychological disorders has attracted much less scientific study, a deficit in the literature given that ruptures to bonds are so pervasive among war-affected and forcibly displaced populations.4 The present study focused on the antecedents of complicated grief and its relationship to other common stress-related disorders such as PTSD and depression among Bosnian refugees resettled in Australia. War-affected persons from the former Yugoslavia comprise the largest group of refugees admitted to Australia in the 1990s.7 Several characteristics of this group facilitated study of complicated grief and its antecedents. Previous studies in other resettlement countries have documented the wide range of traumatic events, including losses suffered by Bosnian refugees,5 and 8 and our own professional experiences in working with this group in Australia since the early 1990s9 support the impression that many of these refugees were exposed to multiple traumatic deaths of family members. Deaths often occurred in horrific circumstances with the killings being witnessed by the surviving family,6 and 8 conditions that would be expected to lead to severe grief reactions.10 One limiting factor of research in the area has been the lack of consensus about the criteria for diagnosing abnormal or complicated grief. Considerable variability is evident in how grief-related reactions have been conceptualized and measured. Historically, Freud11 characterized normal grief as a response of painful dejection, loss of interest, and inhibition of activities. He emphasized the close link between unresolved grief and risk to depression. Subsequently, links between grief and a wider range of disorders including depression, anxiety, PTSD, and adjustment disorders have been identified.12 and 13 More recently, clearer formulations have been provided to distinguish complicated grief from other psychiatric disorders.14, 15, 16, 17, 18 and 19 Complicated grief has been described as the intensification of the normal bereavement reaction to the level where the person is overwhelmed, resorts to maladaptive behavior, or remains incessantly in a state of social dysfunction without progression of the mourning process towards completion.10 and 20 Features of complicated grief form a cohesive cluster that is predictive of future psychiatric morbidity.14, 16, 17, 18 and 21The afflicted person experiences persisting symptoms of separation distress such as yearning, longing, and searching for the deceased, as well as excessive loneliness; traumatic symptoms such as intrusive preoccupations and images associated with the deceased person; disbelief, severe pangs of emotion, denial, feelings of purposelessness and futility; and a sense of numbness or detachment. Other features include feeling persistently shocked, stunned, or dazed, the sense that life is empty and unfulfilling, a fragmented sense of security, and neglect of essential activities at work and at home.14 Yet, some of the conceptualizations applied, especially where terms such as “traumatic grief” are adopted,22 appear to blur the distinction between grief and other trauma-related disorders, particularly PTSD. The potential for overlap between traumatic grief and PTSD is particularly relevant to refugees whose experiences are complex and often involve life-threat as well as traumatic losses.4 Even in civilian settings, where most relevant studies have been undertaken, questions remain about the phenomenologic similarities of PTSD and complicated grief.15 and 23 More recent research has attempted to distinguish the two syndromes.14, 16, 17 and 18 Conceptually, PTSD is represented as being distinct from complicated grief in that the former arises from life-threatening events and the latter from loss of a loved one. In grief, the trauma is typically a separation trauma and therefore includes symptoms of separation distress such as yearning and searching, features not included in DSM-IV criteria for PTSD. In complicated grief, it is the absence of the deceased that is the source of the distress, rather than fears that the traumatic event will be re-experienced as in PTSD. Hypervigilance in grief is provoked by cues relating to the deceased rather than to fundamental insecurities associated with fears of re-experiencing past trauma. At the same time, experiences of life-threat and loss often occur concurrently in settings of civil warfare.4 For example, it was common for Bosnian survivors to have witnessed the violent deaths of families while being in mortal danger themselves.5 Hence, it seems possible that intrusive memories of these events might include characteristics of both PTSD-related images as well as aspects of grief. The relationship between grief and depression also requires clarification. Traumatic loss has been identified as a contributory factor to depression,8, 22 and 24but in refugees, the persistence of depression may be mediated by other more immediate postmigration stresses. 8For example, one study25 indicated that premigration trauma experiences (including death of family) were associated with PTSD,25 whereas ongoing postmigration stresses were more closely linked to depression. The present study aimed to examine the antecedents of grief and its relationship to PTSD and depression in Bosnian refugees. The analysis builds on previous reports based on the same dataset in which the traumatic antecedents of depression26 and PTSD27 were examined separately. In particular, we wished to examine whether the specific pattern of life-threat leading to PTSD yielded by an earlier analysis27 differed from that of grief, which, we postulated, would be closely and specifically linked with traumatic loss. We also wished to investigate for possible phenomenologic overlap of complicated grief with PTSD and depression, respectively.

Results Demographic characteristics The sample consisted of 77 women (61%) and 49 men (39%). The mean age was 47 (range, 18 to 88) years. Seventy-seven percent were married and 92% had at least one child. Eighty-seven percent had completed high school and 13% held a university degree. Sixty-two percent had been employed in factory or sales work in Bosnia and 14% were professionals. Ninety-six percent were unemployed in Australia and 74% spoke little or no English. Subjects originated from 25 different cities and villages across Bosnia-Herzegovina. Ninety-two percent of the subjects had resided in Australia for more than 3 years. The average time since exposure to the most severe traumatic experience (the anchor point for assessing PTSD) was 5 years (range, 2 to 7 years). All subjects were authorized refugees with permanent residency status in Australia. Principal components analysis In a previous analysis of the data,44 principal components analysis (PCA) was used to extract dimensions from the 31 trauma items. Four components emerged with Eigenvalues greater than 1.0, together accounting for 55% of the total variance. The four components each accounted for 31%, 10%, 8%, and 6% of the variance, respectively. The four components were: dimension 1—human rights violation, including incarceration in a concentration camp, being tortured by beating, burning and electric shock, or being forced to witness others being tortured; dimension 2—dispossession and eviction, which included being forced out of their city or home, being evicted from the country, losing all belongings, and being dismissed from jobs; dimension 3—threat to life, including experiences such as proximity to death, exposure to killing, continuous threat of harm, critical shortages of food and water, and absence of medical help when severely ill; dimension 4—traumatic loss of family, which included forced separations and/or witnessing the killing of a close family member. Table 1 provides a detailed description of these loss items. Table 1. Loss Experiences No. (%) Spouse killed 12 (8%) Parent killed 5 (4%) Sibling killed 22 (18%) Child killed 5 (4%) Secondary relatives killed 65 (52%) Friends/colleagues killed 31 (25%) Table options An intercorrelation matrix yielded low to moderate associations among the four trauma dimensions with the highest correlation being between human rights violation and threat to life (factors 3 and 1) (r= .36, P < .001), followed by threat to life and dispossession and eviction (factors 2 and 3) (r= .25, P < .01), and human rights violation and dispossession and eviction (factors 2 and 1) (r = .20, P < .05). Core bereavement items Thirty-one percent of the sample scored above the specified threshold for complicated grief reaction. In relation to the subscales, 39% of the sample experienced images and thoughts about the deceased, 18% experienced acute separation anxiety, and 26% experienced specific symptoms of grief. Pearson correlational analyses revealed that all three subscales were significantly associated with the total grief score (scale 1, r = .73, P< .001; scale 2, r = .80, P < .001; scale 3, r = .79, P < .001). There were no associations between grief and the demographic variables of age, gender, having children, employment status, education level and length of stay in Australia. Only widowhood caused by the war was associated with higher scores on the CBI (χ2 = 103.2, P < .01). Antecedents of grief Logistic regression analyses were conducted to assess the impact of the four trauma dimensions on complicated grief and its three subscales. The results are shown in Table 2. Table 2. Trauma Dimensions Associated With Complicated Grief and Its Three Subscales Odds Ratio Complicated Grief (n = 38) Scale 1: Images and Thoughts (n = 49) Scale 2: Acute Separation (n = 23) Scale 3: Grief (n = 32) Human rights violations 1.2 2.0∗ .99 1.2 Dispossession and eviction 1.2 1.1 1.4 .83 Threat to life .86 .89 1.5 1.4 Traumatic loss 2.6‡ 2.1∗ 6.0‡ 4.4† ∗ P < .01. † P < .001. Table options Traumatic loss was the most consistent predictor of complicated grief and all its subscales, yielding statistically significant odds ratios ranging from 2.1 to 6.0. The other three trauma dimensions did not produce significant associations with grief, with the exception of human rights violations, which yielded a significant association with subscale 1 (images and thoughts). PTSD and depression Rates of PTSD were 63%, with 75% reaching DSM-IV threshold for the intrusions domain, 51% for avoidance, and 58% for hyperarousal. A series of chi-square analyses were undertaken comparing grief and its three subscales with PTSD and its three subscales (re-experiencing, avoidance and hyperarousal) (Table 3). Table 3. Associations of Complicated Grief and Its Three Subscales With PTSD and Its Three Subscales, and With Depression Complicated Grief Images and Thoughts (scale 1) Acute Separation (scale 2) Grief (scale 3) PTSD (n = 79) .04 .14 .11 .11 Re-experiencing scale 1 .21∗ .22∗ .22∗ .25∗ Avoidance scale 2 .09 .13 .13 .14 Hyperarousal scale 3 .09 .17 .10 .04 Depression (n = 58) .43† .34† .30† .19∗ NOTE. Phi coefficient: ∗ P < .05. † P < .01. Table options The absence of an association between PTSD and grief was noteworthy, with the exception of a low order relationship (all correlations <0.25) between the PTSD intrusions subscale and all indices of grief. To facilitate statistical analysis, dysthymia and major depressive disorder were combined into a single depression category (n = 50, 40%). Complicated grief and depression were statistically related across all subscales of grief, but particularly the global grief index (r = 0.43, P < .01), (Table 3). In order to assess whether grief-related depression might form a distinct pattern, the sample was subcategorized into those with depression but no grief (n = 28) and those with depression with grief (n = 30). The group with grief alone (n = 8) was not large enough to allow inclusion in these analyses. Traumatic loss dimension was related to grief-depression (P < .01) but not to the subgroup with depression alone.