رواسازی نسخه پرسشنامه ایتالیایی از سوگ پیچیده (ICG): مطالعه مقایسه بیماران CG در مقابل اختلال دو قطبی، اختلال استرس پس از حادثه و افراد سالم

Abstract Background A minority (9%–20%) of bereaved individuals experience symptoms of persistent intense grief associated with significant distress and impairment. This recently identified distinct post-loss syndrome has been variously named complicated grief, prolonged grief disorder, traumatic grief and persistent complex bereavement disorder. The Inventory of Complicated Grief (ICG) is a self-report instrument used to reliably identify this syndrome. We undertook a study to: 1) validate the Italian version of the ICG; 2) examine its performance in a clinical of bereaved individuals with complicated grief, post-traumatic stress disorder, bipolar disorder and healthy controls. Methods Study participants included 171 bereaved individuals clinically diagnosed with complicated grief (n=64); post-traumatic stress disorder (n=72); bipolar disorder (n=35) and 58 bereaved healthy controls. Assessments included the Structured Clinical Interview for DSM-IV Axis I disorders (SCID-I/P) and the Italian version of the ICG.

1. Introduction The loss of a loved one represents one of the most distressing life events that greatly affect physical and psycho-social well-being [1], [2], [3] and [4]. A number of studies have shown higher rates of disability and medication use in bereaved individuals compared to their non-bereaved counterparts, with the former being at a heightened risk of death [5], [6], [7], [8], [9], [10], [11] and [12]. This morbidity and mortality is likely related to acute grief, the initial response to bereavement, and its persistent form, variously designated as complicated, traumatic, prolonged or persistent complex grief. Most people adapt to loss without long term sequelae. However, a notable minority currently estimated between 7% and 20% of bereaved subjects, finds coping painful and difficult, and goes on to develop clinically disabling grief symptoms [13], [14], [15], [16], [17], [18], [19] and [20]. Within the last decades several studies have focused on the distinction between normal and “pathological” grief processes [[12], [16], [21] and [22]], attempting to establish the characteristics of “unresolved” grief [4], [12], [23], [24] and [25] and leading to the definition of a new specific disorder, variously named “complicated grief” (CG), “traumatic grief”, prolonged grief disorder, or persistent complex bereavement disorder [18], [23], [26], [27] and [28]. In the absence of consensus about the best designation for this syndrome, we use the term complicated grief in this paper. Increasing evidence underpins CG as characterized by a specific set of symptoms that differ from other possible pathological reactions that occur in the aftermath of a loss, such as Major Depressive Disorder (MDD) and Post-Traumatic Stress Disorder (PTSD) [18] and [29]. While evidence in support of a unique syndrome of complicated grief was clearly sufficient to establish independent diagnosis in the DSM-5, uncertainties still remain regarding the best diagnostic criteria, the time frame in which the diagnosis should be made and the name for the condition. Given the lack of consensus in these areas, the DSM-5 work-group collocated the condition in adjustment disorders section and in Section 3 that includes those conditions for which criteria are in need of further research as Persistent Complex Bereavement Disorder (PCBD) [4], [15], [18], [30], [31] and [32]. For this reason, the use of validated measures to assess CG in clinical samples would be decisive in order to distinguishing it from other disorders such as depressive disorders or PTSD. The Inventory of Complicated Grief (ICG), a 19-item questionnaire with good psychometric properties, has been used by a large number of studies to assess the presence of CG [24]. Criteria for identifying a clinically significant condition involving grief were unavailable until the development of the ICG [24]. The original study indicated an ICG total score of ≥25 as the top quartile and proposed this as a cut score for identifying CG [24]; however, in a treatment study funded by the National Institute of Mental Health (MH60783; MH70741; principal investigator MKS) we utilized an ICG score of 30 or higher [18] and demonstrated differential treatment response among individuals we identified as having CG. The ICG assesses the current presence of core CG symptoms that have been associated with impairment and poor outcomes [24] such as: intense yearning and preoccupation with the deceased; anger and bitterness about the death; shock and disbelief; enstrangement from others; hallucinations of the deceased; behavior change, including avoidance or proximity seeking behaviour [33]. Although still debated, most of these symptoms have been integrated in recently introduced criteria for PCBD in the DSM-5. A modified, shorter version of the original ICG is also available, named Inventory of Complicated Grief-Revised (ICG-R), consisting of 15 questions with a 5-point Likert-scale, a functional criterion and a duration criterion of six months [34]. Because ICG may be an outcome for international trials and studies on patients with these symptoms, there is an advantage in having it available in other languages including Italian. Therefore, we decided to validate the ICG for use in Italy, adopting the original version in order to have more informative data accordingly to the still on-going debate over the core symptoms of CG. This process we used entailed a formal translation as well as a cultural validation obtained by submitting the ICG questionnaire to a large group of Italian patients and condicting an assessment of its reliability and psychometric properties. The aim of the present study was to validate the reliability, factor structure and internal consistency of an Italian version of the ICG, besides providing additional empirical evidence of CG diagnosis through analysis of CG symptoms in a clinical data set consisting of bereaved individuals with CG, PTSD, bipolar disorder (BD) and healthy controls. In this same sample we also aimed at identifying the most discriminant ICG total score cut-off, and exploring possible underlying components of the same scale.

3. Results The sample included 229 subjects, 75 (32.8%) males and 154 (67.2%) females with a mean age of 51.16 ± 13.99 years. Demographic and clinical characteristics of the study samples are provided in Table 1. The diagnostic groups were homogeneous for the proportion of females and for age. The bereaved person was the spouse in 8.8% of cases, a child (son/daughter) in 8.4%, a parent in 35.3%, a sibling in 6.5%, a grandparent in 20.9%, a nephew/niece in 2.8%, another relative in 5.6%, and a friend/significant other in 11.6% of cases. The ANOVA analysis showed significantly different total ICG scores among diagnostic groups [F(3.228) = 94.19, p < .001]. Specifically, post-hoc Games–Howell pairwise comparisons indicated that the mean total ICG score was significantly lower in controls than in all other study groups, while it was significantly higher in patients with CG than in all the other three groups. No statistically significant difference emerged between BP and PTSD patients (see Table 2). Table 2. ANOVA overall group and Post-Hoc Games–Howell pairwise comparisons. Mean ± SD Significant pairwise differences (p < .05) CTL 5.43 ± 6.681 PTSD, CG, BP PTSD 17.76 ± 15.420 CTL, CG CG 40.64 ± 10.597 CTL, PTSD, BP BP 15.00 ± 13.204 CTL, CG ANOVA F-Test: F(3.228) = 94.19, p < .001. Table options The scale demonstrated a high level of internal consistency. Cronbach’s alpha value computed on the whole sample was 0.947; each item had a substantive correlation with the total and provided a relevant contribution to the scale because alpha decreased when each item in turn was deleted. The ROC analysis applied to the total ICG scores showed an AUC = 0.93 (p < .001) and that the best discriminant cut-off is 30 with a sensitivity of 82.8% and a specificity of 87.9% (see Fig. 1). ROC curve. Fig. 1. ROC curve. Figure options All ICG symptoms showed very high odds ratios with the indicator of having CG or not from a minimum of 3.25 (item 9) to a maximum of 26.00 (item 4) (see Table 3). Eigenvalues associated with the tetrachoric correlation matrix for the ICG in the full sample yielded three values larger than 1: 12.0, 1.9, and 1.3. The first large eigenvalue indicates 63% (12/19) of the variability in the data and can be explained with just one dimension, whereas 10% and 7% can be explained with the addition of two and three dimensions, respectively. The goodness of fit statistics indicated very good fit for the one-factor model (CFI and TLI = .99, RMSEA = .041) and all of the factor loadings are > 0.50. The more complex two- and three-factor models also have good fit (Table 4). The two-factor model separated out a factor representing hallucinations of the deceased with larger loadings on Q11, 14, and 15 but also cross loaded on Q9 about being hard to trust people. The three factor model similarly separated out the hallucination factor while the other two factors separated between one which included anger, bitterness, shock and disbelief, and another which identified yearning and preoccupation with the deceased but also exhibited several cross-loadings. Despite some interpretability of the additional factors the more parsimonious one-factor model was determined to be the best overall model. Table 3. Odds ratios and 95% C.I. of endorsing the ICG symptoms. ITEM SYMPTOMS ODDS RATIO 95% C.I. 1 I think about this person so much that it’s hard for me to do the things I normally do… 6.50 3.34–12.63 2 Memories of the person who died upset me 12.32 6.25–24.28 3 I feel I cannot accept the death of the person who died 17.17 8.43–34.97 4 I feel myself longing for the person who died 26.00 11.34–59.60 5 I feel drawn to the places and things associated with the person who died 11.43 5.78–22.61 6 I can’t help feeling angry about his/her death… 8.46 4.36–16.43 7 I feel disbelief over what happened 11.76 5.99–23.08 8 I feel stunned or dazed over what happened 12.22 6.16–24.27 9 Ever since he/she died it is hard for me to trust people… 3.25 1.39–7.57 10 Ever since he/she died I feel like I’ve lost the ability to care about people or I feel distant from people I care about… 7.42 2.71–20.33 11 I have pain in the same area of my body or have some of the same symptoms as the person who died… 12.46 3.38–45.87 12 I go out of my way to avoid reminders of the person who died… 4.07 1.56–10.66 13 I feel that life is empty without the person who died… 23.78 11.24–50.39 14 I hear the voice of the person who died speak to me… 6.63 1.66–26.51 15 I see the person who died stand before me… 4.16 1.13–15.28 16 I feel that is unfair that I should live when this person died… 8.96 3.07–26.10 17 I feel bitter over this person’s death… 9.47 4.90–18.29 18 I feel envious of others who have not lost someone close 6.12 2.85–13.18 19 I feel lonely a great deal of the time ever since he/she died… 13.70 6.63–28.34 Table options Table 4. Exploratory factor analysis results. 1-factor 2-factor 3-factor 1 1 2 1 2 3 1 I think about this person so much that it’s hard for me to do the things I normally do… 0.800⁎ 0.673⁎ 0.236 0.162 0.703⁎ 0.000 2 Memories of the person who died upset me 0.874⁎ 0.861⁎ 0.041 0.512⁎ 0.455⁎ −0.040 3 I feel I cannot accept the death of the person who died 0.891⁎ 0.962⁎ −0.103 0.809⁎ 0.156 0.021 4 I feel myself longing for the person who died 0.932⁎ 0.850⁎ 0.162 0.494⁎ 0.510⁎ 0.018 5 I feel drawn to the places and things associated with the person who died 0.772⁎ 0.584⁎ 0.338 0.320 0.405⁎ 0.280 6 I can’t help feeling angry about his/her death… 0.849⁎ 0.852⁎ 0.012 0.735⁎ 0.094 0.232 7 I feel disbelief over what happened 0.903⁎ 0.939⁎ −0.057 0.893⁎ −0.069 0.362 8 I feel stunned or dazed over what happened 0.937⁎ 0.951⁎ −0.007 0.736⁎ 0.176 0.258 9 Ever since he/she died it is hard for me to trust people… 0.696⁎ 0.441⁎ 0.436⁎ 0.436⁎ 0.085 0.530⁎ 10 Ever since he/she died I feel like I’ve lost the ability to care about people or I feel distant from people I care about… 0.815⁎ 0.677⁎ 0.252 0.095 0.819⁎ −0.099 11 I have pain in the same area of my body or have some of the same symptoms as the person who died… 0.673⁎ 0.243 0.677⁎ −0.021 0.471⁎ 0.545⁎ 12 I go out of my way to avoid reminders of the person who died… 0.503⁎ 0.748⁎ −0.406 1.051⁎ −0.487 −0.027 13 I feel that life is empty without the person who died… 0.980⁎ 0.821⁎ 0.295 0.336 0.717⁎ 0.007 14 I hear the voice of the person who died speak to me… 0.681⁎ 0.321 0.571⁎ 0.433⁎ −0.007 0.640⁎ 15 I see the person who died stand before me… 0.648⁎ −0.003 0.967⁎ 0.150 −0.001 1.022⁎ 16 I feel that is unfair that I should live when this person died… 0.798⁎ 0.590⁎ 0.361 −0.036 0.895⁎ −0.033 17 I feel bitter over this person’s death… 0.780⁎ 0.820⁎ −0.054 0.756⁎ 0.009 0.202 18 I feel envious of others who have not lost someone close 0.659⁎ 0.705⁎ −0.057 0.540⁎ 0.182 0.003 19 I feel lonely a great deal of the time ever since he/she died… 0.913⁎ 0.708⁎ 0.357 0.013 1.020⁎ −0.174 RMSEA 0.041 0.033 0.025 CFI/TLI 0.990/0.989 0.994/0.993 0.997/0.996 ⁎ Factor loadings >.40.