شیوع، همبودی و خطرات ناشی از اختلال سوگ و داغداری طولانی مدت در میان بزرگسالان داغدیده چینی

Abstract Few epidemiological studies have investigated prolonged grief disorder (PGD) in the general population of Asian countries, including China. The aim of this study was to explore the rates and risks of PGD, and the association between PGD, post-traumatic stress disorder (PTSD), depression and anxiety in bereaved Chinese adults. The PG-13, PTSD Checklist-Civilian Version (PCL-C), Zung Self-Rating Depression Scale (SDS) and Zung Self-Rating Anxiety Scale (SAS) were administered to 445 subjects. Prevalence within the general population of China was 1.8% (i.e., 8/445). Among the eight subjects who met the PGD diagnosis, 75%, 87.5% and 75% scored above the cut-off point on the PCL-C, SDS and SAS, respectively, although a portion remained free from comorbidity. ANOVA, correlation analysis and stepwise multiple regression analysis demonstrated that kinship to deceased, age of the deceased, religion belief and cause of death were predictive of prolonged grief. A small proportion of bereaved persons may exhibit PGD. There is a substantial but far from complete overlap between PGD and the other three diagnoses. Bereaved parents and the widowed have high risk of PGD. These findings highlight the need for prevention, diagnosis and treatment for PGD patients.

Introduction The loss of a loved one is one of the most common adverse life events. Although the death of a significant other can be a painful experience, the majority of adults recover over time. However, a portion continue to grieve for an extended period of time and begin to exhibit symptoms of a state known as prolonged grief disorder (PGD) (Prigerson et al., 2009 and Morina et al., 2010), which is a disorder defined by Prigerson et al. (2009) who have developed and empirically tested consensus and diagnostic criteria for PGD. These symptoms were also variously labeled pathological grief (Horowitz et al., 1980), traumatic grief (TG) (Prigerson et al., 1999), and complicated grief (CG) (Kersting and Wagner, 2012). PGD results in substantial distress and impairment, worsens one׳s quality of life, and has been linked to excess medical morbidity and suicide rates (Stroebe et al., 2007). Recently, PGD was proposed as a new category for the International Classification of Diseases-11 (ICD-11) (Maercker et al., 2013). Most studies on the epidemiology of PGD were conducted in developed countries (e.g., America, Germany and the Netherlands) and among special population groups (e.g., older adults, widows, orphans, and caregivers of patients in a vegetative state); the prevalence of PGD in these studies ranged from 2.4% to 38.3% (Momartin et al., 2004, Goldsmith et al., 2008, Kersting et al., 2011, Newson et al., 2011, Guarnerio et al., 2012 and Schaal et al., 2012). However, little attention has been focused on a general population sample in a non-western culture. To our knowledge, no studies have been conducted in China, which has the largest population in the world and a unique Eastern culture. The study in China would not only permit cross-cultural comparisons but also greatly improve our understanding of PGD. Whether PGD should be considered a separate diagnostic entity has been the subject of controversy in the last decade (Schaal et al., 2012), as bereaved persons often present with symptoms of depression, posttraumatic stress disorder (PTSD) and anxiety (Newson et al., 2011). Previous studies have documented a substantial association among PGD, PTSD and depression (Boelen and van den Bout, 2005 and Morina et al., 2010). For example, moderate to high correlations were found between PGD and depression/PTSD (Boelen and van den Bout, 2005, Boelen and Prigerson, 2007 and Schaal et al., 2012), while a low to moderate correlation was found between PGD and anxiety symptoms (Boelen and van den Bout, 2005). Although PGD shares some overlapping symptoms with depression, anxiety and PTSD, it largely exhibits distinct symptoms (Bonanno et al., 2007). Compared to PTSD, depression and anxiety, PGD is characterized by an intense yearning for a deceased individual (Prigerson et al., 2009). A variety of studies have indicated that yearning loads highly on the grief factor, but not on depression or anxiety factors (Prigerson et al., 2009). The majority of people with PGD did not meet the criteria of major depression, PTSD or anxiety (Schaal et al., 2012). However, little is known about the association and comorbidity among PGD, PTSD and depression in China. A large number of studies have investigated the influence of various socio-demographic and bereavement-related factors (e.g., gender, income, relationship to the deceased, and cause of death) on prolonged grief symptom severity, but with mixed results (Prigerson et al., 2002 and Neria et al., 2007). In terms of demographic variables, the relation between PGD and gender/age has been inconsistently reported. Some studies have found that gender (Neria et al., 2007) and age (Prigerson et al., 2002) are predictors for the development of grief reactions, whereas other authors have documented no associations between grief and gender (Momartin et al., 2004 and Boelen and van den Bout, 2005) or age (Momartin et al., 2004). While a considerable number of studies have investigated time since loss as a potential predictor for grief symptom level; however, some studies have found no significant association between time since death and the severity of prolonged grief symptoms (Prigerson et al., 2002 and Schaal et al., 2010). Using Prigerson׳s criteria defining PGD (Prigerson et al., 2009) in bereaved Chinese adults, the aim of this study was to (1) explore the prevalence rates of PGD; (2) examine PGD and its relationship to symptoms of PTSD, depression and anxiety and report comorbidity among participants who met PGD criteria; and (3) identify the risk factors associated with the development of PGD.

Results 3.1. Reactions and prevalence of prolonged grief The mean of the prolonged grief-score of the sample was 17.79±7.45, and the range was 11–55. Fig. 1 reports the frequency of the PGD symptoms for the sample. The separation distress frequency is higher than the other symptoms (Fig. 1). Percentage of PGD symptoms in bereaved adults (n=445) according to the PG-13. Fig. 1. Percentage of PGD symptoms in bereaved adults (n=445) according to the PG-13. Figure options According to Prigerson׳s criteria of PGD, eight persons met the criteria for a diagnosis of PGD (Table 2). The point prevalence of PGD is 1.80% among the sample. Table 2. The characteristics of eight PGD patients. I.D. Age1 Gender Religion Education Marital Kinship Time Age2 Cause 1 27 F No Undergraduate Single Other 2.33 28 Medical 2 58 M No Middle school Married Child 1.17 25 Traumatic 3 27 F Buddhism Undergraduate Single Other 5.08 70 Traumatic 4 49 F NO Undergraduate D/W Child 6.75 19 Medical 5 32 F No Undergraduate D/W Spouse 1.58 30 Medical 6 58 F NO Undergraduate Married Child 3.92 25 Medical 7 37 F Christianity Undergraduate Married Spouse 0.58 42 Traumatic 8 47 F Buddhism Postgraduate Married Parents 0.58 86 Traumatic Note: Age1=age of the participant; F=female; M=male; Religion=religious belief; Education=educational background; Marital=marital status; D/W=divorced/widowed; Kinship=kinship to the deceased; Time=time since loss (years); Age2=age of the deceased; Cause=cause of death; Traumatic=traumatic (suicide, accident, etc.). Table options 3.2. Comorbidity of PGD, PTSD, depression and anxiety Of the subjects, 6.50% (n=29) had a PCL-C score ≥50; 38.2% (n=170) had an SDS score ≥41; and 20% (n=89) had an SAS score ≥41. Among the eight subjects who met the PGD diagnosis, six (i.e., 75%), seven (i.e., 87.5%), six (i.e., 75%) scored above the cut-off point for PCL-C, SDS and SAS, respectively. The co-occurrence of PTSD, depression and anxiety symptoms accounted for 75% of the eight PGD individuals. The correlation coefficients between PG-13 and PCL-C, SDS, and SAS were 0.71, 0.43, and 0.53, respectively (p<0.05). 3.3. Risk factors of PGD PG-13 scores were significantly related to the participant׳s age (r=0.112, p<0.05), time since loss (r=−0.103, p<0.05), and age of the deceased (r=−0.331, p<0.01). The variables marital status, religious belief, education background, kinship to the deceased, and cause of death had a significant effect on the severity of the prolonged grief symptoms ( Table 3). The divorced/widowed grief reactions were significantly higher than those of the single and married. The prolonged grief scores of those with Buddhist beliefs were significantly higher than the scores for people without religion. High school and postgraduate degrees were associated with higher prolonged grief scores than those with undergraduate degrees. Those who lost children had the most serious grief reactions. The PG-13 score of subjects whose loss was the result of traumatic causes was significantly higher than those whose loss was due to medical reasons ( Table 3). The variables gender and place of residence did not have a significant effect on prolonged grief symptom severity, and subjective family economic status was unrelated to the level of PG symptoms (r=−0.038, p>0.05). Table 3. ANOVA of PG-13 score of risk factors. Variables M (S.D.) d.f. F η2 Variables M (S.D.) d.f. F η2 Gender 1,433 0.106 0.000 Marital 2,441 9.74** 0.042 Male 17.48 (6.68)a Single 17.46 (6.33)b Female 17.77 (7.50)a Married 17.81 (8.49)b Place 2,436 0.443 0.002 D/W 27.36 (16.55)a Village 18.17 (6.70)a Kinship 5,437 26.21** 0.231 Town 17.22 (6.63)a Parents 18.49 (7.12)c Urban/city 17.78 (8.20)a Child 37.63 (15.93)a Religion 3,439 4.12* 0.027 Spouse 28.45 (15.74)b No 17.53 (6.90)b Sibling 21.15 (6.03)c Buddhism 23.94 (12.01)a Grandparents 15.93 (4.56)d Christianity 18.50 (7.78)ab Other 19.57 (8.41)c Islamism 19.58 (13.79)ab Cause 1,406 35.97** 0.081 Education 2,438 5.33* 0.024 Traumatic 23.35 (11.64)a Middle school 20.80 (9.65)a Medical 17.03 (6.33)b Undergraduate 17.12 (7.00)b Postgraduate 19.09 (7.34)a Note: Place=place of residence; Religion=religious belief; Education=educational background; Marital=marital status; D/W=divorced/widowed; Kinship=kinship to the deceased; Time=time since loss (years); Cause=cause of death; Traumatic=traumatic (suicide, accident, etc.). Values in the same column within the same risk factors followed by different superscript letters are significantly different based on least-squared difference multiple comparison. ⁎ p<0.05. ⁎⁎ p<0.01. Table options According to above ANOVA and correlation analysis results (Table 3), these variables (i.e. age of participants, no & Buddhism, middle school & undergraduate, postgraduate & undergraduate, single & divorced/widowed, married & divorced/widowed, time since loss, age of deceased, child & parents, spouse & parents, grandparents & parents, medical & traumatic) were entered into the stepwise regression model. The stepwise regression analysis identified child & parents, spouse & parents, age of deceased, no & Buddhism and medical & traumatic as significant predictor variables. The model explained 28.1% variance in PG symptoms (F (5, 378)=29.58, P<0.001). Child & parents had the most predictive utility, showing that the one who lost child had higher scores on PG symptoms than who lost parents. Spouse & parents emerged as the second potent predictor of PG symptoms, indicating that the one who lost spouse had more severe grief reactions than who lost parents. Age of deceased emerged as a statistically significant predictor of PG symptoms, with younger deceased associated with more severe grief reactions. No & Buddhism and medical & traumatic also significantly affect PG symptoms ( Table 4). Other variables, such as age of participants, time since loss, single & divorced/widowed, married & divorced/widowed, middle school & undergraduate, postgraduate & undergraduate, grandparents & parents did not contribute significantly to the PG symptom variance. Table 4. Stepwise regression analysis on PG symptom with socio-demographic and bereavement-related variables. These variables have significant effects on PG symptom according to ANOVA results (i.e. Table 3) or significantly correlated with PG symptom. Variables PG symptom score β t rpartial Child & parents 0.294 6.330*** 0.310 Spouse & parents 0.230 5.186*** 0.258 Age of deceased −0.171 −3.241*** −0.164 NO & Buddhism −0.107 −2.437** −0.124 Medical & traumatic −0.116 −2.288* −0.117 Note: β=standardized regression weight; rpartial=partial correlation coefficient; Child & parents and Spouse & parents are two dummy variables of kinship to the deceased with parents as a reference group. NO & Buddhism is a dummy variable of religious belief with Buddhism as a reference group. Medical & traumatic is a dummy variable of cause of death with traumatic as a reference group. ⁎ p<0.05. ⁎⁎ p<0.01. ⁎⁎⁎ p<0.001.