خوش بینی در غم و اندوه طولانی مدت و افسردگی بدنبال از دست دادن چیزی: مطالعه طولی سه موج

There is considerable evidence that optimism, the predisposition to have generalized favorable expectancies for the future, is associated with numerous desirable outcomes. Few studies have examined the association of optimism with emotional distress following the death of a loved one. Doing so is important, because optimism may be an important target for interventions for post-loss psychopathology. In the current study, we examined the degree to which optimism, assessed in the first year post-loss (Time 1, T1), was associated with symptom levels of prolonged grief and depression six months (Time 2, T2) and fifteen months (Time 3, T3) later, controlling for baseline symptoms and also taking into account positive automatic cognitions at T1. Findings showed that higher optimism at T1 was associated with lower concurrent prolonged grief and depression severity. Higher optimism at T1 was also inversely related with depression symptom severity at T2 and T3, but not prolonged grief severity at T2 and T3. Implications of these findings are discussed.

Optimism refers to “an individual difference variable that reflects the extent to which people hold generalized favorable expectancies for their future” (Carver et al., 2010). Optimism is considered a trait, relatively stable over time; yet, variations in optimism (over briefer and longer periods) have been documented and there is some evidence that optimism is amenable to change via cognitive behavioral interventions (Carver et al., 2010). There is considerable evidence that optimism has desirable consequences; it is associated with faster recovery from illness (Carver et al., 2003), lower mortality in old age (Giltay et al., 2004), and has protective effects following exposure to mild (Chang and Sanna, 2003) and severe (Britt et al., 2011 and Kivimäki et al., 2005) stressful life events. Few studies have explored the role of optimism in psychological functioning following the death of a loved one, most of them indicating that optimism has a protective impact. Specifically, Rogers et al. (2005) found that optimism was associated with constructive coping among people who lost a loved to HIV/AIDS. Ai et al. (2006) found personal loss in the 9/11 attacks to be associated with more severe posttraumatic stress disorder (PTSD) symptoms among individuals low in dispositional optimism, but not those who scored high on optimism. Harper et al. (2013) found optimism to be concurrently associated with less severe complicated grief reactions among parents who lost a child. Wagner et al. (2007) examined optimism as an outcome of online therapy for complicated grief; somewhat in contrast with the aforementioned findings, they did not find evidence that baseline optimism was associated with greater reduction in symptoms over time. In fact, unexpectedly, baseline optimism was significantly associated with a weaker reduction in symptoms of avoidance (and unrelated with other symptoms, including intrusive symptoms, depression, and generalized anxiety). None of these studies have investigated to what extent optimism is a prospective predictor of lower emotional distress following loss. Studying this issue is important because it enhances our understanding of underlying mechanisms of distress following loss and informs us about the potential usefulness of trying to enhance optimism in the treatment of disturbed grief. The present study—conducted in The Netherlands—used a prospective design to study the association of optimism with symptoms of Prolonged Grief Disorder (PGD) and depression, representing the two most prevalent and debilitating psychological syndromes that may occur following bereavement (Maercker et al., 2013). PGD—criteria of which were proposed and tested by Prigerson et al., 2009—is a clinical condition including persistent separation distress and difficulties accepting the loss and moving on without the lost person causing significant distress and disability, at least 6 months following the loss. PGD symptoms are distinct from, yet correlated with loss-related depression. Provisional epidemiological studies suggest that PGD occurs in 10–20% of bereaved individuals (Shear, 2015). PGD will likely be included in the forthcoming revision of the International Classification of Diseases and Related Health Problems (ICD-11) and resembles the condition “Persistent Complex Bereavement Disorder (PCBD)” included in the appendix of the fifth edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-5), as a condition requiring further research (Maercker et al., 2013). In a three-wave study, we examined the degree to which optimism, assessed in the first year post-loss (Time 1, T1), was associated with symptom levels of PGD and depression six months later (Time 2, T2), and fifteen months (Time 3, T3) later, controlling for baseline symptoms. At T1, we also assessed current positive automatic cognitions using the Automatic Thoughts Questionnaire-Positive version (ATQ-P; Ingram and Wisnicki 1988). The ATQ-P taps the frequency of self-referent positively valenced cognitions during the past week, reflecting state-like fluctuations in positive thinking (Dozois, 2007). We considered current positive automatic cognitions to elucidate the relative importance of optimism as trait-like positive future thinking versus more transient state-like positive thinking in predicting distress. Based on prior evidence that optimism buffers the impact of stressful events, we predicted that higher dispositional optimism in the first year of bereavement would be associated with lower PGD symptom severity and depression symptom severity, concurrently and prospectively.

3.1. Symptom severity scores Mean scores on the ICG were: T1, M=31.5 (S.D.=12.5), T2, M=27.0 (S.D.=12.4), and, T3, M=24.4 (S.D.=13.3). Scores declined significantly over time (F(2, 134)=33.61, p<0.001), with significant differences both between T1 and T2 (p<0.001) and between T2 and T3 (p<0.01). Mean scores on the SCL-depression scale were: T1, M=39.3 (S.D.=13.7), T2, M=35.2 (S.D.=12.8), and, T3, M=32.8 (S.D.=13.4). Scores declined significantly over time (F(2, 134)=33.61, p<0.001), with significant changes between T1 and T2 (p<0.001) and a trend toward a significant decline between T2 and T3 (p=0.07). At T1, T2, and T3, 67.4%, 46.5%, and 41.9%, of all participants, respectively, had a >25 score on the ICG which is indicative of PGD “caseness” ( Prigerson et al., 1995). 3 3.2. Variation in symptom severity as a function of demographic and loss-related variables We examined to what extent PGD and depression symptom severity scores at T1, T2, and T3 differed as a function of socio-demographic variables (i.e., age, gender, number of years of education) and loss-related variables (i.e., time since loss, relationship to the deceased, cause of loss). Because we wished to control for relevant background variables (the ones associated with symptom scores) in our subsequent regression analyses, we aimed to reduce Type II error and did not control for alpha inflation. Age was inversely associated with T1 PGD severity (r= −0.14, p<0.05) and T1 depression severity (r= −0.21, p<0.01) and positively associated with T2 depression severity (r=0.16, p<0.05), T3 depression severity (r=0.32, p<0.001) and T3 PGD severity (r=0.18, p<0.05). Number of years of education was inversely associated with T1 PGD severity (r= −0.15, p<0.05), T3 PGD severity (r= −0.18, p 0.05), and T3 depression severity (r= −0.26, p<0.01). T3 depression severity varied as a function of kinship (F(2, 134)=4.98, p<0.01) which was due to participants confronted with the loss of a partner/spouse having higher scores than participants who lost a relative other than a partner/spouse or child (p<0.01). PGD and depression severity scores at T1, T2 and T3 did not differ as a function of gender, cause of the loss, or time since loss. 3.3. Regression analyses predicting PGD symptom severity at T1, T2, and T3 We carried out three regression analyses predicting PGD scores at T1, T2, and T3, respectively. In each of these analyses, independent variables were included in distinct blocks, representing (i) relevant background/loss-related variables (i.e., the ones associated with symptoms at T1, T2, and T3); (ii) positive automatic cognitions, and (iii) optimism. In the regression predicting PGD scores at T2 and T3, baseline PGD severity was also controlled. Outcomes are summarized in Table 1; the first column shows the ΔR2 (and the second column the associated F-test) for each block when entered as a first block to the equation and thus represents the percentage of variance in the dependent variable explained by this block, when not taking into account the variance explained by the other variables in the equation. The third column shows the ΔR2 (and the fourth column the associated F-test) for each block of variables when entered as a last step to the equation and thus represents the percentage of variance in the dependent variable explained by this block, after controlling for the variance explained by the other independent variables in the equation. The fifth through seventh columns display the B, SE, and β of the independent variables when these were entered to the regression models simultaneously. Table 1. Summary of regression analyses predicting prolonged grief symptom severity. ΔR2 when entered as first block ΔF when entered as first block ΔR2 when entered as last block ΔF when entered as last block B in final model SE B in final model β in final model DV=prolonged grief at T1 Block 1: 0.058 6.93⁎⁎ 0.030 4.70⁎ Age −0.11 0.05 −0.13⁎ Years of education −0.63 0.24 −0.16⁎⁎ Block 2: 0.125 32.47⁎⁎⁎ 0.022 7.05⁎⁎ Positive automatic cognition −0.11 0.04 −0.18⁎⁎ Block 3: 0.241 72.03⁎⁎⁎ 0.106 33.53⁎⁎⁎ Optimism −0.88 0.15 −0.38⁎⁎⁎ DV=prolonged grief at T2 Block 1: 0.540 184.10⁎⁎⁎ 0.354 120.55⁎⁎⁎ PGD at T1 0.74 0.07 0.69⁎⁎⁎ Block 2: 0.076 12.84⁎⁎⁎ <0.001 <1 Positive automatic cognition −0.01 0.04 −0.02 Block 3: 0.177 33.71⁎⁎⁎ 0.004 10.31 Optimism −0.18 0.16 −0.07 DV=prolonged grief at T3 Block 1: 0.469 118.49⁎⁎⁎ 0.310 85.14⁎⁎⁎ PGD at T1 0.72 0.08 0.66⁎⁎⁎ Block 2: 0.047 3.29⁎ 0.054 7.45⁎⁎⁎ Age 0.22 0.06 0.24⁎⁎⁎ Years of education 0.06 0.28 0.02 Block 3: 0.072 10.32⁎⁎ 0.003 <1 Positive automatic cognition −0.04 0.05 −0.06 Block 4: 0.145 22.74⁎⁎⁎ 0.002 <1 Optimism −0.15 0.20 −0.05 ⁎ p<0.05. ⁎⁎ p<0.01. ⁎⁎⁎ p<0.001. Table options The regression predicting PGD scores at T1 yielded a significant model; R2=0.290, F(4, 228)=22.86, p<0.001. All three blocks of variables explained a unique proportion of variance in PGD scores when controlling for the other variables. The regression predicting PGD scores at T2 also yielded a significant model; R2=0.545; F(3, 158)=61.86, p<0.001. PGD scores at T1, positive automatic cognitions, and optimism explained significant variance in PGD scores at T2 when entered to the equation as first blocks; PGD severity at T1 was the only variable explaining variance in PGD scores at T2 when controlling for the other variables in the equation. Finally, the regression model predicting PGD scores at T3 was also significant R2=0.527; F(5, 135)=28.96; p<0.001. All blocks of variables explained variance in PGD scores at T3 when entered to the equation as first blocks; however, in the equation with all variables entered, PGD scores at T1 and the participant׳s age were the only variables explaining a unique proportion of variance in PGD scores at T3. 3.4. Regression analyses predicting depression symptom severity at T1, T2, and T3 We carried out three similar regression analyses now including depression scores at T1, T2, and T3 as dependent variables. Outcomes are summarized in Table 2. All three regression models were significant (T1-depression, R2=0.358; F(3, 229)=42.07; T2-depression, R2=0.546; F(4, 158)=46.35; T3-depression, R2=0.634; F(6, 136)=14.59; p<0.001). All blocks explained unique variance in depression severity at T1, T2, and T3 when entered as a first blocks to the equations. Most importantly, lower LOT-scores were concurrently and prospectively associated with higher depression scores, when controlling for the other variables in the equations. Specifically, as shown in the third column of Table 2, optimism explained 12.8% of the variance in depression severity at T1, beyond age (the only background variables associated with T1-depression) and positive automatic cognitions. Optimism predicted 1.9% of the variance in T2-depression and 1.8% in T3-depression beyond relevant background variables, positive automatic cognitions, and baseline depression. Table 2. Summary of regression analyses predicting depression symptom severity. ΔR2 when entered as first block ΔF when entered as first block ΔR2 when entered as last block ΔF when entered as last block B in final model SE B in final model β in final model DV=depression at T1 Block 1: 0.042 10.05⁎⁎ 0.018 6.36⁎ Age −0.13 0.05 −0.14⁎ Block 2: 0.214 62.02⁎⁎⁎ 0.046 16.13⁎⁎⁎ Positive automatic cognition −0.17 0.04 −0.25⁎⁎⁎ Block 3: 0.283 90.11⁎⁎⁎ 0.128 45.15⁎⁎⁎ Optimism −1.03 0.15 −0.41⁎⁎⁎ DV=depression at T2 Block 1: 0.454 130.77⁎⁎⁎ 0.263 89.28⁎⁎⁎ Depression at T1 0.58 0.06 0.59⁎⁎⁎ Block 2: 0.027 4.34⁎ 0.067 22.67⁎⁎⁎ Age 0.23 0.05 0.26⁎⁎⁎ Block 3: 0.119 21.17⁎⁎⁎ 0.007 2.29 Positive automatic cognition −0.06 0.04 −0.09 Block 4: 0.236 24.07⁎⁎⁎ 0.019 6.57⁎ Optimism −0.41 0.16 −0.16⁎ DV=depression at T3 Block 1: 0.240 42.65⁎⁎⁎ 0.130 28.25⁎⁎⁎ Depression at T1 0.43 0.08 0.42⁎⁎⁎ Block 2: 0.137 7.03⁎⁎⁎ 0.134 9.73⁎⁎⁎ Age 0.27 0.08 0.29⁎⁎ Years of education −0.48 0.31 −0.12 Deceased is partner/spouse 1.27 2.17 0.05 Block 3: 0.039 5.54⁎ 0.001 <1 Positive automatic cognition −0.03 0.05 −0.04 Block 4: 0.253 11.17⁎⁎⁎ 0.018 3.96⁎ Optimism −0.43 0.22 −0.16⁎ ⁎ p<0.05. ⁎⁎ p<0.01. ⁎⁎⁎ p<0.001.