اثر غیر مستقیم حمایت اجتماعی بر آسیب شناسی روانی پس از حوادث از طریق خوددلسوزی

Abstract Following a traumatic event, external resources such as social support facilitate recovery. The mechanism underlying this relation is not well understood, however. Self-compassion is a positive coping strategy that has been negatively related to post-trauma psychopathology in prior work. It was hypothesized that the external resource of social support increased the internal resource of self-compassion, which resulted in decreased psychopathology. The current study tested the hypothesis that the association between social support and posttraumatic stress disorder (PTSD), generalized anxiety disorder (GAD), and depression symptoms had an indirect pathway via self-compassion. Using a community sample of individuals exposed to potentially traumatic events, social support was positively related to self-compassion. Self-compassion was negatively related to PTSD, GAD, and depression symptoms. Self-compassion mediated the relation between social support and PTSD, GAD, and depression symptoms. These results suggest that social support may reduce symptoms of PTSD, GAD, and depression through increased self-compassion in those who experienced a trauma.

Introduction Exposure to potentially traumatic events (PTEs) is associated with increased risk for psychopathology including posttraumatic stress disorder (PTSD; Zatzick et al., 2007), depression (Elhai, Contractor, Palmieri, Forbes, & Richardson, 2011), and generalized anxiety disorder (GAD; Grant, Beck, Marques, Palyo, & Clapp, 2008). Not all individuals exposed to a PTE develop psychopathology, however. Several models provide a theoretical framework for the different post-trauma outcomes. The emotion processing theory proposes that limited processing of the PTE results in elevated psychopathology (Foa & Kozak, 1986). Cognitive models suggest that maladaptive interpretations of PTEs results in a persistent belief that one is under threat. This sustained belief results in negative outcomes (Ehlers & Clark, 2000).The conservation of resources model proposes that psychological health is a function of the retention of resources, broadly defined as variables that an individual values, such as mastery of a skillset, self-esteem, and professional roles (Hobfoll, 1989). PTEs deplete resources, which results in psychopathology. Each of these models proposes that exposure to PTEs alters internal processes that increase vulnerability to psychopathology. Considerable evidence supports social support, an external process, as a protective factor against post-PTE psychopathology (Brewin et al., 2000, Neria et al., 2010 and Ozer et al., 2003). Defined as the provision of empathy and care by others (Ullman & Filipas, 2001), social support was negatively related to PTSD symptoms in combat veterans (Pietrzak, Johnson, Goldstein, Malley, & Southwick, 2009), natural disaster survivors (Feder et al., 2013), and women with histories of childhood abuse and intimate partner abuse (Schumm, Briggs-Phillips, & Hobfoll, 2006). Elevated social support was associated with reduced symptoms of PTSD, GAD, and depression for civilians in warzones (Mugisha et al., 2015 and Neria et al., 2010) and in samples of individuals exposed to a range of PTEs (Kwako et al., 2011 and Lian et al., 2014). The precise mechanism by which social support protects against pathology, however, is unknown. The quality of social support is presumed to influence internal perceptions of the individual, which in turn reduces psychopathology. That is, negative social interactions increase negative self-perceptions whereas positive support increases positive self-perceptions. Victims of interpersonal traumas who experienced negative social reactions (e.g. blaming the victim) reported greater shame and PTSD symptoms than those who had positive social interactions (Turner et al., 2013 and Ullman et al., 2007). Thus, poor social support may promote maladaptive internal perceptions about the event, which results in psychopathology. It is posited that improved social support would improve internal perceptions, which then reduces psychopathology. An internal resource that is a protective factor against psychopathology is self-compassion. Self-compassion is conceptualized as the experience of supporting oneself during difficult times, which closely mirrors the benefit of social support. Self-compassion involves three distinct elements: being kind and understanding towards oneself in times of difficulty, being mindfully aware of painful thoughts and feelings to prevent over-identification with problematic emotions, and seeing one's struggles as part of a broader human experience (Neff, 2004). Self-compassion is positively associated with life satisfaction, emotional intelligence, and social connectedness in healthy adults (Neff, 2004). Self-compassion was negatively related to depressive symptoms and anxiety symptoms in community (Neff, 2003) and clinical samples (Krieger, Altenstein, Baettig, Doerig, & Holtforth, 2013). Higher self-compassion is also associated with less social anxiety (Werner et al., 2012). Self-compassion may protect against post-PTE psychopathology. In a sample of children who witnessed a fire, those with high self-compassion showed lower depressive, suicidal, PTSD, and panic symptoms relative to those with low self-compassion (Zeller, Yuval, Nitzan-Assayag, & Bernstein, 2014). In a similar study, adults with a history of child maltreatment and low self-compassion had greater psychological distress, problematic alcohol use, and increased suicidality compared to those with high self-compassion (Tanaka, Wekerle, Schmuck, & Paglia-Boak, 2011). Disorder specific psychopathology was not assessed in this study, however. Self-compassion was negatively related to symptoms of avoidance but not re-experiencing, hyperarousal, or overall PTSD scores in a large sample of undergraduates with a history of PTE exposure (Thompson & Waltz, 2008). Finally, self-compassion was negatively related to PTSD symptoms after controlling for psychological inflexibility in a large undergraduate sample (Seligowski, Miron, & Orcutt, 2014). The kindness offered in positive social interactions mirrors the kindness offered to oneself in self-compassionate individuals. The emphasis of self-compassion on experiencing suffering as part of humanity relates to the feelings of inclusion and community imparted from social support. Using the emotion processing theory, social support may allow for processing of the event in a manner that promotes self-compassion and thus symptom reduction. Within the cognitive model, social support may present the individual with alternate interpretations of the event that promote self-compassion and reduce symptoms. Consistent with the conservation of resources model, social support is an external resource that may increase the internal resource of self-compassion to protect against psychopathology after a PTE. Thus, social support and self-compassion are hypothesized to be part of a broader resilience process. The present study examined self-compassion as a pathway by which social support is associated with psychopathology in a community sample of individuals exposed to a PTE. It was hypothesized that self-compassion was negatively related to PTSD, GAD, and depression symptoms in adults with a history of PTEs. Self-compassion was hypothesized to be positively associated with social support. Finally, it was hypothesized that self-compassion would account for a significant portion of the relation between social support and PTSD, GAD, and depression symptoms. Relevant covariates such as gender, age, racial/ethnic status, income level, and exposure to multiple types of trauma were included in the analysis. Additional relations concerning PTSD symptom clusters were explored as indicated by relevant prior literature.

3. Results Descriptive statistics and correlations are presented in Table 3. Using the recommended diagnostic cutoff of 38 for the PCL-5, 19.9% of the sample likely met criteria for current PTSD. Using the recommended cutoff of 10 on the GAD-7 and the PHQ-8, 26.2% of the sample met criteria for GAD and 32.9% met criteria for a major depressive episode. These estimates are consistent with other population estimates of psychopathology in those with exposure to multiple PTEs (Zatzick et al., 2007), yet are higher than what is observed in the general population (Kessler et al., 1995 and Kilpatrick et al., 2013). Comorbidity was common with 11.7% of the sample likely having a single disorder, 13.9% of the sample likely having two disorders, and 13.2% of the sample likely having three disorders. Analyses were conducted with the total sample and then replicated with two randomly selected sub-samples to determine the reliability of the observed relations. The findings were replicated in each sample and thus the results from the total sample are presented. Table 3. Descriptive statistics and bivariate correlations for study variables. 1 2 3 4 5 6 7 8 9 1. SC – 2. PTSD −.37⁎⁎ – 3. GAD −.54⁎⁎ .66⁎⁎ – 4. Depression −.57⁎⁎ .68⁎⁎ .84⁎⁎ – 5. SS .36⁎⁎ −.30⁎⁎ −.32⁎⁎ .56⁎⁎ – 6. PTSD-B −.27⁎⁎ .91⁎⁎ .57⁎⁎ .46⁎⁎ −.22⁎⁎ – 7. PTSD-C −.25⁎⁎ .79⁎⁎ .46⁎⁎ .45⁎⁎ −.16⁎⁎ .75⁎⁎ – 8. PTSD-D −.39⁎⁎ .94⁎⁎ .61⁎⁎ .66⁎⁎ −.33⁎⁎ .78⁎⁎ .66⁎⁎ – 9. PTSD-E −.38⁎⁎ .91⁎⁎ .67⁎⁎ .67⁎⁎ −.29⁎⁎ .74⁎⁎ .61⁎⁎ .82⁎⁎ – M 37.14 21.24 6.36 7.10 62.09 5.27 2.89 6.80 5.71 SD (9.99) (8.91) (5.54) (5.88) (16.42) (5.08) (2.58) (7.15) (5.80) Note: SC = self-compassion as measured by the self-compassion scale. PTSD = PTSD symptoms as measured by the PCL-5. GAD = generalized anxiety disorder symptoms as measured by the GAD-7. Depression = major depressive disorder symptoms as measured by the PHQ-8. SS = Social support as measured by the MSPSS. PTSD-B = PTSD re-experiencing symptoms. PTSD-C = PTSD avoidance symptoms. PTSD-D = PTSD numbing symptoms. PTSD-E = PTSD hyperarousal symptoms. ⁎ p < .05. ⁎⁎ p < .01. Table options Controlling for age, gender, income, and exposure to multiple types of traumas, self-compassion was negatively related to PTSD symptoms (b = − 0.65, p < .001). Controlling for the same covariates, social support was positively related to self-compassion (b = 0.22, p < .001) and negatively related to PTSD symptoms (b = − 0.34, p < .001). Using a bootstrapped 99% confidence interval based on 5000 samples, there was a significant indirect effect of social support on PTSD symptoms via self-compassion (indirect effect = − 0.12; 99% CI: − 0.17 to − 0.09; Table 4). A Sobel test also supported the indirect effect of social support on PTSD symptoms via self-compassion (z = − 5.87, p < .001). These results suggest that self-compassion accounted for 38% of the total effect of social support on PTSD symptoms. Table 4. A, B, C′, and AB paths for models examining the association between social support, self-compassion, and psychopathology. Dependent variable 95% CI B SE LLCI ULCI Self-compassion SS (A-path) 0.22⁎⁎ 0.02 0.17 0.27 Age 0.16⁎⁎ 0.4 0.07 0.25 Gender − 1.25 0.79 − 3.28 0.78 Income 0.11 0.33 − 0.74 0.96 Trauma exposure − 0.37 0.33 − 1.21 0.48 PTSD SC (B-path) − 0.57⁎⁎ 0.07 − 0.71 − 0.42 SS (C′-path) − 0.22⁎⁎ 0.05 − 0.31 − 0.13 AB path − 0.12+ 0.02 − 0.17 − 0.09 Age − 0.18⁎⁎ 0.07 − 0.35 − 0.01 Gender 4.75⁎⁎ 1.45 1.01 8.49 Income 0.27 0.61 − 1.29 1.83 Trauma exposure 0.27 0.61 − 1.29 1.83 Depression SC (B-path) − 0.29⁎⁎ 0.02 − 0.33 − 0.25 SS (C′-path) − 0.08⁎⁎ 0.01 − 0.11 − 0.06 AB path − 0.06+ 0.01 − 0.08 − 0.05 Age − 0.03 0.02 − 0.07 0.02 Gender 0.79 0.40 − 0.24 1.82 Income − 0.39 0.17 − 0.82 0.04 Trauma exposure 0.37 0.17 − 0.53 0.80 GAD SC (B-path) − 0.27⁎⁎ 0.02 − 0.31 − 0.23 SS (C′-path) − 0.05⁎⁎ 0.12 − 0.07 − 0.02 AB path − 0.06+ 0.01 − 0.08 − 0.04 Age − 0.04 0.02 − 0.09 0.01 Gender 0.75 0.39 − 0.26 1.76 Income − 0.32 0.16 − 0.74 0.10 Trauma exposure 0.46⁎⁎ 0.16 0.04 0.88 Note: SC = self-compassion as measured by the self-compassion scale. PTSD = PTSD symptoms as measured by the PCL-5. GAD = generalized anxiety disorder symptoms as measured by the GAD-7. Depression = major depressive disorder symptoms as measured by the PHQ-8. SS = Social support as measured by the MSPSS. PTSD-B = PTSD re-experiencing symptoms. PTSD-C = PTSD avoidance symptoms. PTSD-D = PTSD numbing symptoms. PTSD-E = PTSD hyperarousal symptoms. Trauma exposure refers to the number of types of trauma to which an individual was exposed. Alpha level set at 0.016 for all analyses. + AB path significance determined by the exclusion of 0 in the 99% CI. ⁎⁎ p < .01. Table options Controlling for age, gender, income, and exposure to multiple types of traumas, self-compassion was negatively related to GAD symptoms (b = − 0.29, p < .001). Controlling for the same covariates, social support was negatively related to GAD symptoms (b = − 0.10, p < .001). The total effect of social support on GAD was b = − 0.11, p < .001. Using a bootstrapped 99% confidence interval based on 5000 samples, there was a significant indirect effect of social support on GAD symptoms via self-compassion (indirect effect = − 0.06; 99% CI: −.08 to − 0.04). A Sobel (1982) test also supported the indirect effect of social support on GAD symptoms via self-compassion (z = − 7.68, p < .001). These results suggest that self-compassion accounted for 54% of the total effect of social support on GAD symptoms. Controlling for age, gender, income, and exposure to multiple types of traumas self-compassion was negatively related to depression symptoms (b = − 0.33, p < .001). Controlling for the same covariates, social support was negatively related to depression symptoms (b = − 0.13, p < .001). The total effect of social support on depression symptoms was b = − 0.14, p < .001. Using a bootstrapped 99% confidence interval based on 5000 samples, there was a significant indirect effect of social support on depression symptoms via self-compassion (indirect effect = − 0.06; 99% CI: − 0.08 to − 0.05). A Sobel (1982) test also supported the indirect effect of social support on depression symptoms via self-compassion (z = − 7.71, p < .001). These results suggest that self-compassion accounted for 43% of the total effect of social support on depression symptoms. The associations between DSM 5 PTSD clusters and self-compassion were then examined controlling for age, gender, income, and exposure to multiple types of traumas (Table 5). Self-compassion was significantly negatively associated with re-experiencing (b = − 0.13, p < .001), avoidance (b = − 0.06, p < .001), numbing (b = − 0.27, p < .001), and hyperarousal (b = − 0.21, p < .001). Using a bootstrapped 99% confidence interval based on 5000 samples, self-compassion was tested as a mediator between social support and each of the PTSD symptom clusters. There was a significant indirect effect of social support on reexperiencing symptoms via self-compassion (indirect effect = − 0.02; 99% CI: − 0.04 to − 0.01), avoidance (indirect effect = − 0.01; 99% CI: − 0.02 to − 0.01), numbing (indirect effect = − 0.05; 99% CI: − 0.07 to − 0.03), and hyperarousal (indirect effect = − 0.04; 99% CI: − 0.05 to − 0.03). Table 5. A, B, C′, and AB paths examining the relation between PTSD symptom clusters, self-compassion, and social support. Dependent variable 99% CI B SE LLCI ULCI PTSD-B SC (B-path) − 0.11⁎⁎ 0.02 − 0.15 − 0.06 SS (C′-path) − 0.05⁎⁎ 0.01 − 0.07 − 0.02 AB path − 0.02+ 0.01 − 0.04 − 0.01 Age − 0.06⁎⁎ 0.02 − 0.10 −.01 Gender 1.65⁎⁎ 0.41 0.58 2.71 Income 0.19 0.17 − 0.26 0.63 Trauma exposure 0.34 0.17 − 0.11 0.78 PTSD-C SC (B-path) − 0.06⁎⁎ 0.01 − 0.08 − 0.03 SS (C′-path) − 0.01 − 0.01 − 0.03 0.01 AB path − 0.01+ 0.002 − 0.02 − 0.01 Age − 0.01 0.01 − 0.03 0.02 Gender 0.88⁎⁎ 0.21 0.34 1.43 Income 0.01 0.09 − 0.23 0.23 Trauma exposure 0.23⁎⁎ 0.09 0.01 0.46 PTSD-D SC (B-path) − 0.22⁎⁎ 0.03 − 0.28 − 0.17 SS (C′-path) − 0.10⁎⁎ 0.02 − 0.13 − 0.06 AB Path − 0.05+ 0.01 − 0.07 − 0.03 Age − 0.06 0.03 − 0.12 0.01 Gender 1.31 0.55 − 0.11 2.74 Income 0.07 0.23 − 0.53 0.67 Trauma exposure 0.43 0.23 − 0.16 1.02 PTSD-E SC (B-path) − 0.18⁎⁎ 0.02 − 0.23 − 0.14 SS (C′-path) − 0.06⁎⁎ 0.01 − 0.09 − 0.04 AB path − 0.04+ 0.01 − 0.05 − 0.03 Age − 0.06⁎⁎ 0.02 − 0.11 − 0.03 Gender 0.80 0.45 − 0.37 1.97 Income − 0.04 0.19 − 0.52 0.45 Trauma exposure 0.44 0.19 − 0.05 0.92 Note: SC = self-compassion as measured by the self-compassion scale. PTSD = PTSD symptoms as measured by the PCL-5. SS = Social support as measured by the MSPSS. PTSD-B = PTSD re-experiencing symptoms. PTSD-C = PTSD avoidance symptoms. PTSD-D = PTSD numbing symptoms. PTSD-E = PTSD hyperarousal symptoms. Trauma exposure refers to the number of types of trauma to which an individual was exposed. Alpha level set at 0.016 for all analyses. + AB path significance determined by the exclusion of 0 in the 99% CI. ⁎⁎ p < .01.