بررسی رابطه بین اختلال استرس پس از سانحه و خودآسیبی عمدی: نقش تعدیل کننده اختلال مرزی و اختلال شخصیت گریزی

Abstract Despite increasing evidence for an association between posttraumatic stress disorder (PTSD) and deliberate self-harm (DSH), few studies have examined the factors that moderate this association or the impact of co-occurring personality disorders among individuals with PTSD on DSH frequency. Given the high rates of co-occurrence between PTSD and two personality disorders of particular relevance to DSH, borderline personality disorder (BPD) and avoidant personality disorder (AVPD), this study examined the moderating role of these personality disorders in the association between PTSD and DSH frequency among a sample of substance use disorder patients (N=61). Patients completed structured clinical interviews assessing PTSD, BPD, and AVPD and a questionnaire assessing DSH. Results revealed more frequent DSH among patients with (vs. without) PTSD and provided evidence for the moderating role of AVPD in this association. Specifically, results revealed heightened levels of DSH only among PTSD patients with co-occurring AVPD. Findings are consistent with past research demonstrating that the presence of co-occurring AVPD among patients with other Axis I and II disorders is associated with worse outcomes, and highlight the importance of continuing to examine the moderating role of AVPD in the association between PTSD and a variety of health-risk behaviors.

1. Introduction Posttraumatic stress disorder (PTSD) is an anxiety disorder characterized by the development and persistence of re-experiencing, avoidant, and hyperarousal symptoms following direct or indirect exposure to a traumatic event (Blake et al., 1990). PTSD is a serious clinical concern, associated with considerable functional impairment (Kessler and Frank, 1997), high rates of co-occurring psychiatric disorders (Kessler et al., 1995), and heightened levels of numerous self-destructive and health-compromising behaviors, including suicide attempts (Nepon et al., 2010) and illicit substance abuse (Brady et al., 2004). One particularly clinically-relevant behavior receiving increasing attention among patients with PTSD is deliberate self-harm (DSH), defined as the deliberate, direct destruction of body tissue without conscious suicidal intent (Chapman et al., 2006). Indeed, rates of DSH among individuals with PTSD exceed 50% (Zlotnick et al., 1999, Sacks et al., 2008 and Dyer et al., 2009), and there is evidence to support the role of PTSD symptoms in the development and maintenance of this behavior (Harned et al., 2006 and Bornovalova et al., 2011). Despite evidence for elevated rates of DSH within PTSD, few studies have examined the subset of PTSD patients most at-risk for this behavior. In particular, little research has examined the moderating role of personality pathology in the association between PTSD and DSH or the specific co-occurring personality disorders that may increase the risk for DSH among individuals with PTSD. However, emerging research on the phenomenon of complex PTSD (a multifaceted syndrome encompassing both trauma-related symptoms and personality disturbance; Herman, 1992) suggests that the co-occurrence of PTSD and personality disorders may be especially relevant to DSH. Specifically, complex PTSD includes engagement in self-destructive behavior as a key feature (Herman, 1992 and Dyer et al., 2009) and has been found to evidence stronger associations with DSH than PTSD (Dyer et al., 2009). Thus, this literature highlights the importance of examining the intersection of trauma-related difficulties and personality disorders in the risk for DSH. Two personality disorders that warrant particular consideration in this regard are borderline personality disorder (BPD) and avoidant personality disorder (AVPD), both of which are common among patients with PTSD (Southwick et al., 1993 and Bollinger et al., 2000) and considered relevant to DSH. Indeed, BPD is the disorder most often associated with DSH (Chapman et al., 2006), with as many as 70–75% of individuals with BPD reporting a history of DSH (Gunderson, 2001). Moreover, emerging evidence highlights the relevance of AVPD to DSH, with studies finding an association between AVPD symptoms and DSH among both non-clinical young adult (Klonsky et al., 2003) and incarcerated adult (Haines et al., 1995) samples. The relevance of BPD and AVPD to DSH is further supported by theoretical literature on the pathogenesis of these disorders. Specifically, both BPD and AVPD are considered to be strongly linked to an intolerance of emotional distress and related difficulties regulating distress (Linehan, 1993 and Taylor et al., 2004)—two of the mechanisms implicated in the development and maintenance of DSH (Chapman et al., 2006 and Gratz et al., 2010). Furthermore, there is some evidence to suggest that the co-occurrence of these personality disorders among individuals with PTSD is associated with a number of negative outcomes and maladaptive behaviors (Heffernan and Cloitre, 2000 and Miller and Resick, 2007), including suicidal and other health-risk behaviors (Connor et al., 2002). Although no studies have examined the extent to which these disorders moderate the association between PTSD and DSH in particular, recent findings that BPD pathology moderates the association between DSH and emotional responding (Gratz et al., 2010) highlight the importance of examining the moderating role of personality disorders in the association between DSH and other factors. In considering the moderating roles of BPD and AVPD in the association between PTSD and DSH, one population that may be especially important to study is patients with substance use disorders (SUD). SUD patients have elevated rates of both PTSD and BPD compared to the general population (Trull et al., 2000 and Brady et al., 2004), and PTSD–SUD co-occurrence has been found to be associated with greater impairment and worse outcomes (Najavits et al., 1999 and Back et al., 2000). Furthermore, there is some evidence to suggest heightened risk for DSH among PTSD–SUD patients (compared to those with either disorder alone; Harned et al., 2006). Thus, the goal of the present study was to examine the moderating roles of BPD and AVPD in the association between PTSD and DSH frequency among SUD patients. We hypothesized significant main effects of PTSD, BPD, and AVPD diagnostic status on DSH frequency, such that the frequency of DSH would be higher among SUD patients with (vs. without) these disorders. Furthermore, we hypothesized significant interactions between PTSD and both BPD and AVPD, such that PTSD patients with co-occurring BPD or AVPD would report more frequent DSH than those with PTSD or these personality disorders alone.

3. Results 3.1. Variable transformations The raw scores for DSH frequency were highly positively skewed (4.32) and kurtotic (21.98), requiring transformation (Tabachnick and Fidell, 2007). Following logarithmic transformation, scores on this variable approximated a normal distribution (skewness=1.61, kurtosis=1.71). 3.2. Preliminary analyses Within the present sample, 30% (n=18) of participants reported a history of DSH, 1 49.2% (n=30) met criteria for current PTSD, 39.3% (n=24) met criteria for BPD, and 34.4% (n=21) met criteria for AVPD. A series of correlation analyses were conducted to explore associations between DSH frequency and demographic factors (i.e., age, gender, income, marital status [married vs. not], and employment status [unemployed vs. employed]), depression and anxiety symptom severity, past year substance use severity, and psychotropic medication use (presence vs. absence) in order to identify potential covariates for the analysis of covariance (ANCOVA). DSH frequency was significantly positively associated with depression symptom severity (r=0.30, p<0.05) and anxiety symptom severity (r=0.36, p<0.01), and differed significantly across gender (t[59]=2.01, p<0.05), with women reporting more frequent DSH than men. No other associations were significant (rs<0.22, ps>0.10). Given evidence of substantial overlap between depression and anxiety symptom severity (r=0.78) and high multicollinearity in the model with both included (as evidenced by squared multiple correlations [SMCs] of greater than 0.67 for both depression and anxiety; Tabachnick and Fidell, 2007), only gender and anxiety symptom severity were included as covariates in the primary ANCOVA (resulting in acceptable SMCs for all covariates). 2 3.3. Primary analyses Prior to examining the interactive effects of PTSD, BPD, and AVPD on DSH frequency, we conducted a series of one-tailed independent t-tests to examine differences in DSH frequency as a function of PTSD, BPD, and AVPD status. Power analyses revealed power of >0.60 to detect a medium-sized effect (d=0.50) for PTSD and BPD status and power of 0.58 to detect a medium-sized effect for AVPD status (with N=61 and αs=0.05). As expected, DSH frequency differed significantly across both PTSD (t[59]=−1.71, d=0.44, p<0.05) and AVPD (t[59]=−2.20, d=0.57, p<0.05) status, such that participants with (vs. without) these disorders reported more frequent DSH. Surprisingly, however, there were no significant differences in DSH frequency as a function of BPD status (despite some overlap between a BPD diagnosis and DSH, in the form of one BPD criterion encompassing a history of suicidal and/or DSH behavior; t[59]=−0.09, d=0.02, p>0.10). Next, to explore the interactive effect of PTSD, BPD, and AVPD on DSH frequency, we conducted a 2 (PTSD vs. no PTSD)×2 (BPD vs. no BPD)×2 (AVPD vs. no AVPD) ANCOVA (controlling for gender and anxiety symptom severity) with DSH frequency serving as the dependent variable (see Table 1). Power analyses revealed power of 0.48 to detect a medium-sized effect within this model (α=0.05). Although no significant main effects were found for PTSD, BPD, or AVPD status (when controlling for the other variables in the model), results did reveal a significant interaction between PTSD and AVPD (see Table 1 and Fig. 1). Post-hoc independent t-tests (using adjusted means) were conducted to explore the nature of the interaction. Results demonstrated that participants with co-occurring PTSD and AVPD reported significantly more frequent DSH than all other groups of participants, ts (d.f.s range from 19 to 36)>1.80, ps<0.05. No other interactions were significant. 3 Table 1. ANCOVA examining the interactive effect of PTSD, AVPD, and BPD on frequency of DSH. F (1, 51) p Partial η2 Observed power Gender 1.95 0.17 0.04 0.28 Anxiety symptom severity 6.08⁎ 0.02 0.11 0.68 PTSD 0.72 0.40 0.01 0.13 AVPD 3.12† 0.08 0.06 0.41 BPD 3.14† 0.08 0.06 0.41 PTSD×AVPD 4.49⁎ 0.04 0.08 0.55 PTSD×BPD 0.01 0.95 0.00 0.05 AVPD×BPD 0.28 0.60 0.01 0.08 PTSD×AVPD×BPD 0.60 0.44 0.01 0.12 ⁎ p<0.05. † p<0.10. Table options Interactive effect of PTSD and AVPD on lifetime DSH frequency. Fig. 1. Interactive effect of PTSD and AVPD on lifetime DSH frequency. Figure options Of note, findings remain the same when controlling for depression symptom severity (in addition to gender and anxiety symptom severity), as well as when controlling for anxiety symptom severity only, with only the PTSD×AVPD interaction emerging as significant in these analyses, F s>4.16, View the MathML sourceηp2s>0.07, ps<0.05.