ترومای دوران کودکی و عملکرد روانی فعلی در بزرگسالان مبتلا به اختلال اضطراب اجتماعی

Abstract Etiological models of social anxiety disorder (SAD) suggest that early childhood trauma contributes to the development of this disorder. However, surprisingly little is known about the link between different forms of childhood trauma and adult clinical symptoms in SAD. This study (1) compared levels of childhood trauma in adults with generalized SAD versus healthy controls (HCs), and (2) examined the relationship between specific types of childhood trauma and adult clinical symptoms in SAD. Participants were 102 individuals with generalized SAD and 30 HCs who completed measures of childhood trauma, social anxiety, trait anxiety, depression, and self-esteem. Compared to HCs, individuals with SAD reported greater childhood emotional abuse and emotional neglect. Within the SAD group, childhood emotional abuse and neglect, but not sexual abuse, physical abuse, or physical neglect, were associated with the severity of social anxiety, trait anxiety, depression, and self-esteem.

1. Introduction Social anxiety disorder (SAD) is a common (12.1% lifetime prevalence) (Kessler et al., 2005a and Kessler et al., 2005b) and often debilitating disorder (Lochner et al., 2003 and Schneier et al., 1994) that is characterized by persistent fear of social or performance situations in which an individual is at risk for embarrassment, humiliation, or possible scrutiny by unfamiliar persons (American Psychiatric Association, 2000). SAD affects more than 15 million American adults during any 12-month period (Kessler et al., 2005a and Kessler et al., 2005b). These statistics are particularly compelling in light of evidence suggesting that SAD is associated with significant distress and functional impairment in both work and social domains (Lochner et al., 2003, Rapee, 1995, Schneier et al., 1994 and Sherbourne et al., 2010). SAD may be a risk factor for other clinical disorders with which it commonly co-occurs, including major depression, substance abuse, and other anxiety disorders (Chou, 2009, Lampe et al., 2003, Matza et al., 2003, Ohayon and Schatzberg, 2010 and Randall et al., 2001). In addition, the number of feared social situations reported by individuals with SAD is associated with comorbid major depression, other anxiety disorders, and suicidal ideation (Gabalawy, Cox, Clara, & Mackenzie, 2010). Contemporary models of SAD suggest that the development of SAD is a result of a biological vulnerability coupled with negative social learning experiences (Clark and Wells, 1995, Heimberg et al., 2010 and Rapee and Heimberg, 1997). Findings from family studies demonstrate a strong association between social anxiety in parent and offspring (Bögels et al., 2001 and Lieb et al., 2000), and temperament studies suggest a link between inhibited temperament in childhood and the development of social anxiety in adolescence (Biederman et al., 2001 and Schwartz et al., 1999). The impact of social learning experiences has been suggested as a key environmental factor contributing to the development of the disorder (Rapee & Heimberg, 1997) and has received substantial empirical attention. Prospective studies have found that parental overprotection, rejection, and lack of warmth are associated with offspring SAD (Lieb et al., 2000, Knappe et al., 2009a and Knappe et al., 2009b). One specific social learning factor that has garnered recent attention is childhood trauma. Although extant data indicate that childhood trauma experiences may contribute to the development of SAD (Arrindell et al., 1983, Arrindell et al., 1989 and Bruch and Heimberg, 1994), how childhood trauma impacts adult clinical functioning in SAD remains unknown. The purpose of the current study was to address this gap by comparing histories of childhood trauma between individuals with SAD and non-clinical controls and examining associations between specific types of childhood trauma and adult clinical symptoms in SAD. 1.1. The role of early childhood trauma in the development of SAD Studies investigating childhood trauma in SAD suggest that parental emotional abuse towards the child (e.g., swearing, insulting, denigrating, and non-physical aggressing) and emotional neglect (e.g., emotional deprivation or the absence of feeling special, loved, or being part of a nurturing environment) may be important factors in the development of SAD. For example, in a non-clinical sample, compared to women with low levels of social anxiety, women high in social anxiety reported significantly more paternal rejection, paternal and maternal neglect, and paternal authority-discipline (Klonsky, Dutton, & Liebel, 1990). In another study, Lieb et al. (2000) found that parental rejection was associated with social anxiety in a community sample of adolescents. Studies of adults with SAD also indicate that these individuals report childhood experiences associated with emotional abuse and/or neglect. Individuals with SAD are more likely than controls to perceive their parents as having used shame as a form of discipline (Bruch & Heimberg, 1994). Two separate studies (Arrindell et al., 1983 and Arrindell et al., 1989) found that, compared to non-anxious healthy controls, patients with SAD characterized their parents as rejecting and lacking in emotional warmth, although these findings should be interpreted with caution because of small sample sizes. Simon et al. (2009) recently reported that 56% of individuals with SAD endorsed a history of childhood emotional abuse, and 39% of individuals with SAD endorsed a history of childhood emotional neglect. However, this study did not include a healthy control group, so it remains unclear whether these rates differ from those in non-clinical samples. In addition to emotional abuse and neglect, studies have investigated both (a) rates of SAD in adults with a history of childhood sexual abuse and (b) rates of childhood sexual abuse in adults with SAD. Using the former method, Pribor and Dinwiddie (1992) reported higher rates of SAD in adults with a history of sexual abuse than an age- and race-matched control group (SAD: 46.2% versus healthy controls: 2.8%). Employing the latter method, adults with SAD endorsed higher rates of childhood sexual abuse (SAD: 10.0% versus healthy controls: 5.0%) (Bandelow et al., 2004). However, when familial anxiety was controlled, childhood sexual abuse only predicted a diagnosis of SAD at a trend level. These findings suggest that childhood sexual abuse may not play a unique role in the development of SAD. 1.2. The psychological correlates of early childhood trauma Given that individuals with SAD appear to differ from healthy controls in their exposure to early childhood trauma, one important question is whether these early adverse experiences are correlated with one or more aspects of negative psychological functioning in SAD. In non-clinical samples, studies demonstrate associations between a history of childhood trauma and a number of negative adult experiences including elevated levels of depression, anxiety, substance use, suicidal behaviors, and emotional-behavioral problems (Briere and Elliott, 1994, Briere and Runtz, 1988 and Silverman et al., 1996). More recent studies have evaluated whether distinct forms of childhood trauma are related to specific psychological problems in adulthood. In a sample of young women, physical abuse was related to heightened aggression towards others and sexual abuse was related to maladaptive sexual behavior (e.g., getting into trouble because of sexual behavior, controlling others through the use of sex) (Briere & Runtz, 1990). By contrast, emotional abuse was related to low self-esteem. In a separate study, women who reported a history of emotional neglect reported greater problems in multiple domains (adult attachment styles, anxiety, depression, somatization, paranoia) than those reporting a history of physical abuse (Gauthier, Stollak, Messe, & Aronoff, 1996). Briere and Runtz (1988) found that maternal physical abuse was associated with interpersonal sensitivity and dissociation, whereas paternal emotional abuse was associated with anxiety, depression, interpersonal sensitivity, and dissociation. Gibb and colleagues found that childhood emotional abuse was more strongly related to diagnoses of depression or social anxiety disorder than either physical or sexual abuse (e.g., Gibb, Chelminiski, & Zimmerman, 2007). Despite mounting evidence for the associations between different forms of abuse and various adverse adult clinical symptoms, very few studies have extended this line of inquiry to SAD. Simon et al. (2009) recently examined the relationship between various types of childhood trauma and the severity of social anxiety, global severity of symptoms, disability, resilience, and quality of life in a sample of adults with SAD. A history of childhood emotional abuse or neglect was associated with greater severity of SAD and global symptoms; emotional neglect was also associated with lesser resilience. Childhood sexual abuse was associated with greater disability, whereas childhood physical abuse and neglect were not associated with any of these psychological outcomes. These findings indicate some specificity in the relationship between childhood trauma subtypes and general functional impairment in SAD. However, whether subtypes of childhood trauma might be differentially associated with specific clinical symptoms that have been implicated in the non-clinical literature (e.g., depression, anxiety, self-esteem) remains unknown. Given the evidence that SAD individuals with a history of childhood trauma have poorer treatment outcomes (Alden, Taylor, Laposa, & Mellings, 2006), such knowledge would be useful in developing treatment interventions for this group who does not maximally benefit from current treatments. 1.3. The present study The present study was designed to address two goals. The first goal was to compare differences in the frequency (how often an event occurred) and rates (what percent of the time an event occurred) of different forms of childhood trauma (sexual abuse, physical abuse, physical neglect, emotional abuse, and emotional neglect) in a large sample of individuals with generalized SAD versus a comparison group of healthy control participants (HCs). This addresses limitations in the current literature which include small clinical samples (Arrindell et al., 1983 and Arrindell et al., 1989) and the lack of an HC group (Simon et al., 2009). The second goal was to build upon recent findings linking subtypes of childhood trauma and functional impairment in individuals with SAD (Simon et al., 2009) by examining associations between different forms of childhood trauma and specific adult clinical symptoms (anxiety, depression, and self-esteem) in SAD. We hypothesized that, compared to HCs, individuals with SAD would have greater frequency and rates of childhood emotional abuse, emotional neglect, sexual abuse, physical abuse, and physical neglect. In line with previous findings (Simon et al., 2009), we expected that a greater frequency of childhood emotional abuse and neglect in the SAD sample would be associated with greater severity of current social anxiety. Drawing upon findings in non-clinical samples, we also predicted that greater frequency of childhood emotional abuse and neglect, but not sexual abuse, physical abuse, or physical neglect would be associated with greater anxiety, depression, and lower self-esteem within the SAD group.

. Results 3.1. Preliminary analyses Preliminary analyses were conducted to assess (1) whether there were demographic differences between SAD and HC groups, (2) the associations among childhood trauma variables, (3) the associations among demographic variables and childhood trauma variables, (4) the associations among clinical symptom variables, and (5) between-group differences in the clinical symptom variables. 3.1.1. Demographic variables There was no significant between-group difference in age (SAD: M = 33.47, SD = 8.59, HC: M = 32.60, SD = 9.00, t(130) = 0.48, p = .63) or in number of years of education (SAD: M = 16.69, SD = 2.22, HC: M = 17.48, SD = 2.03, t(122) = 1.72, p = .09), although education data were missing for eight participants (7 SAD, 1 HC). Gender did not differ significantly across the groups, χ2(1, n = 132) = 0.04, p = .85. 3.1.2. Relationships among types of childhood trauma Within the SAD group, all five childhood trauma subscales (sexual abuse, physical abuse, physical neglect, emotional abuse, and emotional neglect) were significantly positively associated with each other, with the exception of the correlations between sexual abuse and either emotional abuse or neglect. Only emotional abuse and emotional neglect were significantly positively associated in the HC group (see Table 1). Table 1. Correlations among age, gender, and Childhood Trauma Questionnaire subscales in individuals with social anxiety disorder (n = 102) (below diagonal) and healthy controls (n = 30) (above diagonal). Age Sexual abuse Physical abuse Physical neglect Emotional abuse Emotional neglect Age – .42* .19 .08 .18 .23 Sexual abuse .19 – .13 .03 .13 −.09 Physical abuse .13 .32** – −.12 .29 −.09 Physical neglect .05 .33** .26** – .18 .11 Emotional abuse .26** .17 .43** .32** – .56** Emotional neglect .35*** .15 .42** .21* .77** – Note. Correlations represent Spearman's rho. * p < .05. ** p < .01. *** p < .001. Table options 3.1.3. Relations among demographic variables and types of childhood trauma Age was significantly positively associated with childhood sexual abuse in the HC group and was significantly positively associated with childhood emotional abuse and emotional neglect in the SAD group (see Table 1). Mann–Whitney tests indicated that males had greater histories of childhood physical abuse than females in the SAD group only (U = 913.00, p < .01). All other comparisons were non-significant (SAD: sexual abuse, U = 1237.50, p = .59; physical neglect, U = 1153.50, p = .90; emotional abuse, U = 1158.50, p = .35; emotional neglect, U = 1090.00, p = .56; HC: sexual abuse, U = 105.50, p = .78; physical abuse, U = 106.00, p = .81; physical neglect, U = 92.50, p = .51; emotional abuse, U = 96.50, p = .51; emotional neglect, U = 92.50, p = .41). 3.1.4. Relations among clinical variables Associations between clinical variables (social anxiety, trait anxiety, depression, and self-esteem) were moderate for both groups (see Table 2). Table 2. Correlations among clinical symptom variables in individuals with social anxiety disorder (n = 102) (below diagonal) and healthy controls (n = 30) (above diagonal). Social anxiety Trait anxiety Depression Self-esteem Social anxiety – .41* .03 −.24 Trait anxiety .38** – .43* −.46* Depression .25* .58* – −.12 Self-esteem −.40** −.64** −.48** – Note. Correlations represent Spearman's rho. * p < .05. ** p < .01. Table options 3.1.5. Differences in social anxiety, trait anxiety, depression, and self-esteem between patients with SAD and HCs Independent-sample t-tests demonstrated significant between-group differences in social anxiety, trait anxiety, depression, and self-esteem (see Table 3). Table 3. Differences in clinical symptom variables between individuals with social anxiety disorder (n = 102) and healthy controls (n = 30). SAD HC t(df) M(SD) M(SD) Social anxiety 53.18 (9.95) 11.37 (7.79) 21.28 (126)*** Trait anxiety 54.26 (9.57) 28.13 (4.89) 19.85 (126)*** Depression 11.91 (9.18) 2.11 (3.34) 8.74 (124)*** Self-esteem 25.51 (5.29) 35.44 (3.25) 12.00 (120)*** *** p < .001. Table options 3.2. Differences in childhood trauma between patients with SAD and HCs Mann–Whitney tests demonstrated that, compared to HCs, individuals with SAD reported greater frequency of emotional abuse (U = 816.00, p < .001), emotional neglect (U = 708.00, p < .001), and total childhood trauma (U = 835.00, p < .001) (see Fig. 1). There was a trend towards greater childhood sexual abuse in the SAD group (U = 1289.00, p = .07). There were no significant between-group differences in childhood physical abuse (U = 1276.00, p = .14) or neglect (U = 1258.50, p = .25). Differences in the endorsed frequencies of childhood trauma among individuals ... Fig. 1. Differences in the endorsed frequencies of childhood trauma among individuals with social anxiety disorder (SAD) and healthy controls (HC). Error bars represent the standard error of measurement. ***p < .001. Figure options See Table 4 for rates of each trauma subtype for both groups. Consistent with the above analyses examining differences in endorsed per-person frequencies of childhood trauma, Mann–Whitney tests demonstrated that, compared with HCs, individuals with SAD had greater rates of childhood emotional abuse (U = 954.00, p < .001) and childhood emotional neglect (U = 1077.00, p < .01). There were no significant between-group differences in rates of childhood sexual abuse (U = 1362.00, p = .14), childhood physical abuse (U = 1320.00, p = .14), or childhood physical neglect (U = 1446.00, p < .32). Table 4. Rates of each childhood trauma subtype in individuals with social anxiety disorder (n = 102) and healthy controls (n = 30). SAD HC n % n % Sexual abuse 18 17.6 2 6.7 Physical abuse 31 30.4 5 16.7 Physical neglect 93 91.2 29 98.7 Emotional abuse 52 51.0 4 13.3 Emotional neglect 37 36.3 2 6.7 Table options 3.3. Relationships between types of childhood trauma and adult clinical symptoms in SAD As shown in Table 5, within the SAD sample, child emotional abuse and neglect were positively correlated with current social anxiety and trait anxiety, and negatively correlated with self-esteem. Childhood emotional neglect was also significantly positively correlated with depression. Childhood sexual abuse, physical abuse, and physical neglect were not correlated with any of these variables. Because of significant associations between age and childhood emotional abuse/neglect found in our preliminary analyses, we examined whether these significant relations remained when controlling for age. Results did not change (emotional abuse–social anxiety, r = .26, p < .05, emotional abuse–trait anxiety, r = .28, p < .01, emotional abuse–self-esteem, r = −.35, p < .001, emotional neglect–social anxiety, r = .25, p < .05, emotional neglect–trait anxiety, r = .40, p < .001, emotional neglect–depression, r = .31, p < .01, emotional neglect–self-esteem, r = −.44, p < .001). We further evaluated whether the observed pattern of results might be an artifact of depressive affect by conducting partial correlations between childhood abuse and neglect with social anxiety, trait anxiety, and self-esteem, controlling for levels of depression. Results did not change (emotional abuse–social anxiety, r = .23, p < .05, emotional abuse–trait anxiety, r = .24, p < .05, emotional abuse–self-esteem, r = −.31, p < .01, emotional neglect–trait anxiety, r = .28, p < .01, emotional neglect–self-esteem, r = −.35, p < .01), with the exception of the relation between childhood emotional neglect and social anxiety, which was no longer significant (r = .19, p > .10). Table 5. Correlations between Childhood Trauma Questionnaire subscales and clinical symptoms in individuals with social anxiety disorder (n = 102). Social anxiety Trait anxiety Depression Self-esteem Sexual abuse .10 .10 .07 −.06 Physical abuse .12 .15 .11 −.13 Physical neglect .07 .04 .13 −.06 Emotional abuse .26* .30** .18 −.35*** Emotional neglect .25* .40*** .31** −.44*** * p < .05. ** p < .01. *** p < .001.