وقایع آسیب زا و اجتماعی استرس زا در زندگی در میان افراد مبتلا به اختلال اضطراب اجتماعی

Abstract This study examined the frequency of reexperiencing, avoidance, and hyperarousal symptoms most often associated with posttraumatic stress disorder (PTSD) among 45 persons with social anxiety disorder and 30 nonanxious controls in response to an extremely stressful social event (which did not satisfy DSM-IV's PTSD Criterion A). Avoidance and hyperarousal in response to reminders of socially stressful events were common among patients; more than one-third would have met criteria for PTSD if these events satisfied DSM-IV PTSD Criterion A. Frequency of this PTSD-like symptom pattern did not differ among patients who did and did not experience another event that did satisfy PTSD Criterion A. Implications of these findings for the treatment of social anxiety disorder are discussed.

. Results Two patients reported experiencing neither a Criterion A event nor a socially stressful event. Because these patients were not representative of the population of interest to the current study, they were excluded from all further analyses. The remaining patient sample (n = 45) was divided into two groups: those who reported experiencing exclusively socially stressful events (group SSE; n = 16; 35.6%) and those who also reported experiencing at least one Criterion A traumatic event (group TE; n = 29; 64.4%). As described previously, this division was made to determine whether persons with a history of at least one Criterion A event experience socially stressful events or endorse levels of social anxiety and other symptoms differently from persons without such a history. Among the normal controls (NC; n = 30), 23.3% (n = 7) reported neither a socially stressful nor a Criterion A event; 23.3% (n = 7) reported a socially stressful event only; 6.7% (n = 2) reported a Criterion A event only; and 46.7% (n = 14) reported both a socially stressful and a Criterion A event. Comparisons were made among the two patient groups and the normal control group and included only those NC participants who reported events relevant to specific comparisons. More specifically, only those normal control participants who reported experiencing a socially stressful event (n = 21) were included in analyses of symptoms associated with socially stressful events (e.g., reexperiencing, avoidance, and hyperarousal symptoms associated with socially stressful events; rates of PTSD-like symptom pattern). Similarly, only those normal control participants who reported experiencing a Criterion A traumatic event (n = 16) were included in analyses of symptoms associated with these events (e.g., reexperiencing, avoidance, and hyperarousal symptoms associated with Criterion A traumatic events; rates of PTSD). The entire sample of normal control participants (n = 30) was included in full-sample analyses (e.g., characteristics of the study sample; severity of social anxiety disorder and depression and degree of life satisfaction). Groups TE, SSE, and NC were compared on continuous and categorical demographic variables using three-group omnibus analyses of variance (ANOVAs) and chi-square (χ2) analyses, respectively (see Table 1). No differences were found on age, annual income, number of years of education, sex, or employment status. However, differences were found on race and marital status. In post hoc 2 × 2 χ2 analyses, the proportions of non-Caucasian participants in the combined patient group (33.33%) and the normal control group (23.33%) did not differ [χ2 (df = 1, n = 75) = 0.87, ns, ES = 0.11]. 4 However, there was a moderate difference in the percentage of non-Caucasian patients in group TE (44.83%) versus group SSE (12.50%) [χ2 (df = 1, n = 45) = 4.85, P < .05, ES = 0.33]. 5 In addition, the three groups differed with a moderate effect size on marital status. However, the total number of participants who were married (n = 6) was so small that the χ2 test was not considered valid (A. Sockloff, personal communication, June 28, 2001). Therefore, conclusions that can be drawn from this result are limited and post hoc analyses were not pursued. 6 Table 1. Characteristics of the study sample TE (n = 29) SSE (n = 16) NC (n = 30) F Age M 27.59 29.38 30.33 0.48 S.D. 9.21 10.10 12.67 Annual income M 19310.86 33200.00 35675.86 2.08 S.D. 17990.97 47000.30 32645.12 Education (years) M 14.39 14.44 15.07 0.59 S.D. 2.69 2.10 2.66 n % n % n % χ2 Sex Female 18 62.07 8 50.00 11 36.67 3.81 Male 11 37.93 8 50.00 19 63.33 Race Caucasian 16 55.17 14 87.50 23 76.67 6.07* Non-Caucasian 13 44.83 2 12.50 7 23.33 Employment Employed 23 79.31 12 75.00 27 90.00 2.01 Not Employed 6 20.69 4 25.00 3 10.00 Marital Status Married 1 3.45 4 25.00 1 3.33 7.99* Single 28 96.55 12 75.00 29 96.67 Note. TE = patients who reported experiencing at least one Criterion A event; SSE = patients who reported experiencing exclusively socially stressful events; NC = normal controls. Sample sizes vary due to missing data. * P < .05. Table options 2.1. Number of reexperiencing, avoidance, and hyperarousal symptoms associated with socially stressful and Criterion A events Two one-way multivariate analyses of variance (MANOVAs) were computed. The first MANOVA included participant group as the independent variable and number of reexperiencing, avoidance, and hyperarousal symptoms endorsed for the worst reported socially stressful event as the dependent variables. The second MANOVA included participant group as the independent variable and number of reexperiencing, avoidance, and hyperarousal symptoms endorsed for the worst reported Criterion A event as the dependent variables. 2.1.1. Socially stressful events All 45 patients and 21 of the 30 normal controls reported experiencing a socially stressful event. In addition, the 29 patients in group TE reported experiencing a Criterion A event. Among patients who reported any reexperiencing, avoidance, and hyperarousal symptoms associated with their respective socially stressful event (n = 37), 94.60% (n = 35) reported experiencing such symptoms for more than 3 months and 5.40% (n = 2) reported experiencing such symptoms for less than 1 month. The MANOVA for reexperiencing, avoidance, and hyperarousal symptoms associated with socially stressful events [Wilks’ λ = 0.71; F(6, 124) = 3.87; P < .001; ES = 0.21] yielded a significant effect for group indicating that groups TE, SSE, and NC differed on mean number of reexperiencing, avoidance, and hyperarousal symptoms experienced in relation to socially stressful events. To clarify the multivariate results, the MANOVA was followed by three ANOVAs, and significant univariate effects were followed by Tukey HSD post hoc comparisons (see Table 2). Groups SSE and TE did not differ, but both endorsed a greater mean number of avoidance and hyperarousal symptoms than did group NC. The group differences in avoidance and hyperarousal symptoms met Cohen's (1990) convention for a moderate effect. The differences among groups in reexperiencing symptoms were not significant in the univariate analysis. Table 2. Number of reexperiencing, avoidance, and hyperarousal symptoms associated with socially stressful events as a function of participant group TE (n = 29) SSE (n = 16) NC (n = 21) F ES Reexperiencing symptoms 0.85 0.00 M 1.41 1.44 0.91 S.D. 1.68 1.37 1.38 Avoidance symptoms 8.17*** 0.33 M 2.86a 2.94a 0.73b S.D. 2.33 2.27 1.42 Hyperarousal symptoms 8.69*** 0.34 M 2.03a 1.63a 0.23b S.D. 1.92 1.75 0.61 Note. TE = patients who reported experiencing at least one Criterion A event; SSE = patients who reported experiencing exclusively socially stressful events; NC = normal controls; ES = effect size. Means in the same row that do not share the same superscript differ at P < .05 in the Tukey honestly significant difference post hoc comparison. *** P < .001. Table options 2.1.2. Criterion A events Patients in group SSE reported no Criterion A events and were excluded from these analyses. Among patients who reported any reexperiencing, avoidance, and hyperarousal symptoms associated with their respective traumatic event (n = 25), 96.00% (n = 24) reported experiencing such symptoms for more than 3 months and 4.00% (n = 1) reported experiencing such symptoms for 1 to 3 months. The MANOVA for reexperiencing, avoidance, and hyperarousal symptoms in connection with Criterion A events [Wilks’ λ = 0.73; F(3, 42) = 5.20; P < .01; ES = 0.30] yielded a significant effect for group indicating that groups TE and NC differed on mean number of reexperiencing, avoidance, and hyperarousal symptoms experienced in relation to Criterion A events. To clarify the multivariate results, the MANOVA was followed by three ANOVAs (see Table 3). Group TE endorsed a greater mean number of reexperiencing, avoidance, and hyperarousal symptoms than did group NC. The effect size for avoidance symptoms was large, and the effect sizes for reexperiencing and hyperarousal symptoms were moderate. Table 3. Number of reexperiencing, avoidance, and hyperarousal symptoms associated with criterion a events as a function of participant group TE (n = 29) NC (n = 16) F ES Reexperiencing symptoms 6.24* 0.34 M 1.72 0.65 S.D. 1.60 1.00 Avoidance symptoms 13.73*** 0.53 M 2.38 0.41 S.D. 2.13 0.62 Hyperarousal symptoms 4.48* 0.28 M 1.28 0.35 S.D. 1.71 0.70 Note. TE = patients who reported experiencing at least one Criterion A event; NC = normal controls; ES = effect size. * P < .05. *** P < .001. Table options 2.2. Rates of a PTSD-like symptom pattern and PTSD Rates of a PTSD-like symptom pattern and PTSD were derived from participants’ PDS-revised responses (see Table 4). A PTSD-like symptom pattern was considered present when all the criteria for PTSD (with the exception of Criterion A) were met for each individual's worst reported socially stressful event. PTSD was considered present when all the criteria for PTSD (including Criterion A) were met for each individual's worst reported qualifying Criterion A event. For the PTSD-like symptom pattern, χ2 tests were conducted comparing the proportions of this symptom pattern in groups TE, SSE, and NC. Groups differed significantly on proportions of this PTSD-like symptom pattern. In post hoc 2 × 2 χ2 analyses, the proportions of patients with this symptom pattern in group TE (34.48%) and SSE (37.50%) did not differ [χ2 (df = 1, n = 45) = 0.04, ns, ES = 0.03]. However, the percentage of participants with the PTSD-like symptom pattern in the combined patient group (n = 16/45; 35.56%) differed from that of group NC (n = 0/21; 0.00%) [χ2 (df = 1, n = 66) = 10.28, P < .001, ES = 0.39]. With respect to PTSD, χ2 analyses were calculated to examine the proportions of PTSD in groups TE and NC. The percentage of patients in group TE with PTSD differed moderately from that of group NC. 7 Table 4. Rates of a PTSD-like symptom pattern and PTSD as a function of participant group TE (n = 29) SSE (n = 16) NC (n = 21) χ2 ES Socially stressful events PTSD-like symptom pattern n 10 6 0 % 34.48 37.50 0.00 No PTSD-like symptom pattern n 19 10 21 % 65.52 62.50 100.00 10.33** 0.39 TE (n = 29) NC (n = 16) χ2 ES Criterion A events PTSD n 8 0 % 27.59 0.00 No PTSD n 21 16 % 72.41 100.00 5.68* 0.32 Note. TE = patients who reported experiencing at least one Criterion A event; SSE = patients who reported experiencing exclusively socially stressful events; NC = normal controls; ES = effect size. * P < .05. ** P < .01. Table options 2.3. Severity of social anxiety disorder and depression, and degree of life satisfaction A MANOVA with group as the independent variable and self-report measures of social anxiety, depression, and life satisfaction as the dependent variables was calculated. The MANOVA yielded a significant effect for group [Wilks’ λ = 0.19; F(10, 120) = 15.37; P < .001; ES = 0.46] indicating that the groups differed on social anxiety, depression, and life satisfaction. To clarify the multivariate results, the MANOVA was followed by five one-way ANOVAs, and significant univariate effects were followed by Tukey HSD post hoc comparisons (see Table 5). Groups SSE and TE did not differ on any measure, but both patient groups had higher scores on the SIAS, SPS, FQS, and BDI, and lower scores on the QOLI than did group NC. The effect sizes for these differences were large. Table 5. Severity of social anxiety disorder and depression, and degree of life satisfaction as a function of participant group TE (n = 29) SSE (n = 16) NC (n = 30) F ES Social Interaction Anxiety Scale 85.75*** 1.12 M 52.04a 47.15a 10.96b S.D. 15.28 14.74 6.38 Social Phobia Scale 86.47*** 1.13 M 38.85a 35.38a 4.64b S.D. 13.69 11.23 3.97 Fear Questionnaire-Social Subscale 45.62*** 0.82 M 22.96a 22.46a 7.21b S.D. 6.98 7.10 5.95 Beck Depression Inventory 30.62*** 0.66 M 15.77a 16.31a 1.86b S.D. 9.63 8.81 2.26 Quality of Life Inventory 27.44*** 0.63 M −0.50a 0.31a 2.04b S.D. 1.66 1.38 0.69 Note. TE = patients who reported experiencing at least one Criterion A event; SSE = patients who reported experiencing exclusively socially stressful events; NC = normal controls; ES = effect size. Means in the same row that do not share the same superscript differ at P < .05 in the Tukey honestly significant difference post hoc comparison. *** P < .001. Table options 2.4. Relationship of type of event to reexperiencing, avoidance, and hyperarousal symptom severity In order to determine whether patients with social anxiety disorder experienced socially stressful and Criterion A events differently, the number of symptoms experienced in connection with both socially stressful and Criterion A events were compared for patients who reported experiencing both classes of events (Group TE; n = 29). Patients endorsed significantly more hyperarousal symptoms in connection with socially stressful events than in connection with Criterion A events (SSE: M = 2.03, S.D. = 1.92; TE: M = 1.28, S.D. = 1.71; t(28) = 2.26, P < .05; ES = 0.41). There were no differences in patients’ report of reexperiencing (SSE: M = 1.41, S.D. = 1.68; TE: M = 1.72, S.D. = 1.60; t(28) = −1.30, ns; ES = 0.19) or avoidance (SSE: M = 2.86, S.D. = 2.33; TE: M = 2.38, S.D. = 2.13; t(28) = 1.26, ns; ES = 0.21) symptoms in response to socially stressful versus Criterion A events. 2.5. Severity of social anxiety, depression, and quality of life in patients with and without a PTSD-like symptom pattern and PTSD Social anxiety, depression, and quality of life scores for patients with and without a PTSD-like symptom pattern were compared. There were no differences between groups on the SIAS [t(39) = −1.46, ns, ES = 0.48]; SPS [t(39) = −0.70, ns, ES = 0.23]; BDI [t(38) = −1.53, ns, ES = 0.49]; QOLI [t(39) = 1.67, ns, ES = 0.55]; or FQS [t(43) = −2.19, ns, ES = 0.67]. A similar comparison was conducted between patients with and without PDS-revised diagnoses of PTSD. There were no differences found between groups on any measure [SIAS: t(25) = −0.69, ns, ES = 0.29; SPS: t(25) = −1.37, ns, ES = 0.58; FQS: t(27) = −1.39, ns, ES = 0.57; BDI: t(24) = −1.91, ns, ES = 0.08; QOLI: t(25) = 1.36, ns, ES = 0.57].