Quality of life following cognitive behavioral treatment for social anxiety disorder: Preliminary findings

There has been little research on quality of life, a multidimensional construct that encompasses psychological functioning, social functioning, physical symptoms, and age-appropriate functioning [Aaronson et al., 1988], among individuals with social anxiety disorder. However, the enormous effect of social anxiety disorder on perceived physical and emotional health, occupational functioning, educational achievement, and interpersonal relationships [Schneier et al., 1994; Wittchen et al., 1999] suggests the importance of this area of study. We present data that follows up on the finding of Safren et al. [1997] that socially anxious patients were significantly improved in selfperceived quality of life immediately following cognitive-behavioral group therapy (CBGT). Two hypotheses were tested: 1) CBGT leads to significant improvements in subjectively perceived quality of life from preto post-treatment and 2) these gains are maintained or additional improvement occurs during the follow-up period. We also examined the relationship between quality of life and symptom severity at pre-treatment, post-treatment, and 6-month follow-up assessments. The sample consisted of 25 patients (age, M = 35.80 years, SD = 11.73; 60% males) with a principal diagnosis of social anxiety disorder/social phobia as assessed by the Anxiety Disorders Interview Schedule-Revised [ADIS-R; DiNardo and Barlow, 1988] or the Anxiety Disorders Interview Schedule for DSM-IV-Lifetime Version [ADIS-IV-L; DiNardo et al., 1994]. All patients received a 12-week course of CBGT for social anxiety [Heimberg and Becker, 2001]. Twelve patients were also studied by Safren et al. [1997] but no follow-up data were included in that report. At each post-treatment assessment, an independent assessor completed a seven-point rating of change from baseline. Patients rated as markedly or moderately improved (ratings of 1 or 2) were classified as treatment responders. The assessor also completed a severity of illness item (1 = normal, 7 = among the most severely ill patients), the International Personality Disorder Examination—Avoidant Personality Disorder (APD) Module [Loranger, 1995], and the Liebowitz [1987] Social Anxiety Scale at all assessment points. Self-report measures of social anxiety were the Social Interaction Anxiety Scale and the Social Phobia Scale [Mattick and Clarke, 1998]. The Beck Depression Inventory [Beck et al., 1979] assessed self-reported depressive symptoms. The Quality of Life Inventory [QOLI; Frisch, 1994] served as the key measure of the importance of, and satisfaction with, 16 different domains of life (e.g., work, health, friends, community). Sixteen patients (64%) were classified as responders and nine (36%) as non-responders. Eighteen (72%) patients met criteria for generalized social anxiety disorder, and nine (36%) patients had probable or definite APD. Mean QOLI scores were significantly lower than the mean of the nonclinical adult sample reported by Frisch [1994] (M = 2.66; SD = 1.20) at all points (Table 1).There were no significant differences in pre-treatment QOLI scores as a function of social anxiety subtype [t(23) = 1.31, ns]. However, patients with probable or definite APD had lower scores (M = –0.42, SD = 1.21) than patients without APD (M = 0.87, SD = 1.42) [t(23) = 2.29, P = .02]. Planned contrasts [Rosenthal et al., 2000] were used to examine our main hypotheses. First, there were significant improvements in QOLI scores from preto post-treatment [t(24] = 2.87, P = .004]. However, a significant change in QOLI scores from post-treatment to follow-up did not occur [t(24) = .34, ns], suggesting that gains were maintained but substantial additional improvements were not realized. Analysis of responders only also failed to reveal significant differences between post-treatment (M = 0.96, SD = 1.78) and follow-up QOLI scores (M = 1.29, SD = 1.80) [t(15) = 0.97, ns], although the means improved across time. We also examined the relationship of the QOLI with measures of social anxiety and depressive symptoms at each assessment (Table 1). The QOLI was significantly correlated only with depression scores at preand post-treatment. At followup, however, QOLI scores were also significantly correlated with self-reported social interaction anxiety and severity of illness. The magnitude of the correlation between the QOLI and social interaction anxiety remained substantial after partialling out depressive symptoms (r = –.59, P < .005), but controlling for social interaction anxiety reduced the follow-up correlation of QOLI and depression to a trend (r = –.39, P < .10). Partialling out depressive symptoms also reduced the r with severity of illness (r = –.36, P < .10). The current findings suggest that cognitive-behavioral treat-