Social Phobia

ANXIETY DISORDERS: LONGITUDINAL COURSE AND TREATMENT

0193-953X/95 $0.00 + .20

SOCIAL PHOBIA Longitudinal Course and Long-term Outcome of Cognitive-Behavioral Treatment Harlan R. Juster, PhD, and Richard G. Heimberg, PhD

When fear associated with social interaction or performance situations produces severe personal distress or functional impairment, social phobia may be diagnosed. This anxiety disorder, when first described by Marks and Gelder, 48 was limited to fear in a number of specific performance situations including public speaking, writing, or eating in public, or using public restrooms while others were present. This conception of social phobia was reflected in the third edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-IIl). 3 More recent conceptualizations suggest that patients with social phobia may fear and avoid many types of social interaction situations in addition to the specific performance situations already noted. A generalized subtype was added with the revision of the diagnostic manual (DSM-IIl-R) 4 to describe individuals who fear most social situations. This conceptualization is virtually unchanged in DSM-IV. 5 Social phobia currently is recognized as potentially debilitating, and the impairment caused by the disorder may be extensive. It may limit educational attainment, career advancement, and social functioning. 6 • 44• 75 Schneier et al 62 assessed impairment in 32 social phobic patients, who were not currently depressed, recruited from an anxiety disorders clinic or from clinicians' private practices. A majority of patients rePreparation of this article was supported in part by grant No. 44119 from the National Institute of Mental Health to the second author.

From the Social Phobia Program, Center for Stress and Anxiety Disorders, University at Albany, State University of New York, Albany, New York

THE PSYCHIATRIC CLINICS OF NORTH AMERICA VOLUME 18 •NUMBER 4 •DECEMBER 1995

821

822

JUSTER & HEIMBERG

ported at least moderate lifetime impairment in education and work, family and social relationships, and participation in organized activities. Social phobia has been associated with substance abuse,6· 61 mood disorders,61 suicidal ideation,63 and suicide attempts. 6 Social phobia is more prevalent than previously believed. The most recent epidemiologic survey (the National Comorbidity Survey),43 conducted with a national sample of more than 8000 participants across the United States, reported that the 12-month prevalence rate of social phobia was 7.9% (6.6% for men and 9.1 % for women). The reported lifetime prevalence rate of 13.3% (11 .l % for men and 15.5% for women) makes social phobia the third most common mental disorder, after major depression and alcohol dependence. Schneier et al63 reported the mean age of onset of social phobia to be 15.5 years at four National Institute of Mental Health E~idemiological Catchment Area Study sites. Closer inspection of these data, however, gives a slightly different slant.39 The distribution of age of onset is highly positively skewed and bimodal. The first peak occurs at less than 5 years of age, and the second peak occurs at about 13 years of age. These data indicate that, for at least some persons, social phobia may be a much more chronic condition than is suggested by reported mean age of onset. Social phobia does appear to follow a chronic and unremitting course. 6· 58 Reich et al58 followed 140 patients in what was characterized as a study of the natural course of social phobia. Despite the fact that most patients (88%) received some form of psychotherapy or pharmacotherapy during their participation in the study, only 11 % achieved full remission of social phobia symptoms after 65 weeks. Unfortunately, psychotherapy was characterized only as individual, group, or family/ couples therapy, so few conclusions can be drawn about its adequacy. Reich et al57 were unable to predict which patients would achieve any level of remission based on a broad array of variables including demographic information, comorbid diagnoses, duration of illness, and a survey of social/emotional functioning. No studies have yet examined fluctuation of social phobia symptoms over time. Studies in several related areas may have implications for further understanding the course of social phobia. These include studies of childhood shyness, behavioral inhibition to the unfamiliar, and family studies of social phobia. (For a fuller discussion, see Heimberg et al. 33) For example, Caspi et al1 4 followed shy children prospectively into adulthood and found that shy boys were slower to marry and to achieve occupational stability in adult life than nonshy boys. Shy girls were more likely to become housewives or to remain at home after childbirth than nonshy girls. Behavioral inhibition to the unfamiliar may occur in 10% to 15% of assessed children and is defined by behavioral withdrawal in the face of novel people, places, and things. 41 Rosenbaum and colleagues60 hypothesized that behavioral inhibition may be linked etiologically to the anxiety disorders, including (especially) social phobia. A longitudinal study of children classified as behaviorally inhibited prior to age 3

SOCIAL PHOBIA

823

years found the majority remained inhibited for almost a decade. These children tended to be introverted and more physiologically reactive to stress than uninhibited children. Other studies have shown a higher likelihood of childhood anxiety disorders and socially related fears in behaviorally inhibited children than in uninhibited children9 and a higher rate of anxiety disorders, especially social phobia, in the parents of behaviorally inhibited children59 than in the parents of uninhibited or normal children. Family studies of social phobia indicate that it is much more common among relatives of social phobia patients than relatives of nonpatients. Fyer and colleagues21 reported that 16% of the relatives of social phobia patients also were diagnosed with social phobia, whereas just 5% of the relatives of nonpatients were so diagnosed. These figures were replicated in a subsequent study examining risk among first-degree relatives of generalized social phobic patients (16%), although the rate of social phobia among relatives of nongeneralized social phobic patients (6%) resembled the rate in the normal control group. 47 REVIEW OF LONG-TERM TREATMENT STUDIES

Studies of the outcome of treatments for social phobia have received increasing attention during the last several years as interest in this disorder has expanded. Several reviews have been completed by our research group on cognitive-behavioral treatments for social phobia.23• 25• 26• 27• 38• 40 This article focuses primarily on long-term outcome data. Therefore, only a subset of the studies covered by the previous reviews is considered. Studies with data on long-term outcome, which have since been conducted, also are reviewed. Although this article is designed to stand alone, it would also be complemented by one of the more recent reviews noted previously. Whereas the focus is on long-term outcome, it remains necessary to discuss post-treatment effects and methodologic issues to clarify the meaning of the longitudinal data. These, however, are kept brief in comparison with previous reviews. In Heimberg's23 early review of 17 empirical studies of the behavioral or cognitive-behavioral treatment of social phobia, only one study included a follow-up assessment as long as 12 months, and that was a single-case study. 67 The average length of follow-up for the 17 studies was a meager 5.12 months. In this article, we review 32 treatment outcome studies of social phobia that included a follow-up period. The range of follow-up for these studies is 1 to 65 months. The average length of follow-up is 9.73 months. This appears to be much higher than was reported in Heimberg's23 review, but it should be noted we have excluded those studies with no follow-up at all. The modal length of follow-up was 6 months (11 studies) followed by 3 months (6 studies). Eight studies followed their patients for at least 1 year after treatment, with the five longest studies conducting assessments 18 (2 studies), 24, 30, and 65 months after the end of treatment.

824

JUSTER & HEIMBERG

The bulk of the treatments reviewed fall into one of four categories: exposure therapy, cognitive restructuring techniques, social skills training, or combinations of these techniques. Exposure therapy refers to any of the class of interventions that use confrontation with a feared stimulus by a patient. Confrontation can occur in real life (in vivo exposure), in role-plays during therapy sessions, or in imagination. Cognitive restructuring techniques attempt to alter patients' perceptions and interpretations of feared situations, their own capacity to perform in feared situations, and outcomes often misperceived as failure. These techniques most often are derived from Rational Emotive Therapy (RET)16 or Beck's Cognitive Therapy,8 although these two treatments include a number of behavioral techniques as well. Social skills training, which presumes that interpersonal anxiety is a result of deficient social skills, incorporates modeling of appropriate behavior, behavioral rehearsal, corrective feedback, social reinforcement, and homework assignments.70 We review evidence for the long-term effectiveness of each of these techniques. We also address the issue of whether cognitive techniques may enhance the efficacy of exposure therapy. Finally, we provide a summary of longterm outcome with recommendations for future research. Exposure Therapy

Although the nature of the mechanism by which exposure results in anxiety reduction remains elusive,7 exposure continues to be one of the most well-studied treatments for social phobia. In this section, we review one uncontrolled study and three studies that compared exposure (as the sole treatment) with alternative treatments such as relaxation therapy and social skills training. Comparisons of exposure with cognitive restructuring techniques are reviewed in a separate section. Followup for these studies varied in length from 4.5 months to 30 months. In an uncontrolled study, Fava et al1 8 treated 10 social phobic patients with eight sessions of self-exposure instructions. Seven patients completed treatment and evidenced significant reductions in both selfreport and independent assessor measures of social anxiety and general anxiety. Gains in general anxiety and self- and other-rated social anxiety were maintained at the 1-year follow-up assessment for six of the seven patients. Al-Kubaisy et aP compared self-directed in vivo exposure (with or without therapist-guided exposure) to relaxation strategies. Twentyeight social phobic patients were assigned randomly to treatments as part of a larger study, which also included agoraphobic and specific phobic patients. Assessments consisted of self-report and independent assessor measures of social fear and avoidance, general anxiety, and depression. In general, exposure fared better than relaxation procedures. At post-treatment, social phobic patients receiving the additional therapist-guided exposure fared better than patients receiving only self-directed exposure instructions, but on only 4 of 27 measures. Patients were

SOCIAL PHOBIA

825

assessed 4.5 months after treatment ended. Therapist-directed exposure appeared to provide little additional advantage over self-directed exposure on measures related to social anxiety. Small samples (only 8 social phobia patients completed each treatment), however, may have minimized power to detect differences between the treatment conditions. Alstrom et al2 compared therapist-directed exposure, dynamically oriented supportive therapy, and relaxation therapy. In addition, all patients received unspecified anxiolytic medication and self-exposure instructions. Forty-two patients were assigned randomly to one of the three treatments or the control therapy. Patients were instructed to remain in the stimulus situation until anxiety abated. Exposure was clearly superior to supportive therapy, relaxation therapy, and to the control condition, in reducing anticipatory and situational anxiety immediately following treatment. At 9-month follow-up, exposure was superior only to the control condition in terms of both anxiety and avoidance. Differences between exposure, supportive therapy, and relaxation therapy evident at post-treatment disappeared at follow-up, partly as a result of clinical deterioration among patients treated with exposure. Specifically, patients treated with exposure showed some return of anticipatory anxiety and avoidance of social situations, although they did not return to baseline levels. Some methodologic problems, described in more detail in earlier reviews (e.g., differences between groups on gender ratio, demographics, and pretreatment impairment) limit the utility of these findings. Probably the most troublesome was the unsystematic administration of treatments within and across conditions. For instance, imaginal exposure was substituted for in vivo exposure for one patient and supplemented in vivo exposure for two others. Furthermore, the control therapy included instructions to patients for self-exposure. Finally, anxiolytic medication was used across conditions, and it is not known how this may have affected patients' response to the various treatments. One strategy meant to improve outcome has been to match treatments to specific patient problems. Wlazlo et aF7 compared individual and group exposure therapy and social skills training for patients judged to have either primary social anxiety versus patients whose anxiety was judged to be secondary to performance deficits. All treatments produced significant within-group changes on most outcome measures. At posttreatment, primary anxious patients receiving group exposure reported reduced fear of social contact and greater ability to refuse requests than patients with social performance deficits treated with group exposure. There were few differences, however, between patients treated with exposure and those treated with social skills training, as the latter treatment also produced significant change. Follow-up for this study included a 3-month post-treatment assessment and a longer-term follow-up conducted within a range of 1.5 to 5.5 years (mean = 2.5 years) following the conclusion of treatment. Most gains achieved during the treatment phase were maintained during follow-up. Overall, exposure and social skills training were equally

826

JUSTER & HEIMBE RG

effective although performance-deficit patients showed their best longterm response to treatment with group exposure. Several problems with this study, noted by Heimberg and Juster,27 include lack of random assignment to treatment conditions, unequal time in therapy across treatments, and inclusion of unrelated techniques in the exposure condition such as modeling and coaching. In addition, instructions for self-exposure were included in all three conditions, thereby reducing the distinctiveness of the treatments. 40 Turner et aF3 compared the efficacy of 20 sessions of a combination of imaginal and in vivo exposure with atenolol, a cardioselective betablocker, and pill placebo (atenolol and placebo were administered double-blind). Seventy-two patients participated, and a broad-based assessment incorporated self-report measures, interviewer ratings, and responses to a behavior test (public speaking). At post-treatment, exposure was significantly more effective than atenolol and placebo on measures associated with the behavior test and composite measures of end-state functioning and improvement. Patients treated with atenolol were not improved significantly on any measure compared with patients receiving placebo at post-treatment. Nearly 60% of atenolol patients and 70% of exposure patients participated in a 6-month follow-up . Patients who were improved at posttest maintained their gains, regardless of treatment received. Exposure patients continued to outperform atenolol patients on two self-report measures, frequency of positive cognitions during the behavior test, and length of time they were willing to continue to give a speech during the behavior test. During follow-up, exposure patients showed additional improvement on the length of time they were willing to speak during the behavior test and positive cognitions during the behavior test. Atenolol patients showed no change on these measures or deteriorated slightly. Although exposure was clearly superior to atenolol in this study, Liebowitz et al45 have noted that atenolol may not be the pharmacotherapy of choice for social phobia. Combined Exposure and Cognitive Restructuring

Kanter and Goldfried 42 used imaginal exposure techniques to provoke anxiety in socially anxious community volunteers. Study participants were trained to respond to imaginally presented anxiety-evoking scenes with either cognitive restructuring (systematic rational restructuring), progressive relaxation (self-control desensitization), or both procedures (combined treatment). Participants in all three treatment conditions evidenced significant improvement over a wait-list group after 7 weeks. The systematic rational restructuring and combined treatments showed some additional advantage over the desensitization group at post-treatment assessment. At a 9-week follow-up, all three treatment groups continued to show marked improvement over their pretreatment level of functioning. Systematic rational restructuring participants

SOCIAL PHOBIA

827

showed greater reductions in trait anxiety, fear of negative evaluation, and irrational beliefs at follow-up than patients receiving self-control desensitization. Follow-up assessment was limited entirely to self-report measures, a common problem in follow-up studies of the treatment of social phobia. Stravynski67 treated a single patient who experienced psychogenic vomiting when visiting his girlfriend in the home of her parents, eating with her parents, and being in the presence of others at theaters and bars. Treatment combined in vivo exposure, social skills training, and cognitive restructuring, all introduced simultaneously. Vomiting was eliminated after 7 sessions, and anxiety in three of four target situations decreased dramatically. Seven months after treatment, the patient continued to perform the target behaviors, and an informal follow-up after 2 years suggested that gains had been maintained. Mersch et al 56 administered a unique combination of RET and paradoxical interventions to three patients with fears that public embarrassment would result from the occurrence of their physiologic symptoms of anxiety (e.g., blushing, sweating, trembling). All patients improved on a broad array of measures, and frequency of feared symptoms was reduced, although anxiety experienced on occurrence of these symptoms remained high. Although these gains were maintained at an 18-month follow-up, methodologic considerations limit conclusions that can be drawn. Specifically, the lack of any control conditions does not allow us to conclude which components of treatment were effective. This is especially difficult because paradoxical techniques incorporated exposure. Heimberg et al29 treated seven social phobic patients in small groups with a combination of imaginal exposure, exposure to simulated social situations, cognitive restructuring, and homework assignments for in vivo exposure. During treatment, cognitive restructuring was conducted immediately following exposure to social situations. These procedures focused on identifying cognitions that indicated a negative response to the situation due to faulty logic or biased negative interpretations. Significant reductions in social anxiety, general anxiety, and fear of negative evaluation were evident immediately after treatment. Patients rated their own performance during a behavior test as higher in quality, and they were rated by others as less anxious. Attributions for negative outcomes became less internal and stable. For six of the seven patients, gains were maintained at a 6-month follow-up on both self-report and behavior test measures. For one patient, anxiety returned to baseline levels. Based on their preliminary study, Heimberg et al3° removed imaginal exposure and further integrated cognitive restructuring and exposure exercises into a protocol referred to as Cognitive Behavioral Group Therapy for Social Phobia (CBGT). Cognitive restructuring is used before, during, and after simulated and in vivo exposures (the latter comprise homework assignments). CBGT is described more thoroughly and case examples are provided in articles by Heimberg et al32 and

828

JUSTER & HEIMBERG

Hope and Heimberg. 35 The initial test of these integrated procedures compared CBGT with an attention-placebo group. This comparison treatment combined educational presentations on topics relevant to social phobia and guided peer support. Forty-nine social phobic patients were assigned randomly to either CBGT or the placebo group. At posttest, CBGT patients reported less anxiety during a behavior test and were rated as less severely impaired by clinical assessors. At the 6month follow-up, CBGT patients maintained their gains and reported more positive and fewer negative thoughts during the behavior test. Eighty-one percent of CBGT patients, compared with 47% of placebo patients, were rated as having made clinically significant improvement at 6-month follow-up. In the longest-term follow-up study to date, Heimberg et al34 reassessed 19 of the patients who participated in the Heimberg et al3° study. Follow-up intervals ranged from 4.5 to 6.25 years (M = 5.5 years). Although these patients were less impaired at pretest than patients who did not participate in this follow-up, there were no differences between participating CBGT and control patients. Nonetheless, generalizations based on these results should be limited to less impaired patient populations. At the long-term follow-up, CBGT patients were rated as less severely impaired and as experiencing less interference at work, in social activities, and in their family life as a result of their social phobia symptoms. Eighty-nine percent of CBGT patients and 44% of placebo patients were judged to be clinically significantly improved. CBGT patients were judged to be barely symptomatic whereas placebo patients were rated as requiring further treatment. CBGT patients also were rated as evidencing more social skills and less anxiety during a behavioral test. Hope et al3 7 used CBGT to treat generalized and nongeneralized social phobic patients. Whereas generalized social phobic patients were more severely impaired than nongeneralized social phobic patients, both groups showed approximately equivalent improvement immediately after treatment. Generalized social phobic patients with and without an additional diagnosis of avoidant personality disorder improved similarly and made clinically significant gains, although they generally remained more impaired after treatment than did nongeneralized social phobic patients. Eighty percent of the original patients participated in a 12month follow-up consisting only of self-report measures. Patients maintained their gains over pretreatment functioning. Three studies have compared cognitive restructuring plus exposure with pharmacologic treatments for social phobia. Two of those studies used Heimberg's CBGT. Gelernter et al22 compared CBGT with the monoamine oxidase inhibitor phenelzine, the triazolobenzodiazepine alprazolam, and pill placebo. Patients assigned to receive medication or placebo also were provided with self-exposure instructions. Patients in all four groups experienced significant improvement on all measures, with few differential effects. Patients treated with phenelzine were more improved than other patients on a measure of trait anxiety at the 2month follow-up, but similarities far outweighed differences. At the

SOCIAL PHOBIA

829

follow-up, patients receiving phenelzine or CBGT maintained their gains while alprazolam patients did not. Several methodologic concerns complicate interpretation of these results including sole reliance on self-report measures and the inclusion of self-exposure instructions in the medication and placebo conditions. Furthermore, although CBGT was based on the protocol of Heimberg (Heimberg RG: Cognitive-Behavioral Treatment of Social Phobia in a Group Setting: A Treatment Manual, ed 2. Unpublished manuscript, 1991), the number of patients in each group was increased from the recommended 6 to 10 patients. In a multisite collaborative study, Heimberg et al31 compared CBGT, phenelzine, pill placebo, and the attention-placebo treatment used by Heimberg et al.3° One hundred thirty-three patients were assigned randomly to one of the four treatments. Only patients judged to be responders to CBGT or phenelzine continued into the later maintenance and follow-up phases of the study. CBGT and phenelzine produced equivalent response rates, superior to the control conditions, after 12 weeks of treatment. Phenelzine was more effective than CBGT on some measures after 12 weeks. During the 6-month untreated follow-up, CBGT patients maintained their gains while a number of phenelzine patients relapsed. Detailed presentation of the results await the published report. The third study using pharmacologic treatments for comparison used a combination of cognitive restructuring, exposure, and relaxation in a brief format to treat 34 musicians with performance anxiety.15 The effect of adding buspirone to this treatment was examined, and patients were assigned randomly to one of four conditions: cognitive-behavioral treatment with buspirone, cognitive-behavioral treatment with placebo, buspirone alone, and placebo alone. Generally, patients treated with cognitive-behavioral therapy experienced improvement on several important outcome variables whereas buspirone was ineffective. Cognitivebehavior therapy plus placebo was the most effective treatment combination, and these patients outperformed all others on a measure of confidence in performing at a 1-month follow-up assessment. Component Analysis of Combined Cognitive Restructuring and Exposure

Well-controlled studies now exist that show that the combination of cognitive restructuring and exposure is a highly effective treatment for social phobia. What these studies do not tell us, however, is whether the combined treatment is more effective than therapies that use exposure in the absence of cognitive restructuring. In this section, we review several studies that address this question. Studies that support the utilization of combined cognitive restructuring and exposure are reviewed first. Butler et al1 3 compared exposure only with exposure plus anxiety management training. 69 The latter is composed of relaxation, distraction, and rational self-talk. Patients in the exposure-only group received a

830

JUSTER & HEIMBERG

filler treatment that included discussion about exposure and occupied the same amount of time as anxiety management training. After treatment, patients in both exposure conditions were superior to a wait-list group, having achieved reductions in phobic severity, anxiety during a behavior test, difficulty during social situations, general anxiety, and depression. Combined treatment patients achieved lower scores on two measures of social anxiety compared with patients receiving exposure only. These patients reported that they relied upon rational self-talk as their primary anxiety management technique. At the 6-month follow-up, combined treatment patients maintained and strengthened their advantage over exposure-only patients. Fully 40% of subjects receiving exposure only sought further treatment in the following 12 months whereas no combined treatment patient did so. Whereas anxiety management training is not a solely cognitive intervention, post hoc analyses conducted by Butler et al1 3 strongly suggest that rational self-talk was the most frequently used coping technique. Problems with procedures in the exposure condition are elaborated elsewhere. 40 Briefly, the total time spent on exposure and its procedures was severely limited. In addition, credibility ratings for the exposure condition were significantly lower than for the combined treatment, possibly due, in part, to the filler treatment. Noting the problems with the Butler et al1 3 study, Mattick and his colleagues 50, 51 conducted two studies examining exposure and cognitive techniques. The first study compared therapist-assisted exposure with and without cognitive restructuring. 50 No filler treatment was used so patients received equivalent amounts of active treatment. Both treatments yielded similar post-treatment outcomes, including reductions in avoidance, phobic severity, and depression. At the 3-month follow-up, patients in the combined treatment group completed a greater percentage of tasks during a behavioral test, evidenced lower avoidance, and higher scores on composite measures of improvement and end-state functioning, than patients treated with exposure alone. In their second study, Mattick et al 51 compared the same conditions as in the previous study, but added a cognitive restructuring only condition and a wait-list group. At post-treatment, patients in the three active treatments completed more tasks during a behavior test and scored lower on self-report measures of phobia severity and avoidance than did patients on the wait-list. Exposure and the combined treatment were more effective than cognitive restructuring alone in terms of the tasks completed during the behavior test. This pattern of effects changed, however, at the 3-month follow-up assessment. At follow-up, patients in the cognitive restructuring and combined treatment conditions continued to improve, whereas patients receiving exposure only appeared to deteriorate. Patients in the combined treatment condition experienced the greatest improvement on percentage of behavioral tasks completed. We now review several studies suggesting that cognitive restructuring may not enhance the outcome of exposure therapy. Biran et al1° examined the effect of adding cognitive therapy to in vivo exposure in

SOCIAL PHOBIA

831

a multiple-baseline-across-subjects design. Two of three women w ith fears of writing in public received five sessions of cognitive restructuring followed by five sessions of exposure while the third patient received exposure only. Exposure, conducted according to a standardized hierarchy of feared writing situations, produced significant increases in the number of writing tasks performed. Cognitive restructuring did not enhance the effectiveness of exposure. Reductions in fear ratings during performance of the exposure situations were less clearly tied to specific treatment. At the 9-month follow-up, avoidance of feared situations remained low. Fear in those situations, however, returned to pretreatment levels. Thus, a desynchrony existed between approach behavior and anxiety, a situation that might portend long-term relapse. This is only speculative, however, as no further follow-up was conducted. Two of the three patients sought additional treatment during follow-up, one for unspecified "social issues," the other for further anxiety reduction. All three indicated some disappointment that their anxiety reduction had not been maintained. This study has been criticized for several reasons.23• 28 Probably the most critical concern involves the total segregation of treatment components in a manner inconsistent with clinical practice and which may reduce treatment effectiveness.28 In a recently completed study (Taylor S, Woody S, Koch WJ, et al: Cognitive restructuring in the treatment of social phobia: Efficacy and modes of action. Submitted for publication, 1995), 65 patients were assigned randomly to receive eight sessions of cognitive restructuring in which exposure was purposely avoided or eight sessions of the filler treatment used by Butler et al. 13 Following an assessment, both groups received exposure with the expectation that subsequent response would be facilitated in the patients initially treated with cognitive restructuring. Cognitive restructuring was clearly more effective than the filler treatment after the initial phase of treatment. There was no facilitation of exposure therapy evident at the end of the second phase of treatment, however. Although only self-report measures were used at the 3-month follow-up, there were no differences between groups. Thus, patients who received the filler treatment were as improved as patients who received cognitive restructuring after both groups were treated with exposure. Hope et al36 compared exposure alone with the complete CBGT protocol to determine the relative importance of the cognitive component. Both conditions were more effective than a wait-list condition after 12 weeks of treatment. Patients treated with CBGT reported reduced anxiety during an individualized behavior test at post-treatment, but this was the only measure in which the full protocol outperformed exposure alone. Differences between exposure and CBGT at post-treatment, most of which favored exposure, disappeared at the 6-month follow-up, at which time the two treatments were equally effective. In this trial, however, CBGT was less effective than in other studies of this protocol.

832

JUSTER & HEIMBERG

Three additional studies have addressed the need for integrating cognitive techniques and exposure. In each case, the cognitive technique used was RET. Scholing and Emmelkamp64 treated 30 social phobic patients with primary fears of blushing, sweating, or trembling. Treatment was provided individually in two 4-week blocks separated by a 4week period of no treatment. Patients were assigned randomly to either in vivo exposure followed by RET, RET followed by in vivo exposure, or integrated RET and in vivo exposure. It is important to note that in the integrated treatment, exposures were never conducted during the session, so the two techniques were not actually administered together. Thus, the primary focus of the session was on changing maladaptive cognitions relative to situations that would be the focus of homework. Outcome was based on composite measures of avoidance of target situations, irrational cognitions, and somatic complaints. All treatments produced substantial improvements with no differences between them after the 3-month follow-up. Follow-up results were consistent with earlier assessments conducted after the first and second treatment blocks. Scholing and Emmelkamp65 conducted an essentially similar study with 73 generalized social phobic patients. Patients were assigned randomly to exposure only, exposure followed by RET, or integrated RET and exposure. Treatments were provided in blocks as in the previous study and included both individual and group modalities. Patients assigned to integrated treatment in the individual modality had no exposures during their sessions. Patients assigned to exposure or integrated RET and exposure in the group modality received exposure during sessions, although discussion of maladaptive cognitions was clearly the primary emphasis in the integrated treatment. Patients in the integrated treatment reported more somatic complaints than patients in the other two treatments after the first treatment block No significant differences were found between treatments after the second block or at the 3-month follow-up. Importantly, 37% of the exposure only and 30% of the RETfollowed-by-exposure patients required further treatment in the 15 months following treatment. Only 15% of the patients in the integrated treatment condition required further treatment. Although the authors reported no statistical analysis of this finding, it is similar to that reported by Butler et al.13 Mersch52 compared exposure alone with an integrated treatment combining in vivo exposure, RET, and social skills training. As in the individual treatments administered by Scholing and Emmelkamp,64• 65 no exposures were conducted during treatment sessions. Thirty-four patients were assigned randomly to either a wait-list control group or to one of the two treatment groups. Sixteen sessions of individual therapy were provided, and follow-up assessments were conducted at 3 months and 18 months. Both treatments were superior to the wait-list control at post-test, but there were no differences between exposure alone and the integrated treatment. At the 18-month follow-up, patients in both treatments achieved additional gains on self-report measures of social anxiety, but they still did not differ from each other.

SOCIAL PHOBIA

833

The results of studies examining the relative efficacy of exposure versus combined treatment are confusing, contradictory, and represent one of the critical issues in the treatment of social phobia. 24 Several studies do not support the need for the integration of exposure and cognitive restructuring10, 37, 52 ' 64' 65 whereas others suggest that combined treatment is superior and that additional gains are more likely during the follow-up period. 13, 5o, 5i An alternative method of examining the relative importance of treatment components is meta-analysis, a technique that compares average effect sizes for particular treatment components across studies. Two meta-analyses examining cognitive-behavioral treatment of social phobia have been conducted to date. In the first study by Feske and Chambless,18a cognitive-behavior therapy (cognitive restructuring plus exposure) and exposure alone were compared and appeared to produce equivalent effects on self-report measures from pretreatment to posttreatment and pretreatment to follow-up. A second meta-analysis by Taylor 70 a came to a different conclusion. Individual effect sizes for exposure, cognitive restructuring, combined exposure and cognitive restructuring, social skills training, placebo, and waiting list were calculated. The post-treatment effect sizes for exposure, cognitive restructuring, combined treatment and social skills training were significantly greater than the effect size for the wait-list control. The effect size for combined cognitive restructuring and exposure was the only one significantly greater than the effect size for placebo (the latter combined pill placebo and attention-placebo). Unfortunately, between-groups analyses were not conducted for follow-up assessments. Within-group analyses showed that the combined exposure plus cognitive restructuring effects remain stable during follow-up, whereas effects for the other treatments increased somewhat. The question of the importance of the integration of exposure and cognitive restructuring in the treatment of social phobia remains unresolved.

Social Skills Training

Marzillier et al49 compared social skills training with systematic desensitization for the treatment of 21 patients who appear to have met current criteria for generalized social phobia. Patients receiving social skills training increased their range of social activities and range of social contacts compared with patients on a waiting list. No other differences were found between treatment groups or between treatment groups and patients on the waiting list. Because of the numerous dropouts in the systematic desensitization group, only within-group analyses could be conducted at the 6-month follow-up . At that assessment, patients who received social skills training had maintained their increased range of social activities. Treatment did not result in improved social

834

JUSTER & HEIMBERG

skills or anxiety reduction, and most measures of clinical improvement showed no change. Trower et aF1 also compared social skills training and systematic desensitization, but unlike Marzillier et al,49 no wait-list or other control condition was included. Using a treatment-matching approach similar to that of Wlazlo et aF7 described earlier, patients were judged to have either demonstrable social skills deficits or to suffer from excessive anxiety. It was hypothesized that a match between treatment and specific patient deficits would provide greater benefit than a mismatch. Both patient groups experienced reductions in severity, social inadequacy, and general anxiety regardless of treatment received. The treatmentmatching hypothesis was supported partially by the finding that patients with skill deficits who received social skills training experienced less difficulty in social situations and a greater frequency of social activities. This pattern was also evident at a 6-month follow-up assessment. Notably, 80% of patients treated with systematic desensitization received additional treatment during follow-up compared with 40% of patients treated with social skills training. Shaw66 treated a group of 10 phobic patients with imaginal exposure and compared their outcome to that achieved by the phobic subgroup treated by Trower et al. 71 Patients treated with imaginal exposure achieved reductions in phobic severity, social inadequacy, and general anxiety similar to those achieved by patients treated with systematic desensitization or social skills training. Gains achieved by patients receiving imaginal exposure were maintained at a 6-month follow-up and remained similar to the other treatment conditions. Although these results are promising, the lack of a control group makes the results of Trower et aF1 and Shaw66 inconclusive. This is particularly true given the results of Marzillier et al,49 in which within-group changes were not reflected in differences between the control group and treatment groups. To determine if cognitive techniques would enhance the effectiveness of social skills training, Stravynski et al68 compared social skills training with and without a cognitive component (based on RET). Patients in both conditions reported increases in social interaction, less depression, and fewer irrational beliefs regarding social situations at post-treatment. Improvement was maintained in both conditions at the 6-month follow-up, with no apparent benefit from the addition of cognitive procedures. This study has been criticized23• 28 for lack of a control condition, inappropriateness of patient samples, and for administration of cognitive therapy in a less than ideal fashion. Mersch et al55 compared RET and social skills training using a treatment-matching paradigm. Patients were classified as behavioral or cognitive reactors based on differential responses to a role-play situation and to a self-report cognitive measure. One-half of the patients were judged to respond during the role-play situation primarily with overt behavioral signs of anxiety (behavioral reactors). The remainder of the patients scored relatively higher on a self-report measure of negative

SOCIAL PHOBIA

835

self-statements about social interactions (cognitive reactors). Patients then were assigned randomly to treatment conditions. At post-treatment, both groups of patients made significant gains regardless of the type of treatment received. Fourteen months after treatment, improvement was maintained across patient and treatment types.53 There were no differences between patients who received the matching treatment and patients who received the nonmatching treatment. Nearly 45% of patients received additional treatment during the follow-up period, suggesting that there was substantial room for further improvement. Patients who sought additional treatment were initially rated as less skilled and more anxious during a behavior test than patients who did not seek additional treatment. Follow-up assessments were limited to self-report measures despite the inclusion of a behavior test during the treatment phase of the study. Two studies that examined social skills training already have been discussed in previous sections of this article.67• 77 Briefly, in a single-case study, Stravynski67 found that the combination of social skills training, cognitive restructuring, and exposure was an effective treatment in reducing anxiety and psychogenic vomiting. This study provides little information about the utility of the social skills training component. Wlazlo et aF7 found that social skills training and exposure were equally effective treatments. One study compared social skills training with and without pharmacotherapy. Falloon et al1 7 treated 16 social phobic patients as defined in DSM-III with either social skills training plus propanolol or social skills training plus placebo. Social skills training was administered in two 6hour "workshops." Within-group changes included reductions in social anxiety and increases in positive self-image. Gains were maintained in the 81 % of patients who responded to the 6-month follow-up. Propanolol did not enhance the effects of social skills training at post-test or follow-up. Evidence for the long-term efficacy of social skills training as a treatment for social phobia is limited. Several studies53• 66• 68• 71• 77 suggested that social skills training has some efficacy, although not more so than other treatments. The only study that included an adequate control group,49 however, found few differences between social skills training and the wait-list. Thus, the lack of a wait-list condition in the other studies severely compromises their findings. The two studies with the longest follow-up period used only self-report measures to assess longterm effectiveness and suffered from other design problems.53• 77 The lack of methodologic rigor in these few studies makes it difficult to draw general conclusions about the long-term effectiveness of social skills training. Social skills training may be making a comeback, however, especially as part of multicomponent treatment packages. Here we report on two recent studies that present preliminary data on the efficacy of treatments that combine several components and that prominently feature social skills training. Feske et al19 used a combination of group

836

JUSTER & HEIMBERG

exposure, relaxation training, systematic desensitization, and social skills training in the treatment of social phobic patients with and without avoidant personality disorder. Both groups of patients improved from pretest to post-test and achieved further gains by the 3-month followup. Patients with avoidant personality disorder began and ended treatment more impaired and remained that way through follow-up, compared with patients without the personality disorder. These findings regarding avoidant personality disorder are consistent with the results reported in one study54 but contrast with those reported in two others.11· 37 In the latter two studies, avoidant personality disorder was not related to treatment outcome among patients with generalized social phobia. Social Effectiveness Therapy combines education, social skills training, imaginal exposure, and therapist-directed exposure. 74 Turner et aF4 used a broad set of assessments including self-report measures, a standardized behavior test, and clinician ratings as well as a composite index of outcome 76 in this uncontrolled study. After 29 sessions of treatment (conducted over 16 weeks), 84% of the 13 patients who completed treatment (59% of the full sample) were classified as having achieved either moderate or high end-state functioning. Turner et aF2 contacted patients 2 years after they completed treatment for follow-up assessment. Eight of the original 13 completed a selfreport battery and five completed a telephone interview. These patients maintained their gains or made some additional gains 21 to 24 months after the conclusion of treatment. SUMMARY

Interest in social phobia has increased dramatically in the past decade, and our knowledge of this previously understudied disorder has increased as well. We now know that social phobia is a chronic condition and that patients with this disorder are unlikely to experience significant improvement without intervention. It is also a highly prevalent condition affecting as many as 13% of the adult population of the Unites States. Although our understanding of the causes of social phobia remains limited, we do know that it is associated with serious impairment and disability in multiple spheres. Thus, the development of treatments with proven long-term efficacy is an important research goal. In this article, we have reviewed studies that examined either exposure, cognitive restructuring, social skills training, or some combination of these treatments. Here, we summarize the major findings of this review. Exposure has fared well as a treatment for social phobia and, in every case, within-group analyses show that patients have improved after treatment. Methodologic problems in some studies, however, limit the conclusions that can be drawn about the comparative efficacy of exposure, social skills training, and relaxation therapy. Conceptual models of social phobia have stressed the importance

SOCIAL PHOBIA

837

of cognitive processes in the development and maintenance of social phobia and much attention has been directed at the long-term efficacy of cognitive-behavioral techniques. It has been hypothesized that exposure plus cognitive restructuring would be a particularly effective combination and several methodologically sound studies have examined this combination. These studies have demonstrated consistently clinically significant within-group changes and superiority to control conditions. Heimberg's CBGT is probably the most widely studied of these treatments. CBGT has been shown to be more effective than an equally credible attention-placebo group. Patients receiving CBGT have maintained their advantage over patients in the attention-placebo group, even 5 years after treatment although flaws in that follow-up study limit generalizability of its results. Generalized and nongeneralized social phobic patients respond equivalently to this highly integrated treatment, and it has been applied effectively by researchers outside the center where it was developed. 22, 46 Despite the successes of combined exposure and cognitive restructuring treatments, it remains unclear as to what the effective component(s) of these and similar treatments are and, therefore, whether or not the integration of therapy components is really necessary. A number of the studies reviewed addressed this question with mixed results. Three studies showed that the combination therapy was superior to either treatment alone.13, 5 o, 5 1 There is also evidence that patients treated with exposure only may show some deterioration during follow-up 2 , 51 whereas patients treated with cognitive restructuring and exposure may continue to improve. 13' 5o, 51 Still, other studies52, 64' 65 found no differences in long-term outcome among exposure alone, RET, or the combined treatment. Hope et al36 found that exposure alone was as effective as an integrated treatment combining exposure and cognitive restructuring (CBGT), and Taylor and colleagues (submitted for publication, 1995) reported that exposure was not enhanced by initial treatment with cognitive restructuring. These results are disappointing in light of all that has been written about the likely benefits of combining cognitive and behavioral therapy in the treatment of social phobia. For example, it has been hypothesized that fear of negative evaluation is a key factor in social phobia12 and that change in this construct should be the goal of treatment. There is some research that supports this claim50, 51 and other evidence that suggests that exposure alone is not particularly effective in producing those changes. 13, 51 Butler12 concluded that the treatment of social phobia is made more difficult when therapy lacks a cognitive focus. For some researchers, the combination of exposure and cognitive restructuring makes intuitive sense and is consistent with recent models that emphasize cognitive processes in the anxiety disorders.32 If the fear of negative evaluation associated with social phobia is modulated by cognitive processes, one role of exposure may be to provoke a negative cognitive reaction, thereby increasing accessibility of these usually covert processes to cognitive restructuring procedures.

838

JUSTER & HEIMBERG

Evidence for the long-term efficacy of social skills training in the treatment of social phobia also is mixed. Patients treated with social skills training appear to fare well immediately following treatment and tend to maintain their gains during follow-ups ranging in length from 3 to 24 months. Closer inspection of the studies reviewed in this section, however, reveals that just one included a no-treatment control group,49 and that study found few differences between the active treatments and the wait-list condition. Despite this lack of well-controlled empirical support, interest may be renewed in social skills training as researchers look for effective treatments for the more difficult social phobia patients, especially those with generalized social phobia and Axis II disorders. Although treatment for these more difficult patients usually has resulted in improvement, these patients begin and end treatment more impaired. Two recent studies tested multicomponent treatment strategies with social skills training playing a prominent role. 19• 72 Although the results appear promising even 2 years after treatment, they are clearly preliminary. Further research on the long-term efficacy of treatments for social phobia is needed. This is especially true for the patients with more generalized social fears and those suffering with Axis II disorders. The solution may lie in the further integration of current treatment options, the refinement of current therapies, or in the development of new approaches to treating this disorder. The use of longer trials of acute treatment and long-term low-frequency maintenance treatment seems worth pursuing. Creative adjuncts to current therapeutic procedures that are already in practice but which require empirical validation include the systematic use of videotape feedback and modeling of feared outcomes. 20 What makes a good long-term outcome study? Although we have chosen to review all studies that included a follow-up regardless of its length, we clearly believe that longer is better. The average length of follow-up for all studies reviewed was approximately 10 months. This figure represents an important positive trend in social phobia outcome research, but it is probably skewed by at least one outlier, a follow-up study conducted 65 months34 after treatment. The majority of studies had follow-up periods of 6 months or less. Follow-up periods of less than 6 months are too brief. A follow-up period of 12 months or more would ensure that all acute effects of treatment, including nonspecific effects, have had the opportunity to dissipate. It would provide an extended time in which to assess further improvement as well as additional treatment received. Furthermore, 12 months provide an adequate amount of time to study the post-treatment course of a disorder that is chronic, difficult to treat, and may h ave a high likelihood of relapse. In addition to length of follow-up, greater attention needs to be paid to follow-up assessment procedures. Assessments at follow-up often have taken a secondary role to the more comprehensive assessments conducted before and immediately after treatment. Follow-up assessments often include fewer and less diverse measures, despite the fact that follow-up is arguably the most important assessment period.

SOCIAL PHOBIA

839

Many studies reviewed used clinician ratings, behavioral tests, and self-report ratings during the acute phase but utilized only self-report measures at follow-up . Given that different assessment modalities may have different sensitivities to changes in treatment, it makes sense that follow-up assessments be at least as broadly based as acute assessments. It also is important to assess and report the extent and nature of additional treatment sought by patients during the follow-up period. In the few studies that examined this variable, differences between various treatments often were revealed. Finally, follow-up assessments often are conducted with a very limited subsample of the original group that participated in acute treatment. We need to be more creative at finding and obtaining the cooperation of patients at these later assessments when interest in treatment may have diminished. More frequent contact may help maintain patients' interest and increase the likelihood of their long-term cooperation. ACKNOWLEDGMENT The authors thank Karen Law for her invaluable assistance in the preparation of this manuscript.

References 1. Al-Kubaisy T, Marks IM, Logsdail S, et al: Role of exposure homework in phobia reduction: A controlled study. Behavior Therapy 23:599-621, 1992 2. Alstrom J, Nordlund C, Persson G, et al: Effects of four treatment methods on social phobic patients not suitable for insight-oriented psychotherapy. Acta Psychiatr Scand 70:97-110, 1984 3. American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, ed 3. Washington, American Psychiatric Association, 1980 4. American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, ed 3-rev. Washington, American Psychiatric Association, 1987 5. American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, ed 4. Washington, American Psychiatric Association, 1994 6. Amies PL, Gelder MG, Shaw PM: Social phobia: A comparative clinical study. Br J Psychiatry 142:174- 179, 1983 7. Barlow DH: Anxiety and Its Disorders: The Nature and Treatment of Anxiety and Panic. New York, Guilford Press, 1988 8. Beck AT: Cognitive Therapy and the Emotional Disorders. New York, International Universities Press, 1976 9. Biederman J, Rosenbaum JF, Hirshfeld DR, et al: Psychiatric correlates of behavioral inhibition in young children of parents with and without psychiatric disorders. Arch Gen Psychiatry 47:21-26, 1990 10. Biran M, Augusto F, Wilson GT: In vivo exposure vs. cognitive restructuring in the treatment of scriptophobia. Behav Res Ther 19:525-532, 1981 11. Brown EJ, Heimberg RG, Juster HR: Social phobia subtype and avoidant personality disorder: Effect on severity of social phobia, impairment, and outcome of cognitivebehavioral treatment. Behavior Therapy 1995 12. Butler G: Exposure as a treatment for social phobia: Some instructive difficulties. Behav Res Ther 23:651-657, 1985 13. Butler G, Cullington A, Munby M, et al: Exposure and anxiety management in the treatment of social phobia. J Consult Clin Psycho! 52:642-650, 1984

840

JUSTER & HEIMBERG

14. Caspi A, Elder CH, Bern DJ: Moving away from the world: Life-course patterns of shy children. Developmental Psychology 24:824-831, 1988 15. Clark DB, Agras WS: The assessment and treatment of performance anxiety in musicians. Am J Psychiatry 148:598-605, 1991 16. Ellis A: Reason and Emotion in Psychotherapy. New York, Lyle Stuart, 1962 17. Falloon IRH, Lloyd GG, Harpin RE: The treatment of social phobia: Real-life rehearsal with nonprofessional therapists. J Nerv Ment Dis 169:180-184, 1981 18. Fava GA, Grandi S, Canestrari R: Treatment of social phobia by homework exposure. Psychother Psychosom 52:209-213, 1989 18a. Feske U, Chambless DL: Cognitive-behavioral versus exposure treatment for social phobia: A meta-analysis. Behavior Therapy, in press 19. Feske U, Perry KJ, Chambless DL, et al: Avoidant personality disorder as a predictor for severity and treatment outcome among generalized social phobics. Journal of Personality Disorders, 1995, in press 20. Flynn T, Pollard CA, Carmin C, et al: Exposure to taped enactments of feared social catastrophe: Successful treatment of four previously refractory cases of social anxiety. Poster presented at the annual meeting of the Association for Advancement of Behavior Therapy, San Diego, November, 1994 21. Fyer AJ, Mannuzza S, Chapman TF, et al: A direct interview family study of social phobia. Arch Gen Psychiatry 50:286-293, 1993 22. Gelernter CS, Uhde TW, Cimbolic P, et al: Cognitive-behavioral and pharmacological treatments for social phobia: A controlled study. Arch Gen Psychiatry 48:938- 945, 1991 23. Heimberg RC: Cognitive and behavioral treatments for social phobia: A critical analysis. Clin Psychol Rev 9:107-128, 1989 24. Heimberg RC: Specific issues in the cognitive-behavioral treatment of social phobia. J Clin Psychiatry 54[suppl 12]:36-45, 1993 25. Heimberg RC, Barlow DH: Psychosocial treatments for social phobia. Psychosomatics 29:27-37, 1988 26. Heimberg RC, Barlow DH: New developments in cognitive-behavioral therapy for social phobia. J Clin Psychiatry 52(suppl 11):21-30, 1991 27. Heimberg RC, Juster HR: Cognitive-behavioral treatment: Literature review. In Heimberg RC, Liebowitz MR, Hope DA, et al (eds): Social Phobia: Diagnosis, Assessment and Treatment. New York, Guilford Press, 1995 28. Heimberg RC, Dodge CS, Becker RE: Social phobia. In Michelson L, Ascher M (eds): Cognitive Behavioral Assessment and Treatment of Anxiety Disorders. New York, Plenum Press, 1987, pp 280-309 29. Heimberg RC, Becker RE, Goldfinger K, et al: Treatment of social phobia by exposure, cognitive restructuring, and homework assignments. J Nerv Mental Dis 173:236- 245, 1985 30. Heimberg RC, Dodge CS, Hope DA, et al: Cognitive behavioral treatment of social phobia: Comparison to a credible placebo control. Cognitive Therapy and Research 14:1- 23, 1990 31. Heimberg RC, Juster HJ, Brown EJ, et al: Cognitive-behavioral versus pharmacological treatment of social phobia: Posttreatment and follow-up effects. Poster presented at the annual meeting of the Association for Advancement of Behavior Therapy, San Diego, November, 1994 32. Heimberg RC, Juster HR, Hope DA, et al: Cognitive behavioral group treatment for social phobia: Description, case presentation and empirical support. In Stein MB (ed): Social Phobia: Clinical and Research Perspectives. Washington, American Psychiatric Press, 1995 33. Heimberg RC, Liebowitz MR, Hope DA, et al (eds): Social Phobia: Diagnosis, Assessment and Treatment. New York, Guilford Press, 1995 34. Heimberg RC, Salzman DC, Holt CS, et al: Cognitive behavioral group treatment for social phobia: Effectiveness at five-year follow-up. Cognitive Therapy and Research 17:325-339, 1993 35. Hope DA, Heimberg RC: Social phobia and social anxiety. In Barlow DH (ed): Clinical

SOCIAL PHOBIA

36. 37. 38. 39. 40.

41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57.

841

Handbook of Psychological Disorders: A Step-By-Step Treatment Manual, ed 2. New York, Guilford Press, 1993, pp 99-136 Hope DA, Heimberg RG, Bruch MA: Dismantling cognitive-behavioral group therapy for social phobia. Behav Res Ther 33:637-650, 1995 Hope DA, Herbert JD, White C: Diagnostic subtype, avoidant personality disorder, and efficacy of cognitive behavioral group therapy for social phobia. Cognitive Therapy and Research 19:319-417, 1995 Hope DA, Holt CS, Heimberg RG: Social phobia. In Giles TR (ed): Handbook of Effective Psychotherapy. New York, Plenum Press, 1993, pp 227-251 Juster HR, Brown EJ, Heimberg RG: Social phobia. In Margraf J (ed): Textbook of Behavior Therapy. Berlin, Springer-Verlag, 1995 Juster HR, Heimberg RG, Holt CS: Social phobia: Diagnostic issues and review of cognitive-behavioral treatment strategies. In Hersen M, Eisler R, Miller P (eds): Progress in Behavior Modification. Sycamore, IL, Sycamore Publishing Company, 30:74-98, 1995 Kagan J, Reznick JS, Snidman N: Biological bases of childhood shyness. Science 240:167-173, 1988 Kanter N, Goldfried M: Relative effectiveness of rational restructuring and selfcontrol desensitization in the reduction of interpersonal anxiety. Behavior Therapy, 10:472-490, 1979 Kessler RC, McGonagle KA, Zhao S, et al: Lifetime and 12-month prevalence of DSMIII-R psychiatric disorders in the United States. Arch Gen Psychiatry 51:8-19, 1994 Liebowitz MR, Gorman JM, Fyer AJ, et al: Social phobia: Review of a neglected anxiety disorder. Arch Gen Psychiatry 42:729-736, 1985 Liebowitz MR, Schneier F, Campeas, R, et al: Phenelzine vs. atenolol in social phobia: A placebo-controlled comparison. Arch Gen Psychiatry 49:290-300, 1992 Lucas RA, Telch MJ: Group versus individual treatment of social phobia. Poster presented at the annual meeting of the Association for Advancement of Behavior Therapy, Atlanta, November, 1993 Mannuzza S, Schneier FR, Chapman TF, et al: Generalized social phobia: Reliability and validity. Arch Gen Psychiatry 52:230-237, 1995 Marks IM, Gelder MG: Different ages of onset in varieties of phobia. Am J Psychiatry 123:218- 221, 1966 Marzillier JS, Lambert C, Kellet J: A controlled evaluation of systematic desensitization and social skills training for socially inadequate psychiatric patients. Behav Res Ther 14:225-238, 1976 Mattick RP, Peters L: Treatment of severe social phobia: Effects of guided exposure with and without cognitive restructuring. J Consult Clin Psycho! 56:251-260, 1988 Mattick RP, Peters L, Clarke JC: Exposure and cognitive restructuring for social phobia: A controlled study. Behavior Therapy 20:3-23, 1989 Mersch PPA: The treatment of social phobia: The differential effectiveness of exposure in vivo and an integration of exposure in vivo, rational emotive therapy and social skills training. Behav Res Ther 33:259-269, 1995 Mersch PPA, Emmelkamp PMG, Lips C: Social phobia: Individual response patterns and the long-term effects of behavioral and cognitive interventions: A follow-up study. Behav Res Ther 29:357-362, 1991 Mersch PPA, Jansen MA, Arntz A: Social phobia and personality disorder: Severity of complaint and treatment effectiveness. Journal of Personality Disorders, 1995, in press Mersch PPA, Emmelkamp PMG, Bagels SM, et al: Social phobia: Individual response patterns and the effects of behavioral and cognitive interventions. Behav Res Ther 27:421-434, 1989 Mersch PPA, Hildebrand M, Lavy EH, et al: Somatic symptoms in social phobia: A treatment method based on rational emotive therapy and paradoxical interventions. J Behav Ther Exp Psychiatry 23:199-211, 1992 Reich J, Goldenberg I, Goisman R, et al: A prospective, follow-along study of the course of social phobia: II. Testing for basic predictors of course. J Nerv Ment Dis 182:297- 301, 1994

842

JUSTER & HEIMBERG

58. Reich J, Goldenberg I, Vasile R, et al: A prospective follow-along study of the course of social phobia. Psychiatry Res 54:249-258, 1994 59. Rosenbaum JF, Biederman J, Hirshfeld DR, et al: Further evidence of an association between behavioral inhibition and anxiety disorders: Results from a family study of children from a non-clinical sample. J Psychiatr Res 25:49- 65, 1991 60. Rosenbaum JF, Biederman J, Pollock RA, et al: The etiology of social phobia. J Clin Psychiatry 55(suppl 6):10-16, 1994 61. Sanderson WC, DiNardo PA, Rapee RM, et al: Syndrome comorbidity in patients diagnosed with a DSM-III-R anxiety disorder. J Abnorm Psycho! 99:308-312, 1990 62. Schneier FR, Liebowitz MR, Garfinkel R, et al: Functional impairment in social phobia. J Clin Psychiatry 55:322-331, 1994 63. Schneier FR, Johnson J, Hornig CD, et al: Social phobia: Comorbidity and morbidity in an epidemiologic sample. Arch Gen Psychiatry 49:282- 288, 1992 64. Scholing A, Emmelkamp PMG: Cognitive and behavioural treatments of fear of blushing, sweating or trembling. Behav Res Ther 31:155-170, 1993 65. Scholing A, Emmelkamp PMG: Exposure with and without cognitive therapy for generalized social phobia: Effects of individual and group treatment. Behav Res Ther 31:667-681, 1993 66. Shaw PM: A comparison of three behaviour therapies in the treatment of social phobia. Br J Psychiatry 134:620-623, 1979 67. Stravynski A: Behavioral treatment of psychogenic vomiting in the context of social phobia. J Nerv Ment Dis 171:448-451, 1983 68. Stravynski A, Marks I, Yule W: Social skills problems in neurotic outpatients: Social skills training with and without cognitive modification. Arch Gen Psychiatry 39:13781385, 1982 69. Suinn RM, Richardson F: Anxiety management training: A nonspecific behavior therapy program for anxiety control. Behavior Therapy 2:498-510, 1971 70. Taylor CB, Arnow B: The Nature and Treatment of Anxiety Disorders. New York, McMillan, 1988 70a. Taylor S: Meta-analysis of cognitive-behavioral treatments for social phobia. J Beh Ther Exp Psychiatry, in press 71. Trower P, Yardley K, Bryant B, et al: The treatment of social failure: A comparison of anxiety-reduction and skills acquisition procedures on two social problems. Behavior Modification 2:41- 60, 1978 72. Turner SM, Beidel DC, Cooley-Quille MR: Two-year follow-up of social phobics treated w ith Social Effectiveness Therapy. Behav Res Ther 33:553-555, 1995 73. Turner SM, Beidel DC, Jacob RG: Social phobia: A comparison of behavior therapy and atenolol. J Consult Clin Psycho! 62:350- 358, 1994 74. Turner SM, Beidel DC, Cooley MR, et al: A multicomponent behavioral treatment for social phobia: Social Effectiveness Therapy. Behav Res Ther 32:381-390, 1994 75. Turner SM, Beidel DC, Dancu CV, et al: Psychopathology of social phobia and comparison to avoidant personality disorder. J Abnorm Psycho! 95:389-394, 1986 76. Turner SM, Beidel DC, Long PJ, et al: A composite m easure to determine the functional status of treated social phobics: The Social Phobia Endstate Functioning Index. Behavior Therapy 24:265-275, 1993 77. Wlazlo Z, Schroeder-Hartwig K, Hand I, et al: Exposure in vivo vs social skills training for social phobia: Long-term outcome and differential effects. Behav Res Ther 28:181-193, 1990

Address reprint requests to Harlan R. Juster, PhD Center for Stress and Anxiety Disorders Pine West Plaza Building No. 4 Albany, NY 12205