Cognitive-Behavioral Group Therapy for Social Phobia in Female Adolescents: Results of a Pilot Study

Cognitive-Behavioral Group Therapy for Social Phobia in Female Adolescents: Results of a Pilot Study CHRIS HAYWARD, M.D., M.P.H., SUSAN VARADY, M.A., ANNE MARIE ALBANO, PH.D., MARGO THIENEMANN, M.D., LYNNE HENDERSON, PH.D., AND ALAN F. SCHATZBERG,M.D.

ABSTRACT Objectives: To examine the efficacy of cognitive-behavioralgroup therapy for adolescents (CBGT-A) in females with social phobia and the effect of this treatment on the risk for major depression. Method: Female adolescents with social phobia ( N = 35) were randomly assigned to treatment (n = 12) or no treatment (n= 23) groups. Assessments were conducted at baseline, after treatment, and at a 1-year follow-up. Results: Eleven subjects completed treatment. Sixteen weeks of treatment produced a significant improvement in interference and reduction in symptoms of social anxiety.There was a significant reduction in the number of subjects meeting DSM-IVcriteria for social phobia in the CBGT-A versus the untreated group; however, at the 1-year follow-up there were no significant differences by treatment condition. There was also suggestive evidence that treatment of social phobia lowers the risk for relapse of major depression among those with a history of major depression. Combining social phobia and major depression as the outcome produced more robust treatment effects in the 1-year follow-up. Conclusions: This pilot study provides evidence for a moderate short-term effect of CBGT-A for treating female adolescents suffering from social phobia and indicates that treatment of social phobia may result in a reduction of major depression. J. Am. Aced. ChiMAdoIesc. Psychiatry 2OOO,39(6):721-726. Key-:

social

phobia, adolescents, female, cognitive-behavioral group therapy, major depression.

Social phobia is becoming increasingly recognized as an important disorder among adolescents (Beidel and Turner, 1998). In population-based studies of adults, the reported peak age of onset for social phobia occurs during adolescence (Burke et al., 1990). Prevalence rates of social phobia in adolescents of high school age range from 5% to 10% (Hayward et al., 1998; Lewinsohn et al., 1993; Wittchen et al., 1999). Social phobia in adolescence is associated with significant impairment, including poor school achievement, difficulties with inti-

Arcqtrd December 14. 1999. From the Dqartmrnt ofI3ychiany and BehaviornlScimer (DK. Haywad Thienemann. and&hatzberg) and the D q a m r n t of Aychlogy (DK Hordmon. Vuitinx&h&r), Sranfird UniversitysStanfim! CA; the D q a m r n t of nyrhOlogy# George Mason University, Faif& VA (Mr. Vardy); and the Dqamnrnt of I3ychorogY. New York Universiv, New York (Dr. Albano). Thrr rcyMh w mrrdcpossibh bygmnnfmm the h f i r d Gnm a Adolarrnrr. the WT Grant FounLtion Faculty &ho&n Award (OK Hapard), and the Pritzkcr Comom'um (Dx &hateberg). Reprint requests to DK Hayward D q a m r n r of Aychiany and Behavioral St-irnrrs, Room 1316, Stanfird University. Stanfird CA 94305-5722; c-mail. Hayward@Lr&d Stanfim! Edu. 0890-8567/oO/3906-0721Q2000 by the Amcricvl Academy of Child and Adolcxcnt Psychiatry.

mate relationships, and secondary alcohol use (Albano et al., 1995a; Scholing and Emmelkamp, 1990). Social phobia during adolescence often persists into adulthood. Furthermore, there is a growing body of literature concerning the comorbidity and relationship between social phobia and major depression. In population-based studies of adults, social phobia is the anxiety disorder that is most comorbid with major depression (Regier et al., 1998). In most cases in which the 2 disorders are comorbid, social phobia develops prior to the onset of major depression (Kessler et al., 1999; Schatzberg et al., 1998). Thus, there has been speculation that social phobia is either a risk factor for the onset of major depression or possibly an early manifestation of major depression during adolescence (Schatzberget al., 1998). In spite of the growing recognition that social phobia is common in adolescence and is associated with significant impairment, including increased risk for major depression, there are very few studies of the relationship between social phobia and major depression in adolescent samples, and even fewer treatment outcome studies of adolescents with social phobia (Ollendick and King, 1998). To our knowledge, there is only one published

J , AM, A C A D . C H I L D ADOLESC. PSYCHIATRY, 39:6.J U N E 2000

72 1

HAYWARD ET AL.

study of cognitive-behavioral group therapy (CBGT) for social phobia in adolescents (Albano et al., 1995b). In this study Albano and Barlow (1996) modified the CBGT for adults (Heimberg et al., 1990, 1993) to take into account the developmental concerns of adolescents. In addition to cognitive restructuring and exposure, training in social skills and social problem-solving were added to the adult version of CBGT. Results of this pilot investigation of the cognitive-behavioral group therapy-adolescent (CBGT-A) protocol demonstrated that social phobia symptoms had decreased to subclinical levels at a 3-month follow-up for 4 of 5 adolescents. At a 1-year follow-up, 4 adolescents were in complete remission and the 5th adolescent reported subclinical symptoms only. The CBGT-A protocol has yet to be evaluated in comparison with an untreated condition. Furthermore, although Albano et al. (1995b) reported a decrease in depressive symptoms after CBGT-A, there is no information about the long-term course of major depressive disorder following treatment of social phobia in adolescents. Therefore, the purpose of this study was to evaluate CBGT-A for social phobia in female adolexents and to compare it with an untreated control condition. A secondary purpose of this study was to determine whether treatment of social phobia was associated with a lower rate of subsequent major depression in those who received treatment versus those who did not. Because of the interest in the effect of treatment on subsequent major depression, the sample was restricted to females, inasmuch as base rates for major depression onset are higher in female than in male adolescents. METHOD Subjects

measures cognitive, somatic, and behavioral responses to potentially fear-producing social situations and asscss~the degree of distress or impairment resulting from these experiences. We used the Social Phobia subscale in this study (Turner et al., 1989). Clark et al. (1994) reported high internal consistency (Cronbach a = .97), as well as reasonable construct and concurrent validity, for the Social Phobia subscale in an adolacent sample. In this sample the Cronbach a for the Social Phobia subscalc was .98. ADirordm Interview Shedub, Child and Pamt Vmiom. The Anxiety Disorders Interview Schedule (ADIS), Child (Silverman and Albano, 1996a) and Parent versions (Silverman and Albano, 1996b), are structured interview schedules for the diagnosis of childhood and adolescent anxiety disorders. The ADIS-IV-(ChildParent version), which uscs DSM-Ncriteria, was used both to determine diagnoses at baseline and as an outcome measure. The diagnosis of social phobia using the ADIS has been shown to have good test-retest reliability (K= 0.73) (Silverman, 1991). The ADIS interference rating scale was the primary outcome measure.This scale is rated from 0 to 8 and is used to assess the impact of social phobia symptoms on the domains of friends, school, and home. A score of 4 or greater is necessary for a diagnosis of social phobia. The diagnosis of major depression was determined using the ADIS and based on DSM-Ncritcria. Diagnostic interviews were conducted by graduate students in clini d psychology who completed a period of training with one of the originators of the interview 0JZI.K. Silverman). At the posttreatment and 1-year follow-up assessments, interviewers were kept blind to information regarding treatment status. Assignment to Treatment and Follow-up After the baseline evaluation, subjects were randomly assigned to CBGT-A or to an untreated condition. Twelve subjects were recruited for each randomization, with 6 subjects randomly assigned to the CBGT-A condition and 6 to an untreated condition. After 2 treatment groups were completed, a third set of 11 subjects were included in the untreated condition. Thus, the final assignment included 12 subjects for the CBGT-A condition and 23 subjects for the untreated condition. One treatment group was co-led by a clinical psychologist and a research assistant; a second p u p was co-led by a child psychiatrist and a research assistant. Social-phobic subjects were evaluated at basdine, after treatment, and 1 year after the treatment. On average, 5 months dapsed between the pretreatment and posttreatment asscssmcnts and 12.9 months between posmcatment and the 1-year follow-up. Length of time between assessments was comparable for the CBGT-A and untreated subjects. The non-socially phobic comparison group was assessed only at baseline and at 18 months. Subjects in the untreated condition were not contacted during the period between baseline and posttreatment. Inquiries at posttreatment indicated that none of the untreated subjects received treatment between the baseline and posttreatment, preserving the integrity of the untreated condition. However, 4 of the untreated subjects received treatment in the community between posttreatment and the 1-year follow-up. Three of these 4 received combined pharmacotherapy and psychotherapy, and 1 received pharmacothcrapy alone.

Subjects in this study were 35 socially phobic female adolacents (mean age = 15.8 i 1.6 years) and 18 non-socially phobic comparison subjects (mean age = 15.2 i 1.4 years). Both groups were recruited through local advertisements. The socially phobic subjects were required to meet diagnostic criteria for DSM-IV social phobia. Subjects were excluded if they currently had major depression; if they had a current or previous history of panic disorder, agoraphobia, substance abuse, or psychotic disorder; or if they were using a psychotropic medication. Notably, subjects who had a previous history of major depression were not excluded. Subjects in the non-socially phobic comparison group were as0 screened for Axis I disorders and.&e same occlusion criteria were applied. All procedures were approved by the Stanford University In6;utional &view Board. All ;;bjects &d at least one parent for each subject provided written informed consent.

us (A.M.A.), who trained the group leaders in the delivery of this pro-

Measures Social Phobia a n d A m i q Inventory. The Social Phobia and Anxiety Inventory (SPAI) is an empirically derived self-report instrument that

tocol. Each session was approximately 1.5 hours in duration. Sessions 1 and 2 focused on providing group members with information about social anxiety and the rationale provided for treatment. Sessions 3 to 8 involved the introduction of skill-building, including social skills,

722

Cognitive-Behaviora'GroupTheraPY-Ado'escent The 16-week CBGT-A protocol we used was developed by one of

J . A M . A C A D . C H I L D ADOLESC. PSYCHIATRY. 3 9 : 6 ,

J U N E ZOO0

T R E A T M E N T OF SOCIAL PHOBIA I N ADOLESCENTS

social problem-solving skills, assertiveness, and cognitive restructuring. Sessions 9 through 15 involved in vivo and simulated within-session exposure to feared social situations. Each group member worked through a hierarchy of feared social situations. During these cxposures, group members were encouraged to apply coping strategies reviewed in previous meetings. There were homework assignments for between-session in vivo exposures. Session 16 consisted of a final exposure, discussion of termination, and plans for follow-up. This treatment protocol did not include parental involvement.

RESULTS

Pretreatment differences across treatment conditions were examined with x2 tests and I tests. Variables assessed included age, previous history of major depression, and scores on the interference from the ADIS and on the SPAI. There were no significant pretreatment differences across conditions. One subject assigned to CBGT-A did not complete treatment. Excluding this subject, attendance for the 16 weeks of group treatment was 8 1% averaged across both groups. One subject in the untreated group was lost to follow-up before the posttreatment assessment. Between the posttreatment assessment and the 1-year follow-up, 1 subject in the treatment group and 4 subjects in the untreated group were lost to follow-up. Posttreatment Outcome

The effects of treatment were analyzed with the xztest or Fisher exact test for dichotomous variables and andysis of covariance for continuous measures. Pretreatment scores served as the covariate. The diagnoses of social phobia or major depression were given if either the subject

TABLE 1 Means and Standard Deviations for the Anxiety Disorders Interview Schedule Interference Rating (0-8) by Treatment Condition (Child and Parent) Pretreatment Posttreatment Child CBGT-A Untreated Parent CBGT-A Untreated

F

pValue

5.2 e2.0 5.1 * 1.7

2.7* 1.7 4.8 * 1.7

10.9 (1,27)

.003

6.8 6.5

3.5 4.8

* 2.5

5.2 (1,26)

.031

1.0

* 0.9

* 1.6

Now CBGT-A, n = 11; untreated, n = 22. CBGT-A = cognitivebehavioral group therapy-adolescent.

or her parent's report was positive on the ADIS. The primary criterion for efficacy at posttreatment was the rating of interference from the ADIS-IV-(Child/Parent version). As shown in Table 1, there were significant reductions in interference in the CBGT-A condition compared with the untreated condition on both the parent and child interferenceratings. O n the 0 to 8 interfirence score, mean scores on ratings decreased nearly 50% in both the child interviews (5.2 2.0 to 2.7 1.7) and parent interviews (6.8 1.0 to 3.5 2.5). There were also significant differences by treatment condition in the SPAI at posttreatment, adjusted for baseline differences (Table 2). It is important to note that the mean scores on the SPAI in the CBGT-A condition at posttreatment were still quite elevated, suggesting continued symptoms (92.8 30). In comparison, the mean baseline scores in the non-socially phobic comparison group were 37.8 18.8.

*

*

*

*

*

*

TABLE 2 Means and Standard Deviations for Social Phobia and Anxiety Inventory (SPAI) Social Phobia Subscale by Treatment Condition Mean

SPAI CBGT-A Untreated

* SD

Pretreatment

Posttreatment

126.1 * 29.7 114.5 * 42.5

92.8 104.3

Pretreatment

SPAI CBGT-A Untreated

* *

126.1 29.7 114.5 42.5 Pretreatment

Non-socially phobic comparison group

Mean * SD

37.9

* 18.8

* 30.8 * 43.3

F

p Value

4.2 (1,31)

.048

0.7 (1,26)

.4

1-Year Follow-up" 96.4 99.2

2

34.6 41.8

1-Year Follow-up" 40.4

* 17.9

Note: CBGT-A, n = 11; untreated, n = 22; Non-socially phobic comparison group, n = 18. CBGT-A = cognitive-behavioral group therapy-adolescent. a At the 1-year follow-up the sample was n = 10 for CBGT-A, n = 18 for the untreated, and n = 16 for the non-socially phobic comparison group.

J . AM. ACAD. C H I L D ADOLESC. PSYCHIATRY. 3936, J U N E 2000

723

HAYWARD ET AL.

As shown in Figure 1, at posttreatment 55% (6/11) of subjects in the treatment condition still met criteria for social phobia on the ADIS, whereas 96% (21/22) of subjects in the untreated condition met criteria for social phobia ( p < .05 on the Fisher exact test).

n

70

One-Year Follow-up

To evaluate the course of illness, subjects were reassessed 1 year after the posttreatment evaluation. Interviewers remained blind to any data regarding'the treatment condition. As shown in Figure 1, there was no significant difference in the frequency of social phobia by condition at 1-year follow-up. In addition, there was no difference in mean scores on the SPAI after 1 year, with adjustment for pretreatment scores (Table 2).

Occurrence of major depression was also compared between the 2 groups. Since some subjects had a previous history of major depression, onset represented both onset of a new episode as well as relapse of major depression. At posttreatment 41% (9/22) of subjects in the untreated condition met criteria for major depression in contrast to only 18% (2/11) of subjects in the treatment condition. This difference was not statistically significant. At the 1year follow-up, 33% (6/18) of untreated subjects demonstrated major depression compared with 20% (2/10) of treated subjects. This difference was also not statistically significant. We then evaluated the effect of treatment on depression relapse among those with a prior history of depression (Fig. 2). The time period for risk of depression was

I

1

Post-Tmatmed 1-YearFollow-up

Flg. 1 Pcrccntagc with social phobia by trutmcnt condition at prctrcatmcnt (CBGT-A, n = 12; untreated, n = 23), posnrutmcnt (CBGT-A, n = 11; untrcatcd, n = 22). and 1-year follow-up (CBGT-A, n = 10; untreated, n = 18). 'Uxs combined parcnt and child diagnostic intcrvicws; positive on cithcr p i t i v c . 'p c .05. Fishcr cxact t a t . CBGT-A = cognitivc-bchavioral group therapy-adolaccnt.

-

724

0

-CBGT-A

U-sd

past History

of YD

CBGT-A

Untrsaad NolFsocbl~

NoHiatory of MD

Phobic Comparison -P

Fig. 2 Occurrcncc of M D bctwccn prctrcatmcnt and 1-year follow-up by trcatmcnt condition and prcvious history of major dcprcssion. Previous history of MD: CBGT-A (n = 6). untrcatcd (n = 11); no history of MD: CBGT-A (n = 5 ) , untrutcd (n = 11); non-socially phobic comparison group (n = 18). CBGT-A = cognitive-bchaviord group therapy-adolcxcnt; M D major dcprcssion.

-

Major Depression

RaTreatment

10

collapsed to include a diagnosis of depression at either the 16-week or 1-year follow-up. There was a suggestion that relapse of major depression was higher among untreated subjects versus the CBGT-A condition: 64% relapse in the untreated condition versus 17% in the CBGT-A condition. This difference was not statistically significant; however, the power to detect significancewas low because of small sample sizes (Fig. 2). These findings raise the possibility that treatment of social phobia may reduce the risk of relapse for major depression. They also support the view that social phobia and major depression are closely linked in some adolescents. Combining Social Phobia and Major Depression as Outcomes

Because of the overlap between major depression and social phobia, we combined major depression and social phobia as a single outcome to evaluate treatment effects (Fig. 3).At posttreatment all subjects in the untreated condition had either major depression or social phobia on the ADIS, whereas 55% of those in the treatment condition had either major depression or social phobia ( p < .05 on the Fisher exact test). After 1 year, 78% of those in the untreated condition had either major depression or social phobia versus 40% in the treatment condition ( p = .09 on the Fisher exact test). Thus, much of the improvement in the CBGT-A group compared with the untreated group was maintained in the 1-year follow-up when social phobia and major depression were combined as the outcome.

J . A M . ACAD. CHILD ADOLESC. PSYCHIATRY, 3 9 : 6 . J U N E ZOO0

TREATMENT OF SOCIAL PHOBIA IN ADOLESCENTS

q78

i!

40

P 40 60[

3 .B

20

0 Untreated

OCBGFA

DISCUSSION

The results provide support for a moderate short-term effect of CBGT-A for the treatment of social phobia in female adolescents. There were considerable residual social-phobic symptoms at posttreatment in spite of statistically significant improvement. Also, the follow-up evaluation suggests that differences between subjects receiving CBGT-A and those in the untreated condition were not present 1 year later, when social phobia was the primary outcome. In the only other study reporting results from CBGT-A .for treatment of social phobia in adolescents,Albano et al. (1995b) reported both short-term gains in social phobia symptoms at 3-month follow-up, as well as long-term gains at 12-month follow-up. It is not clear why the same protocol would yield less robust results in the study reported here; however, in the study by Albano et al. (1995b) there was no comparison group and the sample was very small. In addition, parental involvement was included in the Albano et al. (1995b) study, which may have moderated the effects. One other issue to consider is that only females were enrolled in the current study, whereas a mixed group of males and females was investigated in the original study. It is possible that heterosocial anxiety, which was not a direct target of treatment in this protocol, may have been a variable affecting long-term outcome. In a controlled trial using individual cognitive-behavioral therapy (CBT) for childhood anxiety, Barrett et al. (1996) reported results from a subsample of 19 social phobia. The age range of their sample was 7 to 14 years. At post-

J . AM. A C A D . C H I L D ADOLESC. PSYCHIATRY. 3 9 : 6 . J U N E ZOO0

treatment 61.5% of the social-phobic subjects were free of this diagnosis, and 76.9% were diagnosis-free at a 1-year follow-up. The sample of Barren et al. (1996) is younger than that in the study reported here. Notably, in the study by Barrett et al. (1996), a parallel family anxiety management condition added benefit to the CBT, particularly for younger and female subjects. The contribution of the family component was not analyzed separately for the social-phobic subjects. However, a number of authors have argued for the importance of adding a family component to CBT for child and adolescent anxiety disorders (Albano et al., 1999; Dadds et al., 1992; Kendall, 1994). The results of this study, with respect to major depression, suggest that more investigation concerning the relationship between treating social phobia and altering outcomes and risk for major depression is warranted, as suggested by Kessler et al. (1999). Furthermore, these data suggest that the natural history of social phobia for a significant number of adolescents is complicated by episodes of major depression. These 2 disorders may have synergistic effects, with social anxiety promoting major depression and vice versa. Risk factor and treatment studies that combine social phobia and major depression as one outcome seem warranted, at least in adolescent samples. Limitations

This pilot study has important limitations. The CBGTA protocol is still in development and further revisions have been made since this study was initiated. The sample size is small and restricted to females; thus power and generalizability are both limited. The sample was recruited by advertisement, and therefore the results may not be generalizable to clinical samples. Although the assessmentswere performed with the raters blind to the subjects' treatment status, the subjects themselves were aware of their own treatment status. The subjects' awareness could have differentially affected expectation by treatment condition. The apparent improvement with respect to the diagnosis of social phobia at the 1-year follow-up in subjects in the untreated condition might result from regression to the mean or from the fact that a few of these subjects were receiving treatment within the community. To investigate this possibility, outcomes of these subjects were studied. Four subjects in the untreated condition reported receiving treatment between the posttreatment assessment and the 1-year follow-up; 2 of these 4 met criteria for social phobia at the I-year follow-up. Thus, receiving treatment in the 725

HAYWARD E T AL.

community did not appear to explain improvement in the untreated socially phobic group. Finally, we suggest that hture CBGT-A studies include parents in the treatment program. Enlisting parents as participants in the treatment may increase the effectiveness of the program. Clinical Implications

The pilot study reported here provides evidence for a short-term effect in using CBGT-A to treat adolescents suffering from social phobia. Studies designed to evaluate longer-term treatment maintenance that include booster sessions to enhance short-term gains are warranted. Overall, the investigation of CBGT for adolescent social phobia is in an early stage of development. As judged from the experience from studies of adults, it is likely that both CBGT and pharmacotherapy will be helpful for social phobia (Heimberg et al., 1998). This study also indicates that treatment of social phobia in female adolescents may reduce both social anxiety and the frequency of subsequent major depression, particularly in female adolescents with a previous history of major depression. This is important because of the comorbidity between social phobia and major depression, as well as the significant impairment associated with both of these disorders. REFERENCES Albano AM, Barlow D H (1996), Breaking thc vicious cydc: cognitive bchavi o d group trcatmcnt for socially anxious youth. In: Prychorocial Tmarmmtsf i r Child and Adohcrnt D u o h : EmpiricaUy Rased Snatrg'erf i r Clinical h r t i z e , Hibbs ED, Jcnscn PS, cds. Washington, Dc: American Psychological Assocition, pp 43-62 Albano AM, Dcovcilcr MF, Logsdon-Conradscn S (1999), Cognitivcbchavionl intcrvcntions with socially phobic childrcn. In: Handbook of f3ychotbmpier With Childm and Families, Russ SW, Ollcndick TH, cds. Ncw York: Plcnum, pp 255-280 Albano AM, DiBartotlo PM, Hcimbcrg RG, Barlow D H (1995a).Childrcn and adolcsccnts: assessment and trcauncnt. In: sorial Phobia Diagnorir, Amrrmmt, and Tnatmenr, Hcimbcrg RG, Licbowitz MR, Hopc DA. Schncicr FR, cds.Ncw York: Guilford, pp 387-425 Albano AM, Marten PA, Holt CS. Hcimberg RG, Barlow D H (1995b), Cognitivc-bchavionl group trcatmcnt for social phobia in adolcsccnrr: a prcliminary study.J N m Mort Dir 183:G9-656

726

Barrcn I'M, Dadds MR. Rapcc RM (1996). Family trcatmcnt of childhood anxicty: a wntrollcd trial. J Consult Clin RycholG:333-342 Bcidcl DC. Turnct SM (1998),Shy Children,Phobic Adulrr: Namm and Tmarmt OfSorialPhobirr Washmgton. DC:Amcrican Psychological hocktion sdmcd mcnBurkcKC. Burke JD Jr. Rtpcr DA, Rac DS (1990),Agcat -of ddisordcrsinfivcwmmunitypopuldons.AtrhGnnvrh;rrrru47511-518 Clark DB, Turncr SM. Bcidcl DC, Donovan JE, Kirisci L, Jacob RG (1994), Rcliability and validity of thc Social Phobia and Anxicty Invcnrory for Adolaccnts. PIycholArrCu 6:135-140 Dad& MR, Hurd PM,R a p e RM (1992).Thc rolc of hmily intcrvcntion in thc trcatmcnt of child anxicty disorders: some prcliminary findings. &hau Change 9:171-177 Hayward C, Killcn JD. Kracmcr HC, Taylor CB (1998). Linking sclfrcportcd childhood bchavionl inhibition to adolescent social phobia. J Am AcadChildAdolrrr PIychLzny 37:1308-1316 Hcimbcrg RG, Dodgc CS,Hopc DA, Kennedy CR, Zollo LJ (1990). Cognitivc bchavionl p u p uatmcnt for social phobia: comparison with a crcdiblc placcbo control. Cop Thm Rrr 14:l-23 Hcimbcrg RG, Licbowia MR. Hopc DA ct al. (1998).Cognitive bchavionl group thcnpy vs phcnclzine thcnpy for social phobia. Arch &IPryhiany 5 5: 1 133-1 14 1 Hcimbcrg RG. Salaman DG, Holt CS. Blendell KA (1993), Cognitivcbchavioral group trcatmcnt for social phobia: cffcctivcncss at fivc-ycar follow-up. C~gnThm RCS17:325-339 Kcndall PC (1994),Treating anxicty dicordcrs in childrcn: rcsults of a randomized clinical trial. ]Consult CIin P3yhol62:lWllO Kcsslcr RC, Sung P. Wittchcn H-U, Stcin M, Walters EE (1999). Lifetime wmorbiditicsbctwccn rocial phobia and mood dLordcn in thc US National ComorbiditySurvcy. RyrhDMed 29:555-567 Lcwinsohn I'M, Hops H. Robcm RE, Scdcy J R Andmvs JA (1993),Add& c m t psychopathology.I: p d e n c c and incidcncc of dcprcssion and other DSM-III-R~L~~ in~h&C Kschool nudcnts.JAbnorm b h l 1 0 2 133-144 Ollcndick TH, King NJ (1998). Empirically supported trcatmcnts for children with phobic and anxicty disordcn: currcnt sum.] Clin Childpivhol 27156-167 Rcger DA, Rac DS, Narrow WE, Kaclbcr CT. Schanbcrg AF (1998).Prcvalencc of anxicty dicorders and thcir comorbidity with mood and addictivc disordcrs. BrJRychirrny 173:24-28 Schaabcrg AF, Samson JA, Rothschild AJ, Bond TC, Rcgier DA (1998), McIzan Hospital dcprcssion racarch facility: carly-onsct phobic disordcrs and adult-onsct major depression. BrJI3ychiany Suppl29-34 Scholing A, Emmclkamp PMG (1990).Social phobia: naturc and treatment. In: Handbook of Sorial and Evaluation Anmty, Lcitenbcrg H. cd. Ncw York: Plcnum, pp 269-324 Silvcrman W K (1991), Diagnostic rcliability of anxicty disorders in chiidrcn using structurcd intcrvim. D i o d 5: 105-124 Silverman WK,Albano AM (1936a), TheAnxiety D i r o h I n m h Srhedule fir DSM-N: Chikilnm'nuSchedule. San Antonio, TX Gnywind Silverman WK.Albano AM (1996b). TheAnriq D i r o h I n m b Schedule fir DSM-IV: Pamt I n m b Schedule. San Antonio, TX Graywind Turner SM, Stanlcy MA,Bcidcl DC. Bond L (1989). Thc Social Phobia and Anxicty Invcntory: W N ~ N Cvalidity. ~ &hau Rrr Thn 11:221-234 Winchcn H-U, Stein MB, Kaslcr RC (1999).Social furs and social phobia in a community samplc of adolcsccnts and young adults: prcvalcncc, risk facron and wmorbidity. I3ychol Med 29:309-323

Jm

J . A M . A C A D . CHILD ADOLESC. PSYCHIATRY. 3 9 : 6 , JUNE ZOO0