Regular Article Psychother Psychosom 2014;83:54–61 DOI: 10.1159/000354672
Received: December 19, 2012 Accepted after revision: July 25, 2013 Published online: November 19, 2013
A Randomized Controlled Trial of Individual Cognitive Therapy, Group Cognitive Behaviour Therapy and Attentional Placebo for Adolescent Social Phobia Jo Magne Ingul a, c Tore Aune c, d Hans M. Nordahl b, c
a
Department of Child and Adolescent Psychiatry, and b Department of Psychiatry, Levanger Hospital, Nord-Trøndelag Health Trust, Levanger, and c Department of Psychology, Norwegian University of Science and Technology, Trondheim, d University College of Nord Trøndelag, Levanger, Norway
Key Words Social phobia · Adolescents · Individual cognitive therapy · Group cognitive behavioural therapy
the most commonly used treatments for adolescent SP, we found that individual therapy was the most effective, yielding better effects than both CBTG and AP. © 2013 S. Karger AG, Basel
© 2013 S. Karger AG, Basel 0033–3190/14/0831–0054$39.50/0 E-Mail [email protected] www.karger.com/pps
Introduction
Social phobia (SP) or social anxiety disorder tends to follow a chronic and unremitting course [1]. It is often associated with educational and social impairment among adolescents, carrying with it an increased risk of suicide attempts and substance abuse in adults [2]. In recent decades, several studies have been published documenting the effect of cognitive behaviour therapy (CBT) treatment on children and adolescents with anxiety disorders [3, 4]. Reported treatment efficiency varies across studies, and Rao et al. [5] suggested that there is a need for specialized, disorder-specific treatment, but also saw a need for different treatment approaches across different age groups. Only 6 studies have specifically investigated the treatment of SP among adolescents [6–12]. All studies included some version of a group CBT (CBTG) format. The exception is the study of Herbert et al. [10], who compared individual and group therapeutic interventions based on Jo Magne Ingul Department of Child and Adolescent Psychiatry Levanger Hospital Kirkegt 2, NO–7600 Levanger (Norway) E-Mail jo.magne.ingul @ hnt.no
Downloaded by: Siriraj Medical Library, Mahidol University 202.28.191.34 - 2/16/2015 12:35:10 AM
Abstract Background: Very few studies have investigated the effects of individual disorder-specific treatment of social phobia (SP) in adolescents. The objective of this study was to compare the effects of individual cognitive therapy for SP, group cognitive behavioural therapy (CBTG) and attentional placebo (AP) among adolescents with a primary diagnosis of SP. Methods: A randomized controlled design was used, and a total of 279 adolescents were assessed. Fifty-seven adolescents, between 13 and 16 years old, were allocated to individual cognitive therapy, CBTG or AP. The participants were assessed before treatment, at the end of treatment and at a 12-month follow-up using both self-report and a semi-structured interview. Results: The individual cognitive therapy showed significant reductions in symptoms, impairment and diagnostic criteria both at the end of treatment and at the 12-month follow-up. Compared with CBTG and AP, the individual cognitive therapy group demonstrated significantly greater effects on both symptom reduction and impairment. There were no significant differences between CBTG and AP. Conclusions: In a direct comparison between
RCT of Adolescent Social Phobia
Methods Participants All participants were 8th- to 10th-grade students recruited from a school-based screening (n = 1,748). Out of the 279 assessed, 128 were found eligible for treatment, and 57 completed the treatment (fig. 1). The mean age of the participants who completed the study was 14.5 years, and 32 (56.1%) were girls. All participants had an SP diagnosis as their primary problem; 75.4% had generalized SP and 24.6% had a specific SP (specific SP was defined as a more circumscribed SP, where the adolescent showed severe impairment and symptoms of anxiety in 1–2 specific situations). Approximately 65% had at least 1 comorbid disorder, the most common being specific phobia (49%) and generalized anxiety disorder (28%; table 1). Procedure The study was approved by the Regional Ethical Committee and introduced to the school authorities in all counties by written information and meetings. Recruitment was performed in two waves using a two-stage process. Participants initially consented to participate in the first stage of recruitment and were required to give a second consent before being entered into the second stage. Students who, in the screening, scored above the clinically recommended cut-off on the Social Phobia and Anxiety Inventory for Children (SPAI-C) [23] (n = 277) and 56 randomly selected adolescents, to allow blind assessment, were invited to an interview (n = 333; fig. 1). The interviews were conducted using the Anxiety Disorders Interview Schedule for DSM-IV – Child version (ADIS-C) [24]. Students who fulfilled the inclusion criteria were informed about the treatment study and invited to participate. Adolescents who consented were given written information about the treatment study and a consent form to take home. They were then entered into the second stage of recruitment and randomized. After randomization, adolescents and their parents were contacted, asked for confirmation of their consent and provided additional information. Assessment was conducted before and after treatment and at a 12-month follow-up. Symptom measures were assessed at all 3 points. A clinical interview (ADIS-C) was conducted both before treatment and at the 12-month follow-up. At the follow-up interview, assessors were blinded with respect to treatment conditions, receiving only the name and contact information of each adolescent. Inclusion and Exclusion Criteria To qualify for inclusion, students had to be in grades 8–10 and experiencing SP as their primary problem. The presence of mental retardation or psychoses was considered a ground for exclusion. Participants who were already being treated elsewhere for mental health conditions were also excluded (fig. 1). Estimation of Sample Size The SPAI-C, the main outcome measure, has a maximum score of 52 points and a standard deviation (SD) of 7 points [25]. An effect size of 0.80 was the expected minimum clinical difference. With a 5% significance level and 80% power, 16 adolescents would be required in each group to detect significant differences. With an expected attrition rate of 25%, we decided that at least 20 adolescents would be required in each group.
Psychother Psychosom 2014;83:54–61 DOI: 10.1159/000354672
55
Downloaded by: Siriraj Medical Library, Mahidol University 202.28.191.34 - 2/16/2015 12:35:10 AM
the same protocol. Overall, the studies show significant reductions in both symptoms and recovery rates from pre-treatment to post-treatment [8–12]. Reported followup results indicate that changes were maintained for periods ranging from 6 months [10, 11] to 5 years [13]. The exception is Hayward et al. [9], who reported no significant effect of treatment at a 1-year follow-up. The results from these reported studies show promising results, but there are some potential limitations. The studies are relatively small, with limited statistical power [6, 7, 9]; moreover, one was an uncontrolled pilot study [6] and others compared CBT to a waiting list control [7, 9]. Most studies reported large dropout rates, and in one study, treatment rejecters were included in the control condition [8]. Among adults, both individual and group therapy formats have been reported to be effective [14]. Rodebaugh and Rapee [14] found no difference in outcome between therapy formats. Other studies have reported individual therapy to be superior [15, 16]. Herbert et al. [10] examined differences between therapy formats in adolescents with SP. In this study, the same protocol [17] was used for both individual and group therapy. Both conditions were associated with significant reductions in symptoms and functional impairment, with only small differences between the formats. To our knowledge, no study has reported on adolescent SP using a manual specifically designed for individual therapy. Clark and Wells [18] developed a cognitive therapy for SP in adults. This model focuses on the cognitive processes that maintain SP. Results of the treatment are promising in adults [16, 19]. The status of the Clark and Wells model in children and adolescents is not established; however, studies providing evidence for elements in the model have been reported [20, 21], but also one study disfavouring the validity of the model [22]. In summary, treatment of SP in adolescents has been dominated by studies using a group therapy format. No study has ever investigated individual treatment of SP in adolescents with a manual specially designed for the format. Currently, the effectiveness of the Clark and Wells [18] treatment in adolescents is unknown. In this study, we wanted to test the effectiveness of the Clark and Wells treatment in adolescents, comparing it to the best treatment available, CBTG. To control for the effect of social exposure, a group receiving attentional placebo (AP) was included. With this background, we predicted that (1) individual therapy based on Clark and Wells’ treatment would prove as effective as CBTG in treating adolescents with SP and (2) that both individual cognitive therapy and CBTG would be more effective than AP.
Total screened n = 1,748
Refused to participate n = 52 Other reasons n = 2 (moved)
Potential participants n = 333
Assessed for eligibility n = 279
Did not meet inclusion criteria n = 125 Met exclusion criteria n = 11
Randomized n = 128
Did not want treatment n = 15
Allocated to AP n = 34 No show n = 16
Allocated to CBTG n = 58 No show n = 25
Allocated to CBTI n = 36 No show n = 11
Post-treatment assessment completed n = 16 No data n=2
Post-treatment assessment completed n = 20 No data n = 13
Post-treatment assessment completed n = 21 No data n=4
Follow-up completed (12 months) n = 15
Follow-up completed (12 months) n = 12
show: adolescents who initially agreed to participate but later declined, before treatment started.
Randomization The randomization was performed at each site, because of long travelling distances. Randomization was conducted using a preassigned random schedule generated from the SPSS 15.0 random number generator. Measures Structured Interview The ADIS-C is a semi-structured interview for anxiety disorders and the most common comorbid disorders in children and adolescents using DSM-IV criteria [24]. The ADIS-C has generally shown good to excellent reliability. The ADIS-C has been found to be sensitive to treatment effects [26, 27]. Diagnoses were assessed by clinical psychology graduate students and psychiatric nurses trained by the authors. Following the training period, and before they qualified as assessors, 25% of all training interviews were randomly selected and reassessed. Kappa for the social anxiety diagnosis of the ADIS-C was 0.75.
56
Psychother Psychosom 2014;83:54–61 DOI: 10.1159/000354672
Primary Outcome Measures SPAI-C. The SPAI-C [23] is a 26-item self-report inventory designed to assess symptoms of DSM-IV SP. It has shown good internal consistency, test-retest reliability [23, 28] and construct validity [28]. Cronbach’s α for the SPAI-C in this study was 0.95. The Social Thoughts and Beliefs Scale. The Social Thoughts and Beliefs Scale (STABS) [29] is a 21-item self-report inventory designed to assess cognition in persons with SP. It has shown excellent internal consistency, test-retest reliability and discriminant validity [29, 30]. Acceptable concurrent validity has also been documented [30]. Chronbach’s α for STABS in this study was 0.96. Secondary Outcome Measures Anxiety. The Screen for Child Anxiety Related Emotional Disorders (SCARED) [31] is a 41-item self-report measure appropriate for use with children and adolescents 8–18 years old. It has demonstrated very good psychometric properties in a clinical sample [31] and in a Norwegian community sample [32]. Cronbach’s
Ingul/Aune/Nordahl
Downloaded by: Siriraj Medical Library, Mahidol University 202.28.191.34 - 2/16/2015 12:35:10 AM
Fig. 1. Flow chart of study participants. No
Table 1. Demographic and baseline data by randomized condition
CBTI (n = 21) Female Age, years Living with sibling Yes No Missing data
12 (57.14) 14.98±0.94
CBTG (n = 20)
AP (n = 16)
12 (60.00) 14.30±0.89
8 (50.00) 14.16±1.08
Total (n = 57) 32 (56.14) 14.50±1.01
15 (71.43) 6 (28.57) 0
13 (65.00) 5 (25.00) 2 (10.00)
11 (68.75) 4 (25.00) 1 (6.25)
39 (68.42) 15 (26.32) 3 (5.26)
Living with Mother and father Mother only Father only Mother and father, but not together Missing data
9 (42.86) 7 (14.33) 2 (9.52) 3 (14.29) 0
13 (65.00) 4 (20.00) 1 (5.00) 0 2 (10.00)
13 (81.25) 1 (6.25) 0 1 (6.25) 1 (6.25)
35 (61.40) 12 (21.05) 3 (5.26) 4 (7.02) 3 (5.26)
School grade 8 9 10
6 (28.57) 7 (33.33) 8 (38.10)
6 (30.00) 9 (45.00) 5 (25.00)
8 (50.00) 4 (25.00) 3 (18.75)
20 (35.09) 20 (35.09) 16 (28.07)
16 (76.19)
15 (75.00)
12 (75.00)
43 (75.44)
Generalized SP Interference rating before treatment Comorbidity Specific phobia GAD Agoraphobia Separation anxiety OCD ADD Depression PTSD
4.48±1.37 15 (71.43) 9 (42.86) 7 (33.33) 0 1 (4.76) 0 3 (14.29) 2 (9.52) 1 (4.76)
4.65±1.39 12 (60.00) 9 (45.00) 6 (30.00) 4 (20.00) 0 0 1 (5.00) 2 (10.00) 0
4.25±1.00 10 (62.50) 10 (62.50) 3 (18.75) 0 2 (12.50) 1 (6.25) 1 (6.25) 1 (6.25) 1 (6.25)
4.47±1.27 37 (64.91) 28 (49.12) 16 (28.07) 4 (7.02) 3 (5.26) 1 (1.75) 5 (8.77) 5 (8.77) 2 (3.50)
α in this study was 0.93 for the total SCARED and 0.80 for the social anxiety disorder subscale (SCARED-SAD-sub). The SCARED assesses the DSM-IV symptoms of 4 different diagnoses, and it also provides total and school avoidance scores. Depression. The Children’s Depression Inventory (CDI) is a multidimensional measure quantifying depressive symptomatology in children and adolescents [33]. It has 27 items in which the child must choose 1 of 3 levels and yields scores for 5 factors and a total score. The CDI has shown good psychometric properties [33]. Cronbach’s α for the CDI in this study was 0.98. Quality of Life. To measure the perceived quality of life, the Pediatric Quality of Life Inventory (PedsQL) 4.0 [34] was used. The PedsQL consists of 23 items covering 4 domains – physical, emotional, social and school functioning – and calculates a total quality of life score. The items are scored on a 5-point Likert scale and linearly transformed to a 0–100 scale. The scale has shown adequate psychometric properties [34].
Interventions Treatment and Therapists The AP groups were led by experienced milieu therapists, most of whom were psychiatric nurses and social workers with extensive clinical experience. CBTG was led and supervised by an experienced clinical psychologist (second author) and a team of 4 cotherapists. In the individual CBT condition (CBTI), treatment was led and supervised by a clinical psychologist (first author) and 2 co-therapists. In both CBTG and CBTI conditions, independent experts assessed the therapies using video and audiotapes. They focused on treatment integrity, adherence to protocol and competence of the therapists. No protocol violations were reported, and the competence of the therapists was reported to be high overall.
RCT of Adolescent Social Phobia
Psychother Psychosom 2014;83:54–61 DOI: 10.1159/000354672
Attentional Placebo AP was conducted in groups over 10 sessions of 90 min each. The groups consisted of 4–6 participants. This condition did not include any of the active components in the CBTI or CBTG.
57
Downloaded by: Siriraj Medical Library, Mahidol University 202.28.191.34 - 2/16/2015 12:35:10 AM
Values represent numbers of participants (percentages) or means ± SD, as appropriate. GAD = Generalized anxiety disorder; OCD = obsessive compulsive disorder; ADD = attention deficit disorder; PTSD = post-traumatic stress disorder.
Through social activity, social interaction and social support, the adolescents were exposed to adults, peers and social attention to the same degree as in the other conditions.
Table 2. Outcome measures for the 3 intervention groups before
Group CBT The CBTG was conducted over 10 sessions of 90 min each. The groups consisted of 4–6 participants. The manual was based on The C.A.T. Project Manual for the Cognitive Behavioral Treatment of Anxious Adolescents [35] with some ‘play elements’ from the Social Effectiveness Therapy for Children and Adolescents program [36]. The manual was divided into two parts (online suppl. table 1; for all online suppl. material, see www.karger.com/ doi/10.1159/000354672). Prior to the study, the protocol was tested by the second author in a pilot study, showing significant symptom reductions [37].
CBTI (n = 21) CBTG (n = 20) AP (n = 16)
Individual CBT The CBTI was conducted over 12 sessions of 50 min each. The manual was developed by Clark and Wells [18] as a treatment for adult SP. The language, tempo and type of interventions were adapted for use with adolescents by the first and third authors. The treatment included 3 phases (online suppl. table 1). Statistical Analysis The demographic data were analysed using χ2 tests and analysis of variance. The effect of different treatment conditions on adolescent SP was tested with paired sample t tests. Analysis of clinically significant change was performed using the reliable change index (RCI) [38]. The RCI method evaluates the difference between pretest and post-test scores in a single case, using published norms for clinical and nonclinical populations for the relevant disorder. Values greater than 1.96 are considered to represent statistically reliable change [38]. The values used to calculate RCI here were derived from Aune et al. [25]. Between-group comparisons were performed using factorial analysis of co-variance (ANCOVA) and independent-sample t tests. Factorial ANCOVA assesses group differences on a single dependent variable after the effects of one or more known covariates are statistically removed, thereby minimizing error variance. When there are more than two groups, planned comparisons can be applied, assessing differences in means on the dependent variable using pooled group means adjusted for differences in covariates [39].
SPAI-C Pre Post FU FU ITT STABS Pre Post FU FU ITT SCARED-SAD-sub Pre Post FU FU ITT Interference Pre FU SCARED total Pre Post FU CDI Pre Post FU PedsQL Pre Post FU
26.81±6.77 7.86±5.76 10.64±7.90 10.29±6.68
22.20±9.16 23.16±9.35 18.47±9.99 18.35±9.06
22.40±9.24 17.63±9.49
48.10±13.70 21.12±13.82 22.92±17.87 23.81±15.72
40.41±12.04 36.78±15.94 39.65±15.73 37.70±14.39
40.85±9.37 31.98±10.83
7.90±2.26 3.81±3.63 4.83±3.76 4.81±3.66
6.50±2.80 6.05±3.79 7.00±3.40 6.47±3.39
5.81±2.64 5.67±2.64
4.48±1.37 0.91±1.45
4.65±1.39 2.67±2.47
4.25±1.00
27.72±12.72 14.45±8.45 14.33±12.81
22.60±12.52 25.16±17.10 22.47±14.32
23.13±15.30 22.87±12.84
21.34±3.61 10.24±9.70 12.17±9.48
18.42±11.95 16.35±11.48 14.13±8.85
12.60±7.22 7.04±4.62
34.91±14.39 18.88±12.57 19.35±14.60
27.46±12.46 28.98±16.56 29.55±16.58
24.64±11.18 21.38±11.73
Values represent means ± SD. Pre = Before treatment; Post = after treatment; FU = 12 months after treatment; ITT = intention to treat.
Pre-Treatment Comparisons There were no significant differences between the 3 intervention groups on demographic variables except age, where there was a significant difference [F(2, 52) = 3.72, p < 0.05] between the CBTI group (mean = 14.98, SD = 0.94) and the AP group (mean = 14.16, SD = 1.08; table 1). Comparison of the groups with regard to pre-treatment symptom severity showed no significant differences on the main outcome measures of social anxiety, type of Psychother Psychosom 2014;83:54–61 DOI: 10.1159/000354672
Treatment Attrition A χ2 test showed that the assigned treatment condition did not significantly influence whether the adolescents started treatment or not (χ2 = 1.97, p > 0.05). There was a significant association between the type of treatment received and whether or not they completed treatment (χ2 = 7.79, p < 0.01), with significantly more adolescents completing the CBTI (n = 21, 84.0%) and AP (n = 16, 88.9%) than the CBTG (n = 20, 58.8%). There were no significant Ingul/Aune/Nordahl
Downloaded by: Siriraj Medical Library, Mahidol University 202.28.191.34 - 2/16/2015 12:35:10 AM
social anxiety disorder diagnosis (generalized or specific) or the interference of anxiety symptoms in participants’ daily life (tables 1, 2).
Results
58
treatment, after treatment, at follow-up and intention-to-treat follow-up
Changes in Symptom Severity Within-Condition Changes For the AP group, results were only analysed at the end of treatment. Paired-sample t tests showed that the AP group had a significant reduction in symptoms as measured by the SPAI-C [t(14) = 2.37, p < 0.05] and the STABS [t(13) = 3.61, p < 0.01; online suppl. table 2]. For the CBTG group, there were no significant changes from pre- to post-treatment. At follow-up, there were significant changes in symptoms on the SPAI-C [t(14) = 2.45, p < 0.05] compared to pre-treatment. For CBTI, significant reductions were seen in all 3 social anxiety measures from pre- to post-treatment [SPAI-C: t(20) = 8.47, p < 0.001; STABS: t(20) = 8.12, p < 0.001; SCARED-SAD-sub: t(20) = 4.74, p < 0.001] and from pre-treatment to followup [SPAI-C: t(11) = 4.04, p < 0.01; STABS: t(11) = 3.64, p < 0.01; SCARED-SAD-sub: t(11) = 3.21, p < 0.01; online suppl. table 2]. RCI analysis for participants in the CBTI group showed that 69.8% at the end of treatment and 51.4% at follow-up had an RCI indicating clinically significant improvement. In the CBTG group, 21.1% at the end of treatment and 13% at follow-up had an RCI indicating clinically significant improvement. In the AP group, 27.8% had an RCI indicating clinically significant improvement at the end of treatment. Between-Condition Comparisons Differences between treatment conditions were tested using ANCOVA with planned contrasts. Age, gender and pre-treatment social anxiety and depression levels were used as covariates. There were significant effects of treatment conditions on all 3 measures [SPAI-C: F(2, 44) = 18.60, p < 0.001, d = 2.08; STABS: F(2, 45) = 6.48, p < 0.01, d = 1.07; SCARED-SAD-sub: F(2, 48) = 3.66, p < 0.05, d = 0.81] when the covariates were controlled for. Covariates had no significant effect. Planned contrasts showed the significant differences to be between CBTI and CBTG [SPAI-C: t(35) = 6.34, p < 0.001; STABS: t(35) = 4.18, p < 0.001; SCARED-SAD-sub: t(35) = 2.52, p < 0.05; online suppl. table 3] and between CBTI and AP [SPAI-C: t(32) = 4.25, p < 0.001; STABS: t(32) = 2.07, p < 0.01; SCARED-SAD-sub: t(31) = 2.19, p < 0.05] on all 3 measures. At follow-up, independent-sample t tests showed that there were significant differences between CBTI and RCT of Adolescent Social Phobia
CBTG on 2 of the 3 social anxiety measures [SPAI-C: t(25) = 2.22, p < 0.05; STABS: t(25) = 2.59, p < 0.05; online suppl. table 3]. Intention-to-Treat Analysis Because of missing data at follow-up (34%), intentionto-treat analyses using the last observation carried forward method was carried out. Both within- and betweencondition (CBTI vs. CBTG) analyses using intention-totreat data confirmed the results of the completer analysis (online suppl. tables 2, 3). Recovery and Impairment Ratings Prior to treatment, all participants were diagnosed with SP. At follow-up, 72.7% of participants in the CBTI group and 53.3% of those in the CBTG group had no SP diagnosis. The difference between the 2 groups was not significant. Ratings of impairment, the way in which symptoms interfere with daily lives, showed that there was a significant difference between the CBTI and CBTG groups [t(23.09) = 2.28, p < 0.05] at follow-up. Both conditions showed significant reductions in interference from pre-treatment to follow-up [CBTI: t(10) = 4.98, p < 0.001; CBTG: t(14) = 3.87, p < 0.01; online suppl. table 2]. Secondary Measures The SCARED total score, the CDI and the PedsQL were also tested within each condition. The analysis showed that there were no significant changes at any time for the CBTG group. For the AP group, there were significant changes from pre-treatment to post-treatment on the CDI [t(14) = 2.93, p < 0.05] and PedsQL [t(13) = 2.18, p < 0.05; online suppl. table 2]. All 3 secondary measures were significantly different between pre-treatment and post-treatment in the CBTI condition [SCARED: t(16) = 3.59, p < 0.01; CDI: t(20) = 5.01, p < 0.001; PedsQL: t(20) = 4.32, p < 0.001]. There was also a significant change from pre-treatment to follow-up on the CDI [t(11) = 3.46, p < 0.01].
Discussion
Overall, the results indicate that individual cognitive therapy (CBTI) yields better effects than both CBTG and AP. In-Albon and Schneider [4] reported the average effect size of CBT on symptoms of SP to be 0.80, while Segool and Carlson [40] reported effect sizes of 0.86 on symptoms and 1.56 on impairment. Our results for CBTI were 2.07 and 2.49. Moreover, a recovery rate of 73% is Psychother Psychosom 2014;83:54–61 DOI: 10.1159/000354672
59
Downloaded by: Siriraj Medical Library, Mahidol University 202.28.191.34 - 2/16/2015 12:35:10 AM
differences on any of the clinical measures [SPAI-C: t(124) = 0.48, p > 0.05; STABS: t(117) = 0.57, p > 0.05] between those who did not start treatment and those who finished the treatment.
60
Psychother Psychosom 2014;83:54–61 DOI: 10.1159/000354672
with respect to the conclusions, as noted by Tomba and Bech [44]. However, the main measures, the SPAI-C [45] and the ADIS-C [26, 27], have been shown to be sensitive to change. Assessment was done with adolescents as the only informants. A multi-informant approach can increase the reliability of the diagnoses. However, there is evidence that adolescents should be the primary source when diagnosing SP [46]. Also, Silverman et al. [3] found that efficacy rates of CBT are much larger based on parent (0.91) than child (0.44) ratings. This indicates that children experience less change than their parents. The implication for this study is that we are more likely to underestimate than overestimate change. In conclusion, with the limitations of the study in mind, working with negative perceptual biases maintaining SP seems to provide relatively strong and enduring results. Our results also indicate that individual therapy with a protocol specifically designed to face idiosyncratic problems may be applicable in the treatment of adolescent SP. Both individual and group cognitive therapy are effective, but the study also demonstrates that it is important to control for the effect of natural self-exposure in comparative studies, as seen in the results of the AP condition in the current study.
Acknowledgements This study was supported by a grant from the Liaison Committee between the Central Norway Regional Health Authority and the Norwegian University of Science and Technology.
Disclosure Statement No author or immediate family member has financial arrangements that might represent potential conflicts of interest for the findings presented, nor has any funding source influenced the content of this report.
References
1 Kessler RC, Berglund P, Demler O, Jin R, Merikangas KR, Walters EE: Lifetime prevalence and age-of-onset distributions of DSMIV disorders in the National Comorbidity Survey replication. Arch Gen Psychiatry 2005;62:593–602. 2 Stein MB, Stein DJ: Social anxiety disorder. Lancet 2008;371:1115–1125. 3 Silverman WK, Pina AA, Viswesvaran C: Evidence-based psychosocial treatments for phobic and anxiety disorders in children and adolescents. J Clin Child Adolesc Psychol 2008;37:105–130.
Ingul/Aune/Nordahl
Downloaded by: Siriraj Medical Library, Mahidol University 202.28.191.34 - 2/16/2015 12:35:10 AM
relatively large and in line with the best reported results [3, 4]. To our knowledge, this is the first randomized controlled trial of individual cognitive therapy based on the Clark and Wells [18] treatment for SP in adolescents. The results may be important as they indicate that working on the negative perceptual biases is useful also in adolescents. They may also be important as recent research [41] has shown that treatment with antidepressants has been associated with a substantial risk of adverse effects, including excessive mood elevation and behavioural activation, making it even more important to develop and document good psychotherapeutic methods and ensuring that mental health professionals are trained to deliver these [42]. For the CBTG group, there were significant reductions in both symptoms and interference from pre-treatment to follow-up, which was in accordance with our hypothesis. A long-term treatment efficacy of 53% is in line with what has generally been reported among adolescents [8– 12]. However, there were no significant changes from pre-treatment to post-treatment, and overall, the changes were lower than expected. The results of the study indicate that CBTG is not as effective as CBTI. This could be a format effect, given that individual therapy has been shown to be more effective than group therapy in some adult studies [15, 16]. However, childhood anxiety studies have shown that there are no differences between group and individual formats [4, 10, 43]. This indicates that the format of the therapy is not an entirely sufficient explanation of the differences in the present study. CBTG is based on standard CBT, utilizing standard techniques and understanding of SP, while CBTI makes assumptions with regard to specific maintaining factors, targeting these in treatment. Different models may therefore be an explanation of the differences in results. The study gives several very interesting findings but they have to be interpreted with caution, and there are some limitations that need consideration. As Herbert et al. [10] report, high attrition rates and small samples are the norm in studies of adolescent SP. Although limited, our sample size is larger than estimated to be necessary to have adequate statistical power. Even though we tried to minimize the risk and effects of dropout, a third of those who completed the treatment failed to participate at follow-up. We cannot rule out the possibility that the results may be inflated, as we might expect the successful participants to show up. The present study relied on self-report and clinical interviews for measuring change; this warrants caution
RCT of Adolescent Social Phobia
18 Clark DM, Wells A: A cognitive model of social phobia; in Heimberg RG, Liebowitz MR, Hope DA, Schneier FR (eds): Social Phobia: Diagnosis, Assessment, and Treatment. New York, Guilford Press, 1995, pp 69–93. 19 Clark DM, Ehlers A, Hackmann A, McManus F, Fennell M, Grey N, Waddington L, Wild J: Cognitive therapy versus exposure and applied relaxation in social phobia: a randomized controlled trial. J Consult Clin Psychol 2006;74:568–578. 20 Morgan J, Banerjee R: Social anxiety and selfevaluation of social performance in a nonclinical sample of children. J Clin Child Adolesc Psychol 2006;35:292–301. 21 Cowart MJW, Ollendick TH: Attention training in socially anxious children: a multiple baseline design analysis. J Anxiety Disord 2011;25:972–977. 22 Alfano CA, Beidel DC, Turner SM: Negative self-imagery among adolescents with social phobia: a test of an adult model of the disorder. J Clin Child Adolesc Psychol 2008; 37: 327–336. 23 Beidel DC, Turner SM, Morris TL: A new inventory to assess childhood social anxiety and phobia: The Social Phobia and Anxiety Inventory for Children. Psychol Assess 1995; 7:73–79. 24 Albano AM, Silverman WK: Anxiety Disorders Interview Schedule for DSM-IV: Child Version Clinician Manual. New York, Oxford University Press, 1996. 25 Aune T, Stiles TC, Svarva K: Psychometric properties of the Social Phobia and Anxiety Inventory for children using a non-American population-based sample. J Anxiety Disord 2008;22:1075–1086. 26 Silverman WK, Saavedra LM, Pina AA: Testretest reliability of anxiety symptoms and diagnoses with anxiety disorders interview schedule for DSM-IV: child and parent versions. J Am Acad Child Adolesc Psychiatry 2001;40:937–944. 27 Kendall PC, Flannery-Schroeder E, Panichelli-Mindel SM, Southam-Gerow M, Henin A, Warman M: Therapy for youths with anxiety disorders: a second randomized clincal trial. J Consult Clin Psychol 1997;65:366–380. 28 Beidel DC, Turner SM, Hamlin K, Morris TL: The Social Phobia and Anxiety Inventory for Children (SPAI-C): external and discriminative validity. Behav Ther 2000;31:75–87. 29 Turner SM, Johnson MR, Beidel DC, Heiser NA, Lydiard RB: The Social Thoughts and Beliefs Scale: a new inventory for assessing cognitions in social phobia. Psychol Assess 2003; 15:384–391. 30 Fergus TA, Valentiner DP, Kim HS, Stephenson K: The Social Thoughts and Beliefs Scale: psychometric properties and its relation with interpersonal functioning in a non-clinical sample. Cognit Ther Res 2009;33:425–431.
31 Birmaher B, Khetarpal S, Brent DA, Cully M, Balach L, Kaufman J, Neer SM: The Screen for Child Anxiety Related Emotional Disorders (SCARED): scale construction and psychometric characteristics. J Am Acad Child Adolesc Psychiatry 1997;36:545–553. 32 Leikanger E, Ingul JM, Larsson B: Sex and age-related anxiety in a community sample of Norwegian adolescents. Scand J Psychol 2012; 53:150–157. 33 Kovacs M: CDI: Children’s Depression Inventory: Technical Manual Update. Toronto, MHS, 2003. 34 Varni JW, Seid M, Kurtin PS: PedsQL 4.0: reliability and validity of the Pediatric Quality of Life Inventory version 4.0 generic core scales in healthy and patient populations. Med Care 2001;39:800–812. 35 Kendall PC, Choudhury M, Hudson JL, Webb A: The C.A.T. Project Manual for the Cognitive Behavioral Treatment of Anxious Adolescents. Philadelphia, Temple University, 2000. 36 Beidel DC, Turner SM, Morris TL: Social Effectiveness Therapy for Children and Adolescents. Toronto, MHS, 2004. 37 Aune T: Sosial angstlidelse hos barn og unge. Tidsskr Norsk Psykol 2011;48:12–16. 38 Jacobson NS, Truax P: Clinical significance: a statistical approach to defining meaningful change in psychotherapy research. J Consult Clin Psychol 1991;59:12–19. 39 Tabachnik BG, Fidell LS: Using Multivariate Statistics, ed 5. Boston, Pearson International, 2007. 40 Segool N, Carlson JS: Efficacy of cognitivebehavioral and pharmacological treatments for phobic and anxiety disorders in children and adolescents. Depress Anxiety 2008; 25: 620–631. 41 Offidani E, Fava GA, Tomba E, Baldessarini RJ: Excessive mood elevation and behavioural activation with antidepressant treatment of juvenile depressive and anxiety disorders: a systematic review. Psychother Psychosom 2013;82:132–141. 42 Leckman JF: The risk and benefits of antidepressants to treat pediatric-onset depression and anxiety disorders: a developmental perspective. Psychother Psychosom 2013; 82: 129–131. 43 Flannery-Schroeder EC, Kendall PC: Group and individual cognitive-behavioral treatments for youth with anxiety disorders: a randomized clinical trial. Cognit Ther Res 2000; 24:251–278. 44 Tomba E, Bech P: Clinimetrics and clinical psychometrics: macro- and microanalysis. Psycother Psychosom 2012;81:333–343. 45 Beidel DC, Turner SM, Morris TL: SPAI-C. Social Phobia and Anxiety Inventory for Children. Manual. New York, MHS, 2000. 46 La Greca AM, Lopez N: Social anxiety among adolescents: linkages with peer relations and friendships. J Abnorm Child Psychol 1998;26: 83–94.
Psychother Psychosom 2014;83:54–61 DOI: 10.1159/000354672
61
Downloaded by: Siriraj Medical Library, Mahidol University 202.28.191.34 - 2/16/2015 12:35:10 AM
4 In-Albon T, Schneider S: Psychotherapy of childhood anxiety disorders: a meta-analysis. Psychother Psychosom 2007;76:15–24. 5 Rao PA, Beidel DC, Turner SM, Ammerman RT, Crosby LE, Sallee FR: Social anxiety disorder in childhood and adolescence: descriptive psychopathology. Behav Res Ther 2007; 45:1181–1191. 6 Albano AM, Marten PA, Holt CS, Heimberg RG: Cognitive-behavioral group treatment for social phobia in adolescents: a preliminary study. J Nerv Ment Dis 1995;183:649–656. 7 Baer S, Garland EJ: Pilot study of communitybased cognitive behavioral group therapy for adolescents with social phobia. J Am Acad Child Adolesc Psychiatry 2005;44:258–264. 8 Garcia-Lopez LJ, Olivares J, Turner SM, Beidel DC, Albano AM, Sanchez-Meca J: Results at long-term among three psychological treatments for adolescents with generalized social phobia. II. Clinical significance and effect size. Psicol Conductual 2002;10:371–385. 9 Hayward C, Varady S, Albano AM, Thienemann M, Henderson L, Schatzberg AF: Cognitive-behavioral group therapy for social phobia in female adolescents: results of a pilot study. J Am Acad Child Adolesc Psychiatry 2000;39:721–726. 10 Herbert JD, Gaudiano BA, Rheingold AA, Moitra E, Myers VH, Dalrymple KL, Brandsma LL: Cognitive behavior therapy for generalized social anxiety disorder in adolescents: a randomized controlled trial. J Anxiety Disord 2009;23:167–177. 11 Masia Warner C, Klein RG, Dent HC, Fisher PH, Alvir J, Albano AM, Guardino M: Schoolbased intervention for adolescents with social anxiety disorder: results of a controlled study. J Abnorm Child Psychol 2005;33:707–722. 12 Masia Warner C, Fisher PH, Shrout PE, Rathor S, Klein RG: Treating adolescents with social anxiety disorder in school: an attention control trial. J Child Psychol Psychiatry 2007; 48:676–686. 13 Garcia-Lopez LJ, Olivares J, Beidel D, Albano AM, Turner S, Rosa AI: Efficacy of three treatment protocols for adolescents with social anxiety disorder: a 5-year follow-up assessment. J Anxiety Disord 2006;20:175–191. 14 Rodebaugh TL, Rapee RM: Those who think they look worst respond best: self-observer discrepancy predicts response to video feedback following a speech task. Cognit Ther Res 2005;29:705–715. 15 Mortberg E, Clark DM, Sundin O, Wistedt AA: Intensive group cognitive treatment and individual cognitive therapy vs. treatment as usual in social phobia: a randomized controlled trial. Acta Psychiatr Scand 2007; 115: 142–154. 16 Stangier U, Heidenreich T, Peitz M, Lauterbach W, Clark DM: Cognitive therapy for social phobia: individual versus group treatment. Behav Res Ther 2003;41:991–1007. 17 Albano AM: Treatment of social anxiety in adolescents. Cogn Behav Pract 1995;2:271–298.
Received: December 19, 2012 Accepted after revision: July 25, 2013 Published online: November 19, 2013
A Randomized Controlled Trial of Individual Cognitive Therapy, Group Cognitive Behaviour Therapy and Attentional Placebo for Adolescent Social Phobia Jo Magne Ingul a, c Tore Aune c, d Hans M. Nordahl b, c
a
Department of Child and Adolescent Psychiatry, and b Department of Psychiatry, Levanger Hospital, Nord-Trøndelag Health Trust, Levanger, and c Department of Psychology, Norwegian University of Science and Technology, Trondheim, d University College of Nord Trøndelag, Levanger, Norway
Key Words Social phobia · Adolescents · Individual cognitive therapy · Group cognitive behavioural therapy
the most commonly used treatments for adolescent SP, we found that individual therapy was the most effective, yielding better effects than both CBTG and AP. © 2013 S. Karger AG, Basel
© 2013 S. Karger AG, Basel 0033–3190/14/0831–0054$39.50/0 E-Mail [email protected] www.karger.com/pps
Introduction
Social phobia (SP) or social anxiety disorder tends to follow a chronic and unremitting course [1]. It is often associated with educational and social impairment among adolescents, carrying with it an increased risk of suicide attempts and substance abuse in adults [2]. In recent decades, several studies have been published documenting the effect of cognitive behaviour therapy (CBT) treatment on children and adolescents with anxiety disorders [3, 4]. Reported treatment efficiency varies across studies, and Rao et al. [5] suggested that there is a need for specialized, disorder-specific treatment, but also saw a need for different treatment approaches across different age groups. Only 6 studies have specifically investigated the treatment of SP among adolescents [6–12]. All studies included some version of a group CBT (CBTG) format. The exception is the study of Herbert et al. [10], who compared individual and group therapeutic interventions based on Jo Magne Ingul Department of Child and Adolescent Psychiatry Levanger Hospital Kirkegt 2, NO–7600 Levanger (Norway) E-Mail jo.magne.ingul @ hnt.no
Downloaded by: Siriraj Medical Library, Mahidol University 202.28.191.34 - 2/16/2015 12:35:10 AM
Abstract Background: Very few studies have investigated the effects of individual disorder-specific treatment of social phobia (SP) in adolescents. The objective of this study was to compare the effects of individual cognitive therapy for SP, group cognitive behavioural therapy (CBTG) and attentional placebo (AP) among adolescents with a primary diagnosis of SP. Methods: A randomized controlled design was used, and a total of 279 adolescents were assessed. Fifty-seven adolescents, between 13 and 16 years old, were allocated to individual cognitive therapy, CBTG or AP. The participants were assessed before treatment, at the end of treatment and at a 12-month follow-up using both self-report and a semi-structured interview. Results: The individual cognitive therapy showed significant reductions in symptoms, impairment and diagnostic criteria both at the end of treatment and at the 12-month follow-up. Compared with CBTG and AP, the individual cognitive therapy group demonstrated significantly greater effects on both symptom reduction and impairment. There were no significant differences between CBTG and AP. Conclusions: In a direct comparison between
RCT of Adolescent Social Phobia
Methods Participants All participants were 8th- to 10th-grade students recruited from a school-based screening (n = 1,748). Out of the 279 assessed, 128 were found eligible for treatment, and 57 completed the treatment (fig. 1). The mean age of the participants who completed the study was 14.5 years, and 32 (56.1%) were girls. All participants had an SP diagnosis as their primary problem; 75.4% had generalized SP and 24.6% had a specific SP (specific SP was defined as a more circumscribed SP, where the adolescent showed severe impairment and symptoms of anxiety in 1–2 specific situations). Approximately 65% had at least 1 comorbid disorder, the most common being specific phobia (49%) and generalized anxiety disorder (28%; table 1). Procedure The study was approved by the Regional Ethical Committee and introduced to the school authorities in all counties by written information and meetings. Recruitment was performed in two waves using a two-stage process. Participants initially consented to participate in the first stage of recruitment and were required to give a second consent before being entered into the second stage. Students who, in the screening, scored above the clinically recommended cut-off on the Social Phobia and Anxiety Inventory for Children (SPAI-C) [23] (n = 277) and 56 randomly selected adolescents, to allow blind assessment, were invited to an interview (n = 333; fig. 1). The interviews were conducted using the Anxiety Disorders Interview Schedule for DSM-IV – Child version (ADIS-C) [24]. Students who fulfilled the inclusion criteria were informed about the treatment study and invited to participate. Adolescents who consented were given written information about the treatment study and a consent form to take home. They were then entered into the second stage of recruitment and randomized. After randomization, adolescents and their parents were contacted, asked for confirmation of their consent and provided additional information. Assessment was conducted before and after treatment and at a 12-month follow-up. Symptom measures were assessed at all 3 points. A clinical interview (ADIS-C) was conducted both before treatment and at the 12-month follow-up. At the follow-up interview, assessors were blinded with respect to treatment conditions, receiving only the name and contact information of each adolescent. Inclusion and Exclusion Criteria To qualify for inclusion, students had to be in grades 8–10 and experiencing SP as their primary problem. The presence of mental retardation or psychoses was considered a ground for exclusion. Participants who were already being treated elsewhere for mental health conditions were also excluded (fig. 1). Estimation of Sample Size The SPAI-C, the main outcome measure, has a maximum score of 52 points and a standard deviation (SD) of 7 points [25]. An effect size of 0.80 was the expected minimum clinical difference. With a 5% significance level and 80% power, 16 adolescents would be required in each group to detect significant differences. With an expected attrition rate of 25%, we decided that at least 20 adolescents would be required in each group.
Psychother Psychosom 2014;83:54–61 DOI: 10.1159/000354672
55
Downloaded by: Siriraj Medical Library, Mahidol University 202.28.191.34 - 2/16/2015 12:35:10 AM
the same protocol. Overall, the studies show significant reductions in both symptoms and recovery rates from pre-treatment to post-treatment [8–12]. Reported followup results indicate that changes were maintained for periods ranging from 6 months [10, 11] to 5 years [13]. The exception is Hayward et al. [9], who reported no significant effect of treatment at a 1-year follow-up. The results from these reported studies show promising results, but there are some potential limitations. The studies are relatively small, with limited statistical power [6, 7, 9]; moreover, one was an uncontrolled pilot study [6] and others compared CBT to a waiting list control [7, 9]. Most studies reported large dropout rates, and in one study, treatment rejecters were included in the control condition [8]. Among adults, both individual and group therapy formats have been reported to be effective [14]. Rodebaugh and Rapee [14] found no difference in outcome between therapy formats. Other studies have reported individual therapy to be superior [15, 16]. Herbert et al. [10] examined differences between therapy formats in adolescents with SP. In this study, the same protocol [17] was used for both individual and group therapy. Both conditions were associated with significant reductions in symptoms and functional impairment, with only small differences between the formats. To our knowledge, no study has reported on adolescent SP using a manual specifically designed for individual therapy. Clark and Wells [18] developed a cognitive therapy for SP in adults. This model focuses on the cognitive processes that maintain SP. Results of the treatment are promising in adults [16, 19]. The status of the Clark and Wells model in children and adolescents is not established; however, studies providing evidence for elements in the model have been reported [20, 21], but also one study disfavouring the validity of the model [22]. In summary, treatment of SP in adolescents has been dominated by studies using a group therapy format. No study has ever investigated individual treatment of SP in adolescents with a manual specially designed for the format. Currently, the effectiveness of the Clark and Wells [18] treatment in adolescents is unknown. In this study, we wanted to test the effectiveness of the Clark and Wells treatment in adolescents, comparing it to the best treatment available, CBTG. To control for the effect of social exposure, a group receiving attentional placebo (AP) was included. With this background, we predicted that (1) individual therapy based on Clark and Wells’ treatment would prove as effective as CBTG in treating adolescents with SP and (2) that both individual cognitive therapy and CBTG would be more effective than AP.
Total screened n = 1,748
Refused to participate n = 52 Other reasons n = 2 (moved)
Potential participants n = 333
Assessed for eligibility n = 279
Did not meet inclusion criteria n = 125 Met exclusion criteria n = 11
Randomized n = 128
Did not want treatment n = 15
Allocated to AP n = 34 No show n = 16
Allocated to CBTG n = 58 No show n = 25
Allocated to CBTI n = 36 No show n = 11
Post-treatment assessment completed n = 16 No data n=2
Post-treatment assessment completed n = 20 No data n = 13
Post-treatment assessment completed n = 21 No data n=4
Follow-up completed (12 months) n = 15
Follow-up completed (12 months) n = 12
show: adolescents who initially agreed to participate but later declined, before treatment started.
Randomization The randomization was performed at each site, because of long travelling distances. Randomization was conducted using a preassigned random schedule generated from the SPSS 15.0 random number generator. Measures Structured Interview The ADIS-C is a semi-structured interview for anxiety disorders and the most common comorbid disorders in children and adolescents using DSM-IV criteria [24]. The ADIS-C has generally shown good to excellent reliability. The ADIS-C has been found to be sensitive to treatment effects [26, 27]. Diagnoses were assessed by clinical psychology graduate students and psychiatric nurses trained by the authors. Following the training period, and before they qualified as assessors, 25% of all training interviews were randomly selected and reassessed. Kappa for the social anxiety diagnosis of the ADIS-C was 0.75.
56
Psychother Psychosom 2014;83:54–61 DOI: 10.1159/000354672
Primary Outcome Measures SPAI-C. The SPAI-C [23] is a 26-item self-report inventory designed to assess symptoms of DSM-IV SP. It has shown good internal consistency, test-retest reliability [23, 28] and construct validity [28]. Cronbach’s α for the SPAI-C in this study was 0.95. The Social Thoughts and Beliefs Scale. The Social Thoughts and Beliefs Scale (STABS) [29] is a 21-item self-report inventory designed to assess cognition in persons with SP. It has shown excellent internal consistency, test-retest reliability and discriminant validity [29, 30]. Acceptable concurrent validity has also been documented [30]. Chronbach’s α for STABS in this study was 0.96. Secondary Outcome Measures Anxiety. The Screen for Child Anxiety Related Emotional Disorders (SCARED) [31] is a 41-item self-report measure appropriate for use with children and adolescents 8–18 years old. It has demonstrated very good psychometric properties in a clinical sample [31] and in a Norwegian community sample [32]. Cronbach’s
Ingul/Aune/Nordahl
Downloaded by: Siriraj Medical Library, Mahidol University 202.28.191.34 - 2/16/2015 12:35:10 AM
Fig. 1. Flow chart of study participants. No
Table 1. Demographic and baseline data by randomized condition
CBTI (n = 21) Female Age, years Living with sibling Yes No Missing data
12 (57.14) 14.98±0.94
CBTG (n = 20)
AP (n = 16)
12 (60.00) 14.30±0.89
8 (50.00) 14.16±1.08
Total (n = 57) 32 (56.14) 14.50±1.01
15 (71.43) 6 (28.57) 0
13 (65.00) 5 (25.00) 2 (10.00)
11 (68.75) 4 (25.00) 1 (6.25)
39 (68.42) 15 (26.32) 3 (5.26)
Living with Mother and father Mother only Father only Mother and father, but not together Missing data
9 (42.86) 7 (14.33) 2 (9.52) 3 (14.29) 0
13 (65.00) 4 (20.00) 1 (5.00) 0 2 (10.00)
13 (81.25) 1 (6.25) 0 1 (6.25) 1 (6.25)
35 (61.40) 12 (21.05) 3 (5.26) 4 (7.02) 3 (5.26)
School grade 8 9 10
6 (28.57) 7 (33.33) 8 (38.10)
6 (30.00) 9 (45.00) 5 (25.00)
8 (50.00) 4 (25.00) 3 (18.75)
20 (35.09) 20 (35.09) 16 (28.07)
16 (76.19)
15 (75.00)
12 (75.00)
43 (75.44)
Generalized SP Interference rating before treatment Comorbidity Specific phobia GAD Agoraphobia Separation anxiety OCD ADD Depression PTSD
4.48±1.37 15 (71.43) 9 (42.86) 7 (33.33) 0 1 (4.76) 0 3 (14.29) 2 (9.52) 1 (4.76)
4.65±1.39 12 (60.00) 9 (45.00) 6 (30.00) 4 (20.00) 0 0 1 (5.00) 2 (10.00) 0
4.25±1.00 10 (62.50) 10 (62.50) 3 (18.75) 0 2 (12.50) 1 (6.25) 1 (6.25) 1 (6.25) 1 (6.25)
4.47±1.27 37 (64.91) 28 (49.12) 16 (28.07) 4 (7.02) 3 (5.26) 1 (1.75) 5 (8.77) 5 (8.77) 2 (3.50)
α in this study was 0.93 for the total SCARED and 0.80 for the social anxiety disorder subscale (SCARED-SAD-sub). The SCARED assesses the DSM-IV symptoms of 4 different diagnoses, and it also provides total and school avoidance scores. Depression. The Children’s Depression Inventory (CDI) is a multidimensional measure quantifying depressive symptomatology in children and adolescents [33]. It has 27 items in which the child must choose 1 of 3 levels and yields scores for 5 factors and a total score. The CDI has shown good psychometric properties [33]. Cronbach’s α for the CDI in this study was 0.98. Quality of Life. To measure the perceived quality of life, the Pediatric Quality of Life Inventory (PedsQL) 4.0 [34] was used. The PedsQL consists of 23 items covering 4 domains – physical, emotional, social and school functioning – and calculates a total quality of life score. The items are scored on a 5-point Likert scale and linearly transformed to a 0–100 scale. The scale has shown adequate psychometric properties [34].
Interventions Treatment and Therapists The AP groups were led by experienced milieu therapists, most of whom were psychiatric nurses and social workers with extensive clinical experience. CBTG was led and supervised by an experienced clinical psychologist (second author) and a team of 4 cotherapists. In the individual CBT condition (CBTI), treatment was led and supervised by a clinical psychologist (first author) and 2 co-therapists. In both CBTG and CBTI conditions, independent experts assessed the therapies using video and audiotapes. They focused on treatment integrity, adherence to protocol and competence of the therapists. No protocol violations were reported, and the competence of the therapists was reported to be high overall.
RCT of Adolescent Social Phobia
Psychother Psychosom 2014;83:54–61 DOI: 10.1159/000354672
Attentional Placebo AP was conducted in groups over 10 sessions of 90 min each. The groups consisted of 4–6 participants. This condition did not include any of the active components in the CBTI or CBTG.
57
Downloaded by: Siriraj Medical Library, Mahidol University 202.28.191.34 - 2/16/2015 12:35:10 AM
Values represent numbers of participants (percentages) or means ± SD, as appropriate. GAD = Generalized anxiety disorder; OCD = obsessive compulsive disorder; ADD = attention deficit disorder; PTSD = post-traumatic stress disorder.
Through social activity, social interaction and social support, the adolescents were exposed to adults, peers and social attention to the same degree as in the other conditions.
Table 2. Outcome measures for the 3 intervention groups before
Group CBT The CBTG was conducted over 10 sessions of 90 min each. The groups consisted of 4–6 participants. The manual was based on The C.A.T. Project Manual for the Cognitive Behavioral Treatment of Anxious Adolescents [35] with some ‘play elements’ from the Social Effectiveness Therapy for Children and Adolescents program [36]. The manual was divided into two parts (online suppl. table 1; for all online suppl. material, see www.karger.com/ doi/10.1159/000354672). Prior to the study, the protocol was tested by the second author in a pilot study, showing significant symptom reductions [37].
CBTI (n = 21) CBTG (n = 20) AP (n = 16)
Individual CBT The CBTI was conducted over 12 sessions of 50 min each. The manual was developed by Clark and Wells [18] as a treatment for adult SP. The language, tempo and type of interventions were adapted for use with adolescents by the first and third authors. The treatment included 3 phases (online suppl. table 1). Statistical Analysis The demographic data were analysed using χ2 tests and analysis of variance. The effect of different treatment conditions on adolescent SP was tested with paired sample t tests. Analysis of clinically significant change was performed using the reliable change index (RCI) [38]. The RCI method evaluates the difference between pretest and post-test scores in a single case, using published norms for clinical and nonclinical populations for the relevant disorder. Values greater than 1.96 are considered to represent statistically reliable change [38]. The values used to calculate RCI here were derived from Aune et al. [25]. Between-group comparisons were performed using factorial analysis of co-variance (ANCOVA) and independent-sample t tests. Factorial ANCOVA assesses group differences on a single dependent variable after the effects of one or more known covariates are statistically removed, thereby minimizing error variance. When there are more than two groups, planned comparisons can be applied, assessing differences in means on the dependent variable using pooled group means adjusted for differences in covariates [39].
SPAI-C Pre Post FU FU ITT STABS Pre Post FU FU ITT SCARED-SAD-sub Pre Post FU FU ITT Interference Pre FU SCARED total Pre Post FU CDI Pre Post FU PedsQL Pre Post FU
26.81±6.77 7.86±5.76 10.64±7.90 10.29±6.68
22.20±9.16 23.16±9.35 18.47±9.99 18.35±9.06
22.40±9.24 17.63±9.49
48.10±13.70 21.12±13.82 22.92±17.87 23.81±15.72
40.41±12.04 36.78±15.94 39.65±15.73 37.70±14.39
40.85±9.37 31.98±10.83
7.90±2.26 3.81±3.63 4.83±3.76 4.81±3.66
6.50±2.80 6.05±3.79 7.00±3.40 6.47±3.39
5.81±2.64 5.67±2.64
4.48±1.37 0.91±1.45
4.65±1.39 2.67±2.47
4.25±1.00
27.72±12.72 14.45±8.45 14.33±12.81
22.60±12.52 25.16±17.10 22.47±14.32
23.13±15.30 22.87±12.84
21.34±3.61 10.24±9.70 12.17±9.48
18.42±11.95 16.35±11.48 14.13±8.85
12.60±7.22 7.04±4.62
34.91±14.39 18.88±12.57 19.35±14.60
27.46±12.46 28.98±16.56 29.55±16.58
24.64±11.18 21.38±11.73
Values represent means ± SD. Pre = Before treatment; Post = after treatment; FU = 12 months after treatment; ITT = intention to treat.
Pre-Treatment Comparisons There were no significant differences between the 3 intervention groups on demographic variables except age, where there was a significant difference [F(2, 52) = 3.72, p < 0.05] between the CBTI group (mean = 14.98, SD = 0.94) and the AP group (mean = 14.16, SD = 1.08; table 1). Comparison of the groups with regard to pre-treatment symptom severity showed no significant differences on the main outcome measures of social anxiety, type of Psychother Psychosom 2014;83:54–61 DOI: 10.1159/000354672
Treatment Attrition A χ2 test showed that the assigned treatment condition did not significantly influence whether the adolescents started treatment or not (χ2 = 1.97, p > 0.05). There was a significant association between the type of treatment received and whether or not they completed treatment (χ2 = 7.79, p < 0.01), with significantly more adolescents completing the CBTI (n = 21, 84.0%) and AP (n = 16, 88.9%) than the CBTG (n = 20, 58.8%). There were no significant Ingul/Aune/Nordahl
Downloaded by: Siriraj Medical Library, Mahidol University 202.28.191.34 - 2/16/2015 12:35:10 AM
social anxiety disorder diagnosis (generalized or specific) or the interference of anxiety symptoms in participants’ daily life (tables 1, 2).
Results
58
treatment, after treatment, at follow-up and intention-to-treat follow-up
Changes in Symptom Severity Within-Condition Changes For the AP group, results were only analysed at the end of treatment. Paired-sample t tests showed that the AP group had a significant reduction in symptoms as measured by the SPAI-C [t(14) = 2.37, p < 0.05] and the STABS [t(13) = 3.61, p < 0.01; online suppl. table 2]. For the CBTG group, there were no significant changes from pre- to post-treatment. At follow-up, there were significant changes in symptoms on the SPAI-C [t(14) = 2.45, p < 0.05] compared to pre-treatment. For CBTI, significant reductions were seen in all 3 social anxiety measures from pre- to post-treatment [SPAI-C: t(20) = 8.47, p < 0.001; STABS: t(20) = 8.12, p < 0.001; SCARED-SAD-sub: t(20) = 4.74, p < 0.001] and from pre-treatment to followup [SPAI-C: t(11) = 4.04, p < 0.01; STABS: t(11) = 3.64, p < 0.01; SCARED-SAD-sub: t(11) = 3.21, p < 0.01; online suppl. table 2]. RCI analysis for participants in the CBTI group showed that 69.8% at the end of treatment and 51.4% at follow-up had an RCI indicating clinically significant improvement. In the CBTG group, 21.1% at the end of treatment and 13% at follow-up had an RCI indicating clinically significant improvement. In the AP group, 27.8% had an RCI indicating clinically significant improvement at the end of treatment. Between-Condition Comparisons Differences between treatment conditions were tested using ANCOVA with planned contrasts. Age, gender and pre-treatment social anxiety and depression levels were used as covariates. There were significant effects of treatment conditions on all 3 measures [SPAI-C: F(2, 44) = 18.60, p < 0.001, d = 2.08; STABS: F(2, 45) = 6.48, p < 0.01, d = 1.07; SCARED-SAD-sub: F(2, 48) = 3.66, p < 0.05, d = 0.81] when the covariates were controlled for. Covariates had no significant effect. Planned contrasts showed the significant differences to be between CBTI and CBTG [SPAI-C: t(35) = 6.34, p < 0.001; STABS: t(35) = 4.18, p < 0.001; SCARED-SAD-sub: t(35) = 2.52, p < 0.05; online suppl. table 3] and between CBTI and AP [SPAI-C: t(32) = 4.25, p < 0.001; STABS: t(32) = 2.07, p < 0.01; SCARED-SAD-sub: t(31) = 2.19, p < 0.05] on all 3 measures. At follow-up, independent-sample t tests showed that there were significant differences between CBTI and RCT of Adolescent Social Phobia
CBTG on 2 of the 3 social anxiety measures [SPAI-C: t(25) = 2.22, p < 0.05; STABS: t(25) = 2.59, p < 0.05; online suppl. table 3]. Intention-to-Treat Analysis Because of missing data at follow-up (34%), intentionto-treat analyses using the last observation carried forward method was carried out. Both within- and betweencondition (CBTI vs. CBTG) analyses using intention-totreat data confirmed the results of the completer analysis (online suppl. tables 2, 3). Recovery and Impairment Ratings Prior to treatment, all participants were diagnosed with SP. At follow-up, 72.7% of participants in the CBTI group and 53.3% of those in the CBTG group had no SP diagnosis. The difference between the 2 groups was not significant. Ratings of impairment, the way in which symptoms interfere with daily lives, showed that there was a significant difference between the CBTI and CBTG groups [t(23.09) = 2.28, p < 0.05] at follow-up. Both conditions showed significant reductions in interference from pre-treatment to follow-up [CBTI: t(10) = 4.98, p < 0.001; CBTG: t(14) = 3.87, p < 0.01; online suppl. table 2]. Secondary Measures The SCARED total score, the CDI and the PedsQL were also tested within each condition. The analysis showed that there were no significant changes at any time for the CBTG group. For the AP group, there were significant changes from pre-treatment to post-treatment on the CDI [t(14) = 2.93, p < 0.05] and PedsQL [t(13) = 2.18, p < 0.05; online suppl. table 2]. All 3 secondary measures were significantly different between pre-treatment and post-treatment in the CBTI condition [SCARED: t(16) = 3.59, p < 0.01; CDI: t(20) = 5.01, p < 0.001; PedsQL: t(20) = 4.32, p < 0.001]. There was also a significant change from pre-treatment to follow-up on the CDI [t(11) = 3.46, p < 0.01].
Discussion
Overall, the results indicate that individual cognitive therapy (CBTI) yields better effects than both CBTG and AP. In-Albon and Schneider [4] reported the average effect size of CBT on symptoms of SP to be 0.80, while Segool and Carlson [40] reported effect sizes of 0.86 on symptoms and 1.56 on impairment. Our results for CBTI were 2.07 and 2.49. Moreover, a recovery rate of 73% is Psychother Psychosom 2014;83:54–61 DOI: 10.1159/000354672
59
Downloaded by: Siriraj Medical Library, Mahidol University 202.28.191.34 - 2/16/2015 12:35:10 AM
differences on any of the clinical measures [SPAI-C: t(124) = 0.48, p > 0.05; STABS: t(117) = 0.57, p > 0.05] between those who did not start treatment and those who finished the treatment.
60
Psychother Psychosom 2014;83:54–61 DOI: 10.1159/000354672
with respect to the conclusions, as noted by Tomba and Bech [44]. However, the main measures, the SPAI-C [45] and the ADIS-C [26, 27], have been shown to be sensitive to change. Assessment was done with adolescents as the only informants. A multi-informant approach can increase the reliability of the diagnoses. However, there is evidence that adolescents should be the primary source when diagnosing SP [46]. Also, Silverman et al. [3] found that efficacy rates of CBT are much larger based on parent (0.91) than child (0.44) ratings. This indicates that children experience less change than their parents. The implication for this study is that we are more likely to underestimate than overestimate change. In conclusion, with the limitations of the study in mind, working with negative perceptual biases maintaining SP seems to provide relatively strong and enduring results. Our results also indicate that individual therapy with a protocol specifically designed to face idiosyncratic problems may be applicable in the treatment of adolescent SP. Both individual and group cognitive therapy are effective, but the study also demonstrates that it is important to control for the effect of natural self-exposure in comparative studies, as seen in the results of the AP condition in the current study.
Acknowledgements This study was supported by a grant from the Liaison Committee between the Central Norway Regional Health Authority and the Norwegian University of Science and Technology.
Disclosure Statement No author or immediate family member has financial arrangements that might represent potential conflicts of interest for the findings presented, nor has any funding source influenced the content of this report.
References
1 Kessler RC, Berglund P, Demler O, Jin R, Merikangas KR, Walters EE: Lifetime prevalence and age-of-onset distributions of DSMIV disorders in the National Comorbidity Survey replication. Arch Gen Psychiatry 2005;62:593–602. 2 Stein MB, Stein DJ: Social anxiety disorder. Lancet 2008;371:1115–1125. 3 Silverman WK, Pina AA, Viswesvaran C: Evidence-based psychosocial treatments for phobic and anxiety disorders in children and adolescents. J Clin Child Adolesc Psychol 2008;37:105–130.
Ingul/Aune/Nordahl
Downloaded by: Siriraj Medical Library, Mahidol University 202.28.191.34 - 2/16/2015 12:35:10 AM
relatively large and in line with the best reported results [3, 4]. To our knowledge, this is the first randomized controlled trial of individual cognitive therapy based on the Clark and Wells [18] treatment for SP in adolescents. The results may be important as they indicate that working on the negative perceptual biases is useful also in adolescents. They may also be important as recent research [41] has shown that treatment with antidepressants has been associated with a substantial risk of adverse effects, including excessive mood elevation and behavioural activation, making it even more important to develop and document good psychotherapeutic methods and ensuring that mental health professionals are trained to deliver these [42]. For the CBTG group, there were significant reductions in both symptoms and interference from pre-treatment to follow-up, which was in accordance with our hypothesis. A long-term treatment efficacy of 53% is in line with what has generally been reported among adolescents [8– 12]. However, there were no significant changes from pre-treatment to post-treatment, and overall, the changes were lower than expected. The results of the study indicate that CBTG is not as effective as CBTI. This could be a format effect, given that individual therapy has been shown to be more effective than group therapy in some adult studies [15, 16]. However, childhood anxiety studies have shown that there are no differences between group and individual formats [4, 10, 43]. This indicates that the format of the therapy is not an entirely sufficient explanation of the differences in the present study. CBTG is based on standard CBT, utilizing standard techniques and understanding of SP, while CBTI makes assumptions with regard to specific maintaining factors, targeting these in treatment. Different models may therefore be an explanation of the differences in results. The study gives several very interesting findings but they have to be interpreted with caution, and there are some limitations that need consideration. As Herbert et al. [10] report, high attrition rates and small samples are the norm in studies of adolescent SP. Although limited, our sample size is larger than estimated to be necessary to have adequate statistical power. Even though we tried to minimize the risk and effects of dropout, a third of those who completed the treatment failed to participate at follow-up. We cannot rule out the possibility that the results may be inflated, as we might expect the successful participants to show up. The present study relied on self-report and clinical interviews for measuring change; this warrants caution
RCT of Adolescent Social Phobia
18 Clark DM, Wells A: A cognitive model of social phobia; in Heimberg RG, Liebowitz MR, Hope DA, Schneier FR (eds): Social Phobia: Diagnosis, Assessment, and Treatment. New York, Guilford Press, 1995, pp 69–93. 19 Clark DM, Ehlers A, Hackmann A, McManus F, Fennell M, Grey N, Waddington L, Wild J: Cognitive therapy versus exposure and applied relaxation in social phobia: a randomized controlled trial. J Consult Clin Psychol 2006;74:568–578. 20 Morgan J, Banerjee R: Social anxiety and selfevaluation of social performance in a nonclinical sample of children. J Clin Child Adolesc Psychol 2006;35:292–301. 21 Cowart MJW, Ollendick TH: Attention training in socially anxious children: a multiple baseline design analysis. J Anxiety Disord 2011;25:972–977. 22 Alfano CA, Beidel DC, Turner SM: Negative self-imagery among adolescents with social phobia: a test of an adult model of the disorder. J Clin Child Adolesc Psychol 2008; 37: 327–336. 23 Beidel DC, Turner SM, Morris TL: A new inventory to assess childhood social anxiety and phobia: The Social Phobia and Anxiety Inventory for Children. Psychol Assess 1995; 7:73–79. 24 Albano AM, Silverman WK: Anxiety Disorders Interview Schedule for DSM-IV: Child Version Clinician Manual. New York, Oxford University Press, 1996. 25 Aune T, Stiles TC, Svarva K: Psychometric properties of the Social Phobia and Anxiety Inventory for children using a non-American population-based sample. J Anxiety Disord 2008;22:1075–1086. 26 Silverman WK, Saavedra LM, Pina AA: Testretest reliability of anxiety symptoms and diagnoses with anxiety disorders interview schedule for DSM-IV: child and parent versions. J Am Acad Child Adolesc Psychiatry 2001;40:937–944. 27 Kendall PC, Flannery-Schroeder E, Panichelli-Mindel SM, Southam-Gerow M, Henin A, Warman M: Therapy for youths with anxiety disorders: a second randomized clincal trial. J Consult Clin Psychol 1997;65:366–380. 28 Beidel DC, Turner SM, Hamlin K, Morris TL: The Social Phobia and Anxiety Inventory for Children (SPAI-C): external and discriminative validity. Behav Ther 2000;31:75–87. 29 Turner SM, Johnson MR, Beidel DC, Heiser NA, Lydiard RB: The Social Thoughts and Beliefs Scale: a new inventory for assessing cognitions in social phobia. Psychol Assess 2003; 15:384–391. 30 Fergus TA, Valentiner DP, Kim HS, Stephenson K: The Social Thoughts and Beliefs Scale: psychometric properties and its relation with interpersonal functioning in a non-clinical sample. Cognit Ther Res 2009;33:425–431.
31 Birmaher B, Khetarpal S, Brent DA, Cully M, Balach L, Kaufman J, Neer SM: The Screen for Child Anxiety Related Emotional Disorders (SCARED): scale construction and psychometric characteristics. J Am Acad Child Adolesc Psychiatry 1997;36:545–553. 32 Leikanger E, Ingul JM, Larsson B: Sex and age-related anxiety in a community sample of Norwegian adolescents. Scand J Psychol 2012; 53:150–157. 33 Kovacs M: CDI: Children’s Depression Inventory: Technical Manual Update. Toronto, MHS, 2003. 34 Varni JW, Seid M, Kurtin PS: PedsQL 4.0: reliability and validity of the Pediatric Quality of Life Inventory version 4.0 generic core scales in healthy and patient populations. Med Care 2001;39:800–812. 35 Kendall PC, Choudhury M, Hudson JL, Webb A: The C.A.T. Project Manual for the Cognitive Behavioral Treatment of Anxious Adolescents. Philadelphia, Temple University, 2000. 36 Beidel DC, Turner SM, Morris TL: Social Effectiveness Therapy for Children and Adolescents. Toronto, MHS, 2004. 37 Aune T: Sosial angstlidelse hos barn og unge. Tidsskr Norsk Psykol 2011;48:12–16. 38 Jacobson NS, Truax P: Clinical significance: a statistical approach to defining meaningful change in psychotherapy research. J Consult Clin Psychol 1991;59:12–19. 39 Tabachnik BG, Fidell LS: Using Multivariate Statistics, ed 5. Boston, Pearson International, 2007. 40 Segool N, Carlson JS: Efficacy of cognitivebehavioral and pharmacological treatments for phobic and anxiety disorders in children and adolescents. Depress Anxiety 2008; 25: 620–631. 41 Offidani E, Fava GA, Tomba E, Baldessarini RJ: Excessive mood elevation and behavioural activation with antidepressant treatment of juvenile depressive and anxiety disorders: a systematic review. Psychother Psychosom 2013;82:132–141. 42 Leckman JF: The risk and benefits of antidepressants to treat pediatric-onset depression and anxiety disorders: a developmental perspective. Psychother Psychosom 2013; 82: 129–131. 43 Flannery-Schroeder EC, Kendall PC: Group and individual cognitive-behavioral treatments for youth with anxiety disorders: a randomized clinical trial. Cognit Ther Res 2000; 24:251–278. 44 Tomba E, Bech P: Clinimetrics and clinical psychometrics: macro- and microanalysis. Psycother Psychosom 2012;81:333–343. 45 Beidel DC, Turner SM, Morris TL: SPAI-C. Social Phobia and Anxiety Inventory for Children. Manual. New York, MHS, 2000. 46 La Greca AM, Lopez N: Social anxiety among adolescents: linkages with peer relations and friendships. J Abnorm Child Psychol 1998;26: 83–94.
Psychother Psychosom 2014;83:54–61 DOI: 10.1159/000354672
61
Downloaded by: Siriraj Medical Library, Mahidol University 202.28.191.34 - 2/16/2015 12:35:10 AM
4 In-Albon T, Schneider S: Psychotherapy of childhood anxiety disorders: a meta-analysis. Psychother Psychosom 2007;76:15–24. 5 Rao PA, Beidel DC, Turner SM, Ammerman RT, Crosby LE, Sallee FR: Social anxiety disorder in childhood and adolescence: descriptive psychopathology. Behav Res Ther 2007; 45:1181–1191. 6 Albano AM, Marten PA, Holt CS, Heimberg RG: Cognitive-behavioral group treatment for social phobia in adolescents: a preliminary study. J Nerv Ment Dis 1995;183:649–656. 7 Baer S, Garland EJ: Pilot study of communitybased cognitive behavioral group therapy for adolescents with social phobia. J Am Acad Child Adolesc Psychiatry 2005;44:258–264. 8 Garcia-Lopez LJ, Olivares J, Turner SM, Beidel DC, Albano AM, Sanchez-Meca J: Results at long-term among three psychological treatments for adolescents with generalized social phobia. II. Clinical significance and effect size. Psicol Conductual 2002;10:371–385. 9 Hayward C, Varady S, Albano AM, Thienemann M, Henderson L, Schatzberg AF: Cognitive-behavioral group therapy for social phobia in female adolescents: results of a pilot study. J Am Acad Child Adolesc Psychiatry 2000;39:721–726. 10 Herbert JD, Gaudiano BA, Rheingold AA, Moitra E, Myers VH, Dalrymple KL, Brandsma LL: Cognitive behavior therapy for generalized social anxiety disorder in adolescents: a randomized controlled trial. J Anxiety Disord 2009;23:167–177. 11 Masia Warner C, Klein RG, Dent HC, Fisher PH, Alvir J, Albano AM, Guardino M: Schoolbased intervention for adolescents with social anxiety disorder: results of a controlled study. J Abnorm Child Psychol 2005;33:707–722. 12 Masia Warner C, Fisher PH, Shrout PE, Rathor S, Klein RG: Treating adolescents with social anxiety disorder in school: an attention control trial. J Child Psychol Psychiatry 2007; 48:676–686. 13 Garcia-Lopez LJ, Olivares J, Beidel D, Albano AM, Turner S, Rosa AI: Efficacy of three treatment protocols for adolescents with social anxiety disorder: a 5-year follow-up assessment. J Anxiety Disord 2006;20:175–191. 14 Rodebaugh TL, Rapee RM: Those who think they look worst respond best: self-observer discrepancy predicts response to video feedback following a speech task. Cognit Ther Res 2005;29:705–715. 15 Mortberg E, Clark DM, Sundin O, Wistedt AA: Intensive group cognitive treatment and individual cognitive therapy vs. treatment as usual in social phobia: a randomized controlled trial. Acta Psychiatr Scand 2007; 115: 142–154. 16 Stangier U, Heidenreich T, Peitz M, Lauterbach W, Clark DM: Cognitive therapy for social phobia: individual versus group treatment. Behav Res Ther 2003;41:991–1007. 17 Albano AM: Treatment of social anxiety in adolescents. Cogn Behav Pract 1995;2:271–298.
