Hindawi

Psychiatry Journal
Volume 2020, Article ID 3186450, 7 pages
https://doi.org/10.1155/2020/3186450

Research Article
The Relationship between Symptoms and Social Functioning over
the Course of Cognitive Behavioral Therapy for Social
Anxiety Disorder

Sei Ogawa ,1,2 Risa Imai,2 Masako Suzuki,2 Toshi A. Furukawa,3 and Tatsuo Akechi2
1
Graduate School of Humanities and Social Sciences, Nagoya City University, Nagoya, Japan
2
Department of Psychiatry and Cognitive-Behavioral Medicine, Nagoya City University Graduate School of Medical Sciences,
Nagoya, Japan
3
Department of Health Promotion and Human Behavior, Kyoto University Graduate School of Medicine/School of Public Health,
Kyoto, Japan

Correspondence should be addressed to Sei Ogawa; [email protected]

Received 22 May 2020; Revised 3 August 2020; Accepted 22 September 2020; Published 28 September 2020

Academic Editor: Luca Ferraro

Copyright © 2020 Sei Ogawa et al. This is an open access article distributed under the Creative Commons Attribution License,
which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Background. The present study is aimed at investigating the relationship between changes in symptoms and changes in social
functioning during cognitive behavioral therapy (CBT) for social anxiety disorder (SAD). Methods. Ninety-six patients with
SAD were treated with manualized group CBT. Measures of social anxiety symptoms, depression symptoms, cognition, and
social functioning were administered at baseline and endpoint. Using multiple regression analysis, we examined the associations
between the changes in four aspects (work, home management, social leisure activities, and private leisure activities) of social
functioning as dependent variables and the changes in four factors (social interaction, public speaking, observation by others,
and eating and drinking in public) in social anxiety symptoms, depression symptoms, and cognition as independent variables.
Results. The changes in work functioning were predicted by the changes in the public speaking factor in social anxiety
symptoms. The changes in depression symptoms predicted the changes in home management. The significant predictors of
changes in social leisure activities were the changes in the social interaction factor and depression symptoms. The changes in
private leisure activities were predicted by the changes in the observation by others factor. The changes in cognition predicted
nothing. Conclusion. The present study suggested that the changes in social anxiety or depression symptoms may predict several
aspects of social functioning changes in patients with SAD over the course of CBT. In order to improve social functioning, our
results may be useful for selecting the fear or feared situation in CBT for SAD. Trial Registration. The clinical study registration
number in the Japanese trials registry is UMIN CTR 000031147.

1. Background ders (dysthymia, major depressive episode, posttraumatic

stress disorder, panic disorder, and SAD) [5]. Improvements
Social anxiety disorder (SAD) is a serious disabling disease in social functioning, as well as symptomatic relief, are
associated with remarkable impairment in quality of life important outcomes for the treatment of SAD.
(QOL) or social functioning [1]. Patients with SAD are sig- A large number of randomized controlled trials support
nificantly impaired in social functioning compared with the the use of cognitive behavioral therapy (CBT) for SAD [6].
general population [2]. In addition, patients affected by Meta-analyses have concluded that there is strong empirical
SAD are more financially dependent [3], underemployed, support for the efficacy of CBT for SAD [7]. Moreover, sev-
and less productive at work compared with those without eral studies mentioned that the effects of CBT may extend
the disorder [4]. The impairment caused by SAD is very high, beyond a reduction in social anxiety symptoms to improve-
and SAD is among the five most impairing psychiatric disor- ments in social functioning [8–10]. Although anxiety
2 Psychiatry Journal

symptoms might respond to CBT for anxiety disorder in the 2.2. Treatment. We followed the group CBT manual for SAD
short term, longer treatments seem to be needed to improve developed by Andrews et al. [16] and integrated the Clark
QOL or social functioning. In order to explain this difference, and Wells model of SAD [17] into the manual. The therapists
an analysis of the QOL or social functioning construct in had group discussions once a month to check on therapist
CBT is required [11]. However, there are few studies in sup- adherence to the CBT program and to plan for future ses-
port of the specific mechanisms of change in social function- sions. The program consisted of sixteen weekly sessions by
ing during CBT. two therapists, who were psychiatrists or clinical psycholo-
The cognitive behavioral model is based upon the gists with more than two years of clinical experience. Each
assumption that our cognition influences our behavior or session was scheduled to last 120 minutes. The program
emotions. CBT focuses on unhelpful cognition and behaviors included (i) psychoeducation about SAD including the Clark
and improves QOL or social functioning as mentioned. In and Wells cognitive behavioral model, (ii) behavioral experi-
the previous research of anxiety disorder including panic dis- ments to examine the participant’s catastrophic predictions,
order, generalized anxiety disorder, and posttraumatic stress (iii) attention training to shift focus from unhelpful thoughts
disorder, QOL may change as a result of, or simultaneously and feeling to the task or the external situation, (iv) cognitive
with, anxiety symptom changes, not cognitive changes [12]. restructuring to identify and dispute maladaptive thoughts,
For the purpose of improving QOL or social functioning in and (v) in vivo graded exposures to anxiety-provoking
CBT for these disorders, it may be useful to focus on achiev- situations.
ing symptom improvements in preference to cognitive
changes. However, this study did not include SAD. There- 2.3. Measurements. The patients were assessed with an exten-
fore, additional research is needed to clarify the mechanisms sive questionnaire battery using observer-rated assessments
of changes in social functioning in CBT for SAD. Moreover, and self-report questionnaires at baseline and endpoint in
CBT for SAD is adjusted to particular fears or feared situa- our institute. We assessed the Liebowitz Social Anxiety Scale
tions in daily life. These fears or situations are chosen based (LSAS), the Brief Fear of Negative Evaluation Scale (BFNE),
on what is treatable or the purpose of the treatment. It is the Symptom Checklist-90 Revised (SCL-90-R), and the
important to select the fears or situations for the effective Work, Home Management, Social and Private Leisure Activ-
CBT for SAD. The previous study showed that patients with ities Scale (WHLS).
SAD experienced improved social functioning after CBT
2.3.1. LSAS. The LSAS is a widely used clinician-
[13]. However, there is no study to examine the relationship
administered psychological scale for assessment of SAD
between social fear symptoms and social functioning during
[18]. It is a 24-item scale that is rated in terms of both fear
CBT.
(0-3 indicate none, mild, moderate, and severe, respectively)
The present study is aimed at investigating how changes
and avoidance (0-3 indicate never, occasionally, often, and
in social functioning related to changes in symptoms of
usually, respectively) of performance situations and social
SAD over the course of CBT.
interaction. Moreover, separate exploratory common factor
analyses of the fear and avoidance ratings yielded four similar
2. Methods factors for each: (1) social interaction, (2) public speaking, (3)
observation by others, and (4) eating and drinking in public
This prospective and observational study enrolled consecu-
[19]. The reliability and validity of the Japanese version have
tive SAD patients treated with CBT in our institute. This
been established [20]. The clinicians in charge of the CBT
study was conducted as a single-arm and naturalistic study
program administered LSAS at baseline and endpoint.
in a routine clinical setting.
2.3.2. BFNE. The BFNE is a 12-item self-reported instrument
2.1. Patients. One hundred eleven patients participated in the developed to measure fear of negative evaluation in social sit-
present study from July 2003 to June 2012. Some participants uations [21]. Each item is rated on a 5-point scale of 0 (not at
were referred from mental health professionals, and others all characteristic of me) through 4 (extremely characteristic
sought treatment for SAD on their own. All of the patients of me). This scale is a brief version of the original 30-item
met the criteria for SAD as the primary diagnosis according Fear of Negative Evaluation (FNE) scale [22]. The FNE
to the DSM-IV criteria [14], assessed by the Structured Clin- assesses fear of negative evaluation by others, a core cognitive
ical Interview for DSM-IV (SCID) [15]. All of them fulfilled feature of SAD. Fear of negative evaluation is recognized in
the following criteria: (i) absence of cluster B personality dis- cognitive models of SAD [23]. Furthermore, there is substan-
order, (ii) absence of substance use disorder, (iii) no history tial empirical evidence that fear of negative evaluation is a
of psychosis or bipolar disorder, and (iv) no previous CBT core cognitive feature of SAD [24]. The BFNE is more sensi-
treatment. The main reason for exclusion was that partici- tive to treatment-related change than FNE [25]. Its Japanese
pants with severe psychiatric disease could not attend the version has been validated [26].
group CBT sessions regularly. Although no other additional
structured psychotherapies were allowed during the CBT, 2.3.3. SCL-90-R. The SCL-90-R is a 90-item symptom inven-
there was no restriction on concurrent pharmacotherapy. tory for general psychopathology. The self-report question-
The participants provided their written informed consent naire is divided into ten subscales of somatization,
after receiving a full explanation of the study’s purpose and obsessive-compulsive, interpersonal sensitivity, depression,
procedures. anxiety, hostility, phobic anxiety, paranoid ideation,
Psychiatry Journal 3

psychoticism, and global severity index. Each item is scored Table 1: Demographics and baseline characteristics, mean (SD)
on a five-point scale ranging from 0 (not at all) to 4 (N = 111).
(extremely) [27]. The reliability and validity of the Japanese
Completers Dropouts
version have been shown [28]. The depression subscales have Characteristics p
(n = 96) (n = 15)
been often used as psychiatric outcome measures [29].
Therefore, we used the depression subscale including 13 Sex (% female) 52.1 46.7 0.79
items. Mean age 34.2 (10.8) 32.1 (11.9) 0.48
Onset 19.4 (8.1) 16.3 (5.1) 0.17
2.3.4. WHLS. The WHLS is a self-report questionnaire for Current mood disorder
assessing functional impairments in the domains of work, 17.7 20.0 0.83
(%)
home management, and social and private leisure activities. Current panic disorder
Each item is rated on a 9-point scale of 0 (not at all impaired) 2.1 0 0.57
(%)
through 8 (very severely impaired). Satisfactory reliability Current specific phobia
and construct validity have been demonstrated [30]. The Jap- 2.1 0 0.57
(%)
anese version has been validated [31]. SCL-90-R depression 1.36 (0.82) 1.53 (0.97) 0.46
BFNE 35.5 (10.1) 38.9 (9.0) 0.22
2.4. Statistical Analyses. All the data were examined using the
Statistical Package for the Social Sciences (SPSS) 18.0 for LSAS
Windows [32]. The baseline and endpoint data were used Total 75.5 (22.8) 85.3 (22.8) 0.17
to calculate change scores for each variable, denoted by delta Social interaction factor 38.8 (16.2) 44.4 (11.0) 0.10
(Δ). All the statistical tests were two-tailed, and results were Public speaking factor 21.3 (5.9) 21.9 (6.6) 0.69
considered statistically significant when the p value was less Observation by others
than 0.05. First, we compared the demographic and clinical 11.2 (6.0) 13.7 (6.4) 0.26
factor
data between the patients who completed the program and Eating and drinking in
those who did not, using an independent-samples t-test or public
4.2 (3.3) 5.3 (4.1) 0.26
χ2 test. Second, we used a paired t-test to compare all the WHLS
SAD symptomatology and social functioning scores between Work 5.2 (2.1) 4.5 (1.8) 0.26
the pretreatment and the posttreatment. Third, simple corre-
Home management 2.3 (2.0) 2.7 (2.0) 0.39
lations among change scores in SAD symptoms and cogni-
tions were investigated to identify significant bivariate Social leisure activities 4.8 (2.2) 5.3 (2.1) 0.40
correlation. Fourth, to examine the predictors of changes in Private leisure activities 1.6 (2.0) 1.5 (2.1) 0.93
social functioning outcomes, we performed multiple linear p values were calculated using t statistic for continuous variables and using χ
regression analysis. We used the variables about the changes statistic for categorical variables. SCL-90-R: Symptom Checklist-90 Revised;
in the four factors of LSAS (social interaction, public speak- BFNE: Brief Fear of Negative Evaluation Scale; LSAS: Liebowitz Social
Anxiety Scale; WHLS: Work, Home Management, Social and Private
ing, observation by others, and eating and drinking in pub- Leisure Activities Scale. Completers: the patients who completed the
lic), BFNE, and depression subscale of SCL-90-R as treatment, even when they were absent from a few sessions.
independent variables. The changes in the four subscales of
WHLS (work, home management, social leisure activities,
and private leisure activities) were used as dependent
variables. 3.3. Correlations among Change Scores. Bivariate correlations
of change scores are shown in Table 3. There were significant
2.5. Ethical Approval. The study was performed in accor- correlations among changes in the depression subscale of
dance with the Declaration of Helsinki, and the study’s pro- SCL-90-R, BFNE, and four factors of LSAS. All of the corre-
tocol was approved by the ethics committee of our institute. lations were less than 0.7 meaning that the correlation was
relatively low and might not cause multicollinearity.
3. Results
3.1. Patients. Among 111 patients who started the CBT pro- 3.4. Predictors of the Outcome Changes. In multiple regres-
gram, 96 patients (86.5%) completed the treatment, even sion analysis, the reduction in the public speaking factor of
when they were absent from a few sessions. Fifteen patients LSAS predicted the reduction in the work subscale of WHLS
dropped out prematurely from the treatment. Table 1. shows (Table 4). The reduction in the home subscale was predicted
the baseline demographic and clinical characteristics of the by the reduction in the depression subscale of SCL-90-R
completers and the dropouts. No statistically significant dif- (Table 5). The reduction in the social leisure activities sub-
ferences were seen among the subgroups. scale was predicted by the reduction in the depression sub-
scale of SCL-90-R and the reduction in the social
3.2. Pretreatment and Posttreatment Rating Scale Scores. interaction factor of LSAS (Table 6). The reduction in the
Table 2. presents the pretreatment and posttreatment clinical observation by others factor of LSAS predicted the private
rating scale scores. All the scores were improved significantly leisure activities subscale reduction (Table 7). The change
(p < 0:05). in BFNE did not predict the changes in WHLS subscales.
4 Psychiatry Journal

Table 2: Pretreatment and posttreatment rating scale scores, mean (SD) (n = 96).

Pre-treatment Post-treatment p
SCL-90-R depression 1.36 (0.82) 0.97 (0.77) <0.01
BFNE 35.5 (10.1) 27.7 (9.81) <0.01
LSAS
Total 75.5 (22.8) 55.8 (26.0) <0.01
Social interaction 38.8 (16.2) 29.6 (15.0) <0.01
Public speaking 21.3 (5.9) 15.6 (6.7) <0.01
Observation by others 11.2 (6.0) 7.6 (5.1) <0.01
Eating and drinking in public 4.2 (3.3) 3.0 (2.9) <0.01
WHLS
Work 5.2 (2.1) 3.6 (2.0) <0.01
Home management 2.3 (2.0) 1.4 (1.6) <0.01
Social leisure activities 4.8 (2.2) 3.2 (2.2) <0.01
Private leisure activities 1.6 (2.0) 1.2 (1.7) <0.05
p values were calculated using t statistic. SCL-90-R: Symptom Checklist-90 Revised; BFNE: Brief Fear of Negative Evaluation Scale; LSAS: Liebowitz Social
Anxiety Scale; WHLS: Work, Home Management, Social and Private Leisure Activities Scale.

Table 3: Bivariate correlations among changes of symptom (n = 96).

ΔSCL-90-R depression ΔBFNE ΔLSAS social ΔLSAS public ΔLSAS observation ΔLSAS eating
ΔSCL-90-R depression — 0.53∗∗ 0.43∗∗ 0.39∗∗ 0.42∗∗ 0.27∗∗
ΔBFNE — 0.48∗∗ 0.48∗∗ 0.45∗∗ 0.23∗
ΔLSAS social — 0.64∗∗ 0.69∗∗ 0.46∗∗
∗∗
ΔLSAS public — 0.47 0.49∗∗
ΔLSAS observation — 0.38∗∗
ΔLSAS eating —
∗ ∗∗
SCL-90-R: Symptom Checklist-90 Revised; BFNE: Brief Fear of Negative Evaluation Scale; LSAS: Liebowitz Social Anxiety Scale ( p < 0:05, p < 0:01).

Table 4: Predictors of WHLS work subscale changes (n = 96).

Coefficient Standardized coefficient t p

ΔSCL-90-R depression 0.32 0.13 1.15 0.25
ΔBFNE 0.01 0.07 0.57 0.57
ΔLSAS social interaction 0.02 0.14 0.93 0.35
ΔLSAS public speaking 0.12 0.35∗∗ 2.69 <0.01
ΔLSAS observation by others -0.02 -0.06 -0.42 0.67
ΔLSAS eating and drinking in public -0.03 -0.04 -0.40 0.69
Adjusted R‐square = 0:21, F value = 5:18, degrees of freedom = 95, ∗ p < 0:05, and ∗∗ p < 0:01. SCL-90-R: Symptom Checklist-90 Revised; BFNE: Brief Fear of
Negative Evaluation Scale; LSAS: Liebowitz Social Anxiety Scale; WHLS: Work, Home Management, Social and Private Leisure Activities Scale.

Table 5: Predictors of WHLS home management subscale changes (n = 96).

Coefficient Standardized coefficient t p

∗
ΔSCL-90-R depression 0.53 0.23 1.99 <0.05
ΔBFNE 0.02 0.11 0.90 0.37
ΔLSAS social interaction -0.01 -0.04 -0.23 0.82
ΔLSAS public speaking 0.01 0.03 0.19 0.85
ΔLSAS observation by others 0.03 0.08 0.61 0.54
ΔLSAS eating and drinking in public 0.12 0.17 1.45 0.15
Adjusted R‐square = 0:13, F value = 3:42, degrees of freedom = 95, ∗ p < 0:05, and ∗∗ p < 0:01. SCL-90-R: Symptom Checklist-90 Revised; BFNE: Brief Fear of
Negative Evaluation Scale; LSAS: Liebowitz Social Anxiety Scale; WHLS: Work, Home Management, Social and Private Leisure Activities Scale.
Psychiatry Journal 5

Table 6: Predictors of WHLS social leisure activities subscale changes (n = 96).

Coefficient Standardized coefficient t p

ΔSCL-90-R depression 0.63 0.26∗ 2.51 <0.05
ΔBFNE 0.00 0.10 0.10 0.92
ΔLSAS social interaction 0.04 0.30∗ 2.20 <0.05
ΔLSAS public speaking 0.03 0.11 0.92 0.36
ΔLSAS observation by others -0.03 -0.07 -0.61 0.54
ΔLSAS eating and drinking in public 0.12 0.17 1.66 0.10
Adjusted R‐square = 0:33, F value = 8:68, degrees of freedom = 95, ∗ p < 0:05, and ∗∗ p < 0:01. SCL-90-R: Symptom Checklist-90 Revised; BFNE: Brief Fear of
Negative Evaluation Scale; LSAS: Liebowitz Social Anxiety Scale; WHLS: Work, Home Management, Social and Private Leisure Activities Scale.

Table 7: Predictors of WHLS private leisure activities subscale changes (n = 96).

Coefficient Standardized coefficient t p

ΔSCL-90-R depression 0.38 0.17 1.46 0.15
ΔBFNE -0.02 -0.12 -0.96 0.34
ΔLSAS social interaction -0.01 -0.06 -0.39 0.70
ΔLSAS public speaking 0.01 0.04 0.32 0.75
ΔLSAS observation by others 0.11 0.35∗ 2.54 <0.05
ΔLSAS eating and drinking in public 0.06 0.09 0.80 0.43
∗ ∗∗
Adjusted R‐square = 0:12, F value = 3:18, degrees of freedom = 95, p < 0:05, and p < 0:01. SCL-90-R: Symptom Checklist-90 Revised; BFNE: Brief Fear of
Negative Evaluation Scale; LSAS: Liebowitz Social Anxiety Scale; WHLS: Work, Home Management, Social and Private Leisure Activities Scale.

The number of samples per an independent variable in a A novel finding of our study is that the changes in factors
multivariate analysis should be more than ten [33]. In our in LSAS may predict the social functioning changes in CBT
study, the number of samples was 96 and we use six indepen- for SAD. Cognitive behavioral interventions are tailored to
dent variables in multiple regression analysis. Therefore, the particular fears or feared situations in the patient’s life. The
number of samples per one independent variable was 96/6 situations may be chosen based on what is treatable or the
= 16. The sample size for this study was appropriate. patient’s goal for treatment. For example, in order to improve
especially working in social functioning, we could select the
public speaking task in an exposure session. Our results
4. Discussion may be useful for selecting the situation in tailored treatment.
However, the adjusted R-squares in our results are relatively
The present study demonstrated that the changes in social low. Further investigation is needed to clarify the factors
anxiety or depression symptoms may predict several aspects influencing social functioning.
of social functioning change in patients with SAD over the The present study has several limitations. First, this study
course of CBT, whereas the cognitive change was not related was carried out as a naturalistic study and did not include a
to the change in social functioning. control group. Therefore, one cannot conclude if the signifi-
The present results are consistent with the finding of Oei cant changes in symptomatology and social functioning in
et al. (2014) in CBT for other anxiety disorders that QOL the present cohort might be due to the passage of time rather
change is related to anxiety and depression symptom than CBT for SAD. Second, antidepressant and benzodiaze-
changes. In patients with SAD, social functioning may be pine medications were permitted during CBT treatment.
more closely linked to symptoms while changes in social The information about the amount of drug use during the
functioning might occur independently of cognitive changes. treatment was not collected. Hence, it is impossible to consider
Furthermore, Oei et al. (2014) showed that cognitive changes the effects of medications on CBT. Third, the clinicians in
are not consistently associated with changes in QOL during charge of the CBT program for SAD rated LSAS at baseline
CBT for other anxiety disorders. and endpoint. Accordingly, rating LSAS was not blind or inde-
The present study suggested that the mechanism of pendent. Fourth, the causality cannot be established between
change in QOL or social functioning during CBT may be rel- symptom changes and social functioning changes. Changes
evant to symptom changes rather than cognitive changes. in social functioning may occur as a result of or simulta-
From a clinical point of view, our results suggest that it neously with symptom change. Fifth, this study lacked the data
may be useful to focus on achieving symptom improvements from every session and therefore we cannot analyze the mech-
in preference to cognitive change in order to improve social anism of change over time during CBT. Sixth, the data was
functioning during CBT for SAD. collected from 2003 to 2012 and considered to be rather old.
6 Psychiatry Journal

However, the demographic and clinical characteristics of this Acta Psychiatrica Scandinavica, vol. 109, no. s420, pp. 38–46,
cohort, including sex, mean age, onset, comorbidity, and LSAS 2004.
total score, did not differ much from those of the general SAD [6] F. R. Schneier, “Clinical practice. Social anxiety disorder,” New
population as shown in Table 1. England Journal of Medicine, vol. 355, no. 10, pp. 1029–1036,
2006.
5. Conclusion and Recommendations [7] C. Acarturk, P. Cuijpers, A. Van Straten, and R. de Graaf, “Psy-
chological treatment of social anxiety disorder: a meta-analy-
The present study suggested that the changes in social anxi- sis,” Psychological Medicine, vol. 39, no. 2, pp. 241–254, 2009.
ety or depression symptoms may predict several aspects of [8] J. Cottraux, I. Note, E. Albuisson et al., “Cognitive behavior
social functioning changes in patients with SAD over the therapy versus supportive therapy in social phobia: a random-
course of CBT. For the purpose of improving the social func- ized controlled trial,” Psychotherapy and Psychosomatics,
tioning, our results may be useful for selecting the fear or vol. 69, no. 3, pp. 137–146, 2000.
feared situation over the course of CBT for SAD. [9] E. Mörtberg, D. M. Clark, Ö. Sundin, and A. Å. Wistedt,
“Intensive group cognitive treatment and individual cognitive
Data Availability therapy vs. treatment as usual in social phobia: a randomized
controlled trial,” Acta Psychiatrica Scandinavica, vol. 115,
We also think that data sharing is important. However, the no. 2, pp. 142–154, 2007.
data in this study are not publicly available. Because this [10] P. Carlbring, M. Gunnarsdottir, L. Hedensjo, G. Andersson,
study was started in 2003, and in accordance with the proto- L. Ekselius, and T. Furmark, “Treatment of social phobia: ran-
col at that time, the study participants were informed that the domised trial of internet-delivered cognitive-behavioural ther-
data used in this study will not be provided to third parties apy with telephone support,” British Journal of Psychiatry,
other than the researchers and the participants. vol. 190, no. 2, pp. 123–128, 2007.
[11] S. G. Hofmann, J. Q. Wu, and H. Boettcher, “Effect of
cognitive-behavioral therapy for anxiety disorders on quality
Conflicts of Interest of life: a meta-analysis,” Journal of Consultinf and Clinical Psy-
The authors have declared that no conflicts of interest exist. chology, vol. 82, no. 3, pp. 375–391, 2014.
[12] T. P. S. Oei, N. M. McAlinden, and T. Cruwys, “Exploring
Authors’ Contributions mechanisms of change: the relationships between cognitions,
symptoms, and quality of life over the course of group
This work was conducted in collaboration with all authors. cognitive-behaviour therapy,” Journal of Affective Disorders,
SO was the primary investigator. TA supervised the overall vol. 168, pp. 72–77, 2014.
research project. All authors performed the clinical investiga- [13] C. Suveg, J. N. Kingery, M. Davis, A. Jones, M. Whitehead, and
tion (diagnosis, treatment, and assessment). All authors M. L. Jacob, “Still lonely: social adjustment of youth with and
without social anxiety disorder following cognitive behavioral
approved the final manuscript.
therapy,” Journal of Anxiety Disorders, vol. 52, pp. 72–78,
2017.
Acknowledgments [14] American Psychiatric Association and American Psychiatric
This research was supported by Grants-in-Aid from the Min- Association, Task Force on DSM-IV. Diagnostic and Statistical
Manual of Mental Disorders: DSM-IV, American Psychiatric
istry of Health, Labour and Welfare, Japan, to T.A.F.
Association, Washington, DC, 4th edition, 1994.
[15] M. B. First, Structured Clinical Interview for DSM-IV Axis I
References Disorders : SCID - I : Clinician Version: Administration Book-
[1] M. B. Stein and Y. M. Kean, “Disability and quality of life in let, American Psychiatric Press, Washington, D.C, 1997.
social phobia: epidemiologic findings,” American Journal of [16] G. Andrews, M. Creamer, R. Crino, C. Hunt, L. Lampe, and
Psychiatry, vol. 157, no. 10, pp. 1606–1613, 2000. A. Page, The Treatment of Anxiety Disorders : Clinician Guides
[2] N. M. Simon, M. W. Otto, N. B. Korbly, P. M. Peters, D. C. and Patient Manuals, Cambridge University Press, Cam-
Nicolaou, and M. H. Pollack, “Quality of life in social anxiety bridge, UKNew York, NY, 2nd edition, 2003.
disorder compared with panic disorder and the general popu- [17] D. M. Clark and A. Wells, “A cognitive model of social pho-
lation,” Psychiatric Services, vol. 53, no. 6, pp. 714–718, 2002. bia,” Social Phobia: Diagnosis, Assessment, and Treatment,
[3] F. R. Schneier, J. Johnson, C. D. Hornig, M. R. Liebowitz, and vol. 41, pp. 00022-00023, 1995.
M. M. Weissman, “Social phobia. Comorbidity and morbidity [18] D. M. FRESCO, M. E. COLES, R. G. HEIMBERG et al., “The
in an epidemiologic sample,” Archives of General Psychiatry, Liebowitz Social Anxiety Scale: a comparison of the psycho-
vol. 49, no. 4, pp. 282–288, 1992. metric properties of self-report and clinician-administered for-
[4] H. U. Wittchen, M. Fuetsch, H. Sonntag, N. Muller, and mats,” Psychological Medicine, vol. 31, no. 6, pp. 1025–1035,
M. Liebowitz, “Disability and quality of life in pure and comor- 2001.
bid social phobia. Findings from a controlled study,” European [19] S. A. Safren, R. G. Heimberg, K. J. Horner, H. R. Juster, F. R.
Psychiatry, vol. 15, no. 1, pp. 46–58, 2000. Schneier, and M. R. Liebowitz, “Factor structure of social fears:
[5] The ESEMeD/MHEDEA 2000 investigators, J. Alonso, M. C. the Liebowitz Social Anxiety Scale,” Journal of Anxiety Disor-
Angermeyer et al., “Disability and quality of life impact of ders, vol. 13, no. 3, pp. 253–270, 1999.
mental disorders in Europe: results from the European Study [20] S. Asakura, S. Inoue, F. Sasaki, Y. Sasaki, N. Kitagawa, and
of the Epidemiology of Mental Disorders (ESEMeD) project,” T. Inoue, “Reliability and validity of the Japanese version of
Psychiatry Journal 7

the Liebowitz Social Anxiety Scale,” Seishin Igaku (Clinical

Psychiatry), vol. 44, pp. 1077–1084, 2002.
[21] M. R. Leary, “A brief version of the Fear of Negative Evaluation
Scale,” Personality and Social Psychology Bulletin, vol. 9, no. 3,
pp. 371–375, 2016.
[22] D. Watson and R. Friend, “Measurement of social-evaluative
anxiety,” Journal of consulting and clinical psychology, vol. 33,
no. 4, pp. 448–457, 1969.
[23] D. M. Clark and A. Wells, A Cognitive Model of Social Phobia,
The Guilford Press, New York, 1995.
[24] D. A. Clark and A. T. Beck, Cognitive Therapy of Anxiety Dis-
order Science and Practice, The Guilford Press, 72 Spring
Street, New York, NY 10012, 2010.
[25] K. A. Collins, H. A. Westra, D. J. A. Dozois, and S. H. Stewart,
“The validity of the brief version of the Fear of Negative Eval-
uation Scale,” Journal of anxiety disorders, vol. 19, no. 3,
pp. 345–359, 2005.
[26] R. Ishikawa, K. Sasaki, and I. Fukui, “Standardization of Japa-
nese version of FNE and SADS,” Koudou Ryouhou Kenkyu
(Jap J Behav Ther), vol. 18, pp. 10–17, 1992.
[27] L. R. Derogatis, SCL-90-R : Administration, Scoring & Proce-
dures Manual -II, for the R (evised) Version and Other Instru-
ments of the Psychopathology Rating Scale Series, Clinical
Psychometric Research, Towson, Md, 2nd edition, 1992.
[28] T. A. Furukawa, M. Nakanishi, A. Sakurai, A. Suzuki,
A. Suzuki-Moor, and T. Hamanaka, “Effects of rthyl loflaze-
pate in mood and neurosis-related disorders (ICD-10 JCM) ;
changes in SCL-90-R subscale scores,” Rinsho Seisinigaku
(Clinical Psychiatry), vol. 25, pp. 233–240, 1996.
[29] N. Watanabe, T. A. Furukawa, J. Chen et al., “Change in qual-
ity of life and their predictors in the long-term follow-up after
group cognitive behavioral therapy for social anxiety disorder:
a prospective cohort study,” BMC Psychiatry, vol. 10, no. 1,
2010.
[30] J. C. Mundt, I. M. Marks, M. K. Shear, and J. M. Greist, “The
Work and Social Adjustment Scale: a simple measure of
impairment in functioning,” British Journal of Psychiatry,
vol. 180, no. 5, pp. 461–464, 2002.
[31] Y. Nakano, K. Lee, Y. Noda et al., “Cognitive-behavior therapy
for Japanese patients with panic disorder: acute phase and one-
year follow-up results,” Psychiatry and Clinical Neurosciences,
vol. 62, no. 3, pp. 313–321, 2008.
[32] SPSS for Windows (Version 18.0), SPSS Inc., Chicago, US,
2009.
[33] P. Peduzzi, J. Concato, A. R. Feinstein, and T. R. Holford,
“Importance of events per independent variable in propor-
tional hazards regression analysis. II. Accuracy and precision
of regression estimates,” Journal of Clinical Epidemiology,
vol. 48, no. 12, pp. 1503–1510, 1995.