Behaviour Research and Therapy 76 (2016) 47e56

Contents lists available at ScienceDirect

Behaviour Research and Therapy

journal homepage: www.elsevier.com/locate/brat

Internet-delivered cognitive behavioural therapy for children with

anxiety disorders: A randomised controlled trial
Sarah Vigerland a, *, Brja

nn Ljo

tsson b, Ulrika Thulin a, Lars-Go b, c,
ran Ost
Gerhard Andersson b, d, Eva Serlachius a
a
Department of Clinical Neuroscience, Centre for Psychiatric Research and Education, Karolinska Institutet, Gavlegatan 22, 113 30 Stockholm, Sweden
b g 9, 171 77 Stockholm, Sweden
Department of Clinical Neuroscience, Division of Psychology, Karolinska Institutet, Nobels va
c
Department of Psychology, Stockholm University, 106 91 Stockholm, Sweden
d
Department of Behavioral Sciences and Learning, Linko ping University, 581 83 Linkoping, Sweden

a r t i c l e i n f o a b s t r a c t

Article history: Background: Cognitive behaviour therapy (CBT) has been shown to be an effective treatment for anxiety
Received 11 March 2015 disorders in children, but few affected seek or receive treatment. Internet-delivered CBT (ICBT) could be a
Received in revised form way to increase the availability of empirically supported treatments.
10 November 2015
Aims: A randomised controlled trial was conducted to evaluate ICBT for children with anxiety disorders.
Accepted 11 November 2015
Method: Families (N 93) with a child aged 8e12 years with a principal diagnosis of generalised anxiety
Available online 19 November 2015
disorder, panic disorder, separation anxiety, social phobia or specic phobia were recruited through
media advertisement. Participants were randomised to 10 weeks of ICBT with therapist support, or to a
Keywords:
Internet-delivered treatment
waitlist control condition. The primary outcome measure was the Clinician Severity Rating (CSR) and
CBT secondary measures included child- and parent-reported anxiety. Assessments were made at pre-
Children treatment, post-treatment and at three-month follow-up.
Anxiety disorders Results: At post-treatment, there were signicant reductions on CSR in the treatment group, with a large
between-group effect size (Cohen's d 1.66). Twenty per cent of children in the treatment group no
longer met criteria for their principal diagnosis at post-treatment and at follow-up this number had
increased to 50%. Parent-reported child anxiety was signicantly lower in the treatment group than in
the waitlist group at post-treatment, with a small between-group effect size (Cohen's d 0.45). There
were no signicant differences between the groups regarding child-ratings of anxiety at post-treatment.
Improvements were maintained at three-month follow-up, although this should be interpreted
cautiously due to missing data.
Conclusions: Within the limitations of this study, results suggest that ICBT with therapist support for
children with anxiety disorders can reduce clinician- and parent-rated anxiety symptoms.
Trial registration: Clinicaltrials.gov: NCT01533402.
2015 Published by Elsevier Ltd.

1. Introduction impairments in everyday life of the affected individual as well as in

Epidemiological studies show that 5e10% of children and ado-
lescents suffer from an anxiety disorder, which means that it is one
of the most common mental disorders for this age group (Pine &
Klein, 2008). Anxiety disorders are associated with suffering and
* Corresponding author. Child and Adolescent Psychiatry Research Centre,
Gavlegatan 22, 113 30, Stockholm, Sweden.
E-mail addresses: sarah.vigerland@ki.se (S. Vigerland), brjann.ljotsson@ki.se

tsson), ulrika.thulin@ki.se (U. Thulin), ost@psychology.su.se (L.-G. Ost),
(B. Ljo
gerhard.andersson@liu.se (G. Andersson), eva.serlachius@ki.se (E. Serlachius).
the family and among relatives, and increase the risk for depres-
sion, substance abuse and impairment in social and emotional
functioning during adolescence and early adulthood (Costello,
Angold, & Keeler, 1999; Fichter, Quadieg, Fischer, & Kohlboeck,
2009; Ginsburg, La Greca, & Silverman, 1998; Pine & Klein, 2008).
Thus, childhood anxiety disorders should be identied and treated
as early as possible.
There is strong support for cognitive behaviour therapy (CBT) as
an effective treatment for anxiety disorders in children (James,
James, Cowdrey, Soler, & Choke, 2013), and CBT is regarded to be
the treatment of choice for this group (Dadds & Barrett, 2001).

http://dx.doi.org/10.1016/j.brat.2015.11.006
0005-7967/ 2015 Published by Elsevier Ltd.
48 S. Vigerland et al. / Behaviour Research and Therapy 76 (2016) 47e56
However, the vast majority of children and adolescents with anxi-
ety disorders do not receive evidence-based psychological treat-
ment (Chavira, Stein, Bailey, & Stein, 2005; Costello, He, Sampson,
Kessler, & Merikangas, 2014). It is therefore important to nd
means to increase the availability of CBT for this population.
Computers and the Internet enable us to offer less therapist-
intensive but effective interventions over long distances, which
can increase the availability of evidence based treatments. Internet-
delivered CBT (ICBT) has proven to be an effective method in the
treatment of adults with depression and anxiety disorders
(Andersson, 2014; Hedman, Ljo
tsson, & Lindefors, 2012). Studies
have not only shown efcacy of ICBT compared to no-treatment
control conditions, but have also demonstrated results compara-
ble to face-to-face treatment (Andersson, Cuijpers, Carlbring, Riper,
& Hedman, 2014).
To date, there are several trials evaluating computerised or
internet delivered CBT-interventions for children and adolescents
with anxious or depressive symptoms (Pennant et al., 2015; Reyes-
Portillo et al., 2014; Richardson, Stallard, & Velleman, 2010). ICBT
has some potential advantages compared to computerised CBT, the
most important being the possibility of non-synchronous therapist
support independent of geographical distances, as computerised
approaches often involve face-to face therapist support (Khanna &
Kendall, 2010; Stallard, Richardson, Velleman, & Attwood, 2011).
Wuthrich et al. (2012) conducted a RCT comparing a CD-ROM
delivered treatment with telephone-delivered therapist support
to a waitlist control for adolescents with anxiety disorders, making
it more comparable to ICBT.
A majority of the evaluated ICBT programs are prevention pro-
grams that target children and adolescents who are at risk or have
elevated symptom levels (Reyes-Portillo et al., 2014). The BRAVE-
ONLINE programs, however, have been evaluated for children and
adolescents with diagnosed anxiety disorders and have shown
promising results (Donovan & March, 2014; March, Spence, &
Donovan, 2009; Spence et al., 2011). For example, March et al.
(2009) compared an internet-delivered program for children
(8e12 years) with anxiety disorders to a wait-list control and found
a signicantly larger decrease of clinician rated anxiety severity in
the ICBT group. Results on child- and parent-rated anxiety symp-
toms were mixed at post-treatment. Reyes-Portillo et al. (2014)
considered BRAVE ONLINE to be probably efcacious but
pointed out that the results need to be replicated by other research
groups to further increase the evidence-base level.
We have previously evaluated ICBT for children with specic
phobia in an open trial with promising results (Vigerland et al.,
2013). Although there are a few studies with promising results,
more studies are needed to establish the efcacy of ICBT for chil-
dren with anxiety disorders. The aim of the present study was to
evaluate the efcacy of ICBT for children with anxiety disorders
compared to a waitlist control.
2. Method
2.1. Participants
Participants were self-referred and recruited nationally in
Sweden during 2012 through media advertisement. One parent in
each family was required to be responsible for participation and to
be the primary contact with the research team when needed. In-
clusion criteria for the study were; a) children had to be 8e12 years
of age, b) have a principal diagnosis of either generalised anxiety
disorder, panic disorder, separation anxiety, social phobia or spe-
cic phobia (except for blood-injury, or injection phobia) according
to DSM-IV criteria, c) any psychotropic medications had to be stable
for 3 months prior the treatment, and d) the family had to have
access to a computer and internet and speak Swedish.
Participants were excluded if the child a) had a diagnosed
neurodevelopmental disorder (e.g. autism or attention-decit/
hyperactivity disorder (ADHD)) b), was severely depressed
(dened as 20 on the Child Depression Inventory), or had another
acute psychiatric disorder (e.g. psychosis, suicidal ideation), or c)
was currently involved in psychological treatment. Participants
were also excluded if d) a parent had an on-going substance abuse
or if e) there was an on-going child custody dispute or abuse in the
family, or other family problems that made ICBT unsuitable. Finally,
families were excluded if f) the parent responsible for the treatment
and study participation suffered from serious psychiatric disorder.
The regional ethics committee in Stockholm approved the study.
Written informed consent was obtained from the parents and the
children gave their verbal consent to participate.
2.2. Assessments
2.2.1. Primary outcome measure
The primary outcome measure was the Clinician Severity Rating
(CSR) derived from the Anxiety Disorder Interview Schedule Child
and Parent version (ADIS C/P; Albano & Silverman, 1996). ADIS is a
semi-structured interview conducted with the child and parent
separately to assess diagnostic criteria according to DSM-IV
(American Psychiatric Association, 2000). The severity of each
diagnosis is assessed with the CSR on a 9-point scale (0e8). A score
of 3 or lower is considered as subclinical symptoms whereas a score
of 4 or higher means that the criteria for diagnosis, with regard to
severity, are fullled. The interviewing clinicians initially rate CSR
based on child and parent interviews separately and then agree on
a set of nal composite CSR-scores. The ADIS C/P has shown good to
excellent kappa coefcients and excellent test-retest reliability
(Silverman, Saavedra, & Pina, 2001), and the concurrent validity of
the ADIS C/P has been established (Wood, Piacentini, Bergman,
McCracken, & Barrios, 2002). The telephone administration of
ADIS, conducted with one parent, has yielded good to excellent
agreement with the face-to-face administrated interview
(Lyneham & Rapee, 2005).
2.2.2. Secondary outcome measures
2.2.2.1. Children's Global Assessment Scale (CGAS). CGAS (Shaffer
et al., 1983) is used by clinicians to assess global functioning in
children (scale 0e100) and adolescents, with higher scores indi-
cating higher functioning. It has shown moderate to excellent inter-
rater reliability, good stability over time and good concurrent as
well as discriminant validity (Bird, Canino, & Rubio-Stipec, 1987;
Lundh, Kowalski, Sundberg, Gumpert, & Lande
n, 2010; Shaffer
et al., 1983).
2.2.2.2. The Spence Children's Anxiety Scale (SCAS-C/P). SCAS-C
(Spence, 1998) measures six domains of anxiety; fear of physical
injury, generalised anxiety, obsessive-compulsive disorder, panic/
agoraphobia, separation anxiety and social anxiety. The question-
naire consists of 44 items that are rated on a 4-point scale. Higher
scores indicate high levels of anxiety. Six positive ller items are
included in the children's scale. The internal consistency is high
(Cronbach's alpha 0.93; Spence, Barrett, & Turner, 2003). The
parent version (SCAS-P; Nauta et al., 2004) consists of 38 items,
formulated to correspond to the child version without ller items,
and has shown high internal consistency. Cronbach's alpha in the
current study was 0.87 for SCAS-C and 0.81 for the parent scale.
2.2.2.3. Fear Survey schedule for children eRevised (FSSC-R C/P).
FSSC-R (Ollendick, 1983) consists of 80 items regarding fear of
specic things or situations. Each item is rated on a 3-point scale
 S. Vigerland et al. / Behaviour Research and Therapy 76 (2016) 47e56
with higher scores indicating higher levels of fear. The scale has
high internal consistency and moderate test-retest reliability over a
3-month period. When administered to parents, the wording of the
questions was adjusted. Cronbach's alpha in the current study was
0.93 for FSSCR-C and 0.90 for the parent scale.
2.2.2.4. Penn State worry questionnaire for children (PSWQ-C).
The PSWQ-C/P (Chorpita, Tracey, Brown, Collica, & Barlow, 1997) is
an adaptation of the PSWQ for adults, consisting of 14 items
regarding worry that are rated on a 4-point scale, which has shown
good validity and reliability. When administered to parents, the
wording of the questions was adjusted. Cronbach's alpha in the
current study was 0.99 for PSWQ-C and 0.98 for the parent scale.
2.2.2.5. Separation anxiety inventory for children (SAI-C). The SAI-C
(In-Albon, Meyer, & Schneider, 2013) consists of 12 items that
investigate to what extent the child avoids different relevant situ-
ations. It has shown good validity and reliability. When adminis-
tered to parents, the wording of the questions was adjusted.
Cronbach's alpha in the current study was 0.99 for SAI-C and 0.93
for the parent scale.
2.2.2.6. Social phobia and anxiety inventory (SPAI-C/P). SPAI-C
(Beidel, Turner, & Morris, 1995) is a scale with 26 items that mea-
sures cognitive, somatic and behavioural aspects of social anxiety,
and it has shown good validity and reliability. The parent version
(Higa, Fernandez, Nakamura, Chorpita, & Daleiden, 2006) is iden-
tical to the child version, formulated to correspond to the child
version of, and has shown good internal consistency. Cronbach's
alpha in the current study was 0.97 for SPAI-C and 0.96 for the
parent scale.
2.2.2.7. Quality of Life Inventory e Child version (QOLI e C).
The original QOLI for adults (Frisch, Cornell, Villanueva, & Retzlaff,
1992) covers importance and satisfaction in 16 different life do-
mains, e.g. family, friends, occupation and society and has shown
good reliability and validity when administered in Swedish sam-
ples (Lindner, Andersson, Ost, & Carlbring, 2013; Paunovi

c & Ost,
2004). The Swedish child version contains fewer life domains
(n 10) and child adjusted life domains. The 10 domains are rated
on how important they are (0e2) and how satised the participant
is with them (3 to 3), generating a possible total score of 6 to 6.
Cronbach's alpha in the current study was 0.89.
2.2.2.8. Treatment satisfaction. Participant satisfaction was
measured in children and parents at post-treatment using the
Client Satisfaction Scale (CSS; Ollendick, 2010). It consists of 10
questions, answered on a 5 point rating scale, regarding satisfaction
of treatment outcome; e.g. change in fear, avoidance and whether
or not the participant would recommend the treatment to others.
Higher scores indicate higher satisfaction. Cronbach's alpha in the
current study was 0.75 for the child scale and 0.85 for the parent
scale.
2.2.3. Screening measures
2.2.3.1. Development and well-being assessment (DAWBA).
DAWBA (Goodman, Ford, Richards, Gatward, & Meltzer, 2000) is a
screening instrument containing an interview and questionnaires
for children and adolescents in the ages 5e17 for common mental
health disorders such as anxiety disorders, depression, ADHD,
autism and bipolar disorder. It has shown excellent discriminant
validity and acceptable concurrent validity (Goodman et al., 2000).
In the present study the internet-based version of DAWBA directed
to parents was used, containing both open ended and closed
questions.
49
2.2.3.2. Child Depression Inventory (CDI). The CDI (Kovacs, 1985)
assesses the severity of depressive symptoms in children and ad-
olescents. It consists of 27 items graded from 0 (no symptoms) to 2
(severe symptoms). The questions cover depressed mood, self-
blame, loss of appetite, insomnia, interpersonal relationships and
school adjustment. Internal consistency in a psychiatric sample was
good (Kovacs, 1985) and validity was found to be equivocal
although CDI scores were related to diagnoses in a psychiatric
sample. The CDI has shown good reliability in a Swedish sample
(Ivarsson, Svalander, & Litlere, 2006) and Cronbach's alpha in the
current study was 0.79.
2.2.3.3. Primary care evaluation of mental disorders (Prime-MD).
Parents symptoms of anxiety, depression and other psychiatric
problems were assessed with Prime-MD (Spitzer et al., 1994) at pre-
treatment. The self-report form of the Prime-MD used in this study
screens for depression, anxiety, alcohol problems, somatoform
disorders and eating disorders. It consists of 29 items answered
with yes/no. Two items were removed from the self-report form
(item 4 Pains or problems during menstruation and item 5 Pain
or problems during sexual intercourse) as their content was
judged as irrelevant for the purpose of the questionnaire in the
current study. The self-report form has shown good overall sensi-
tivity when compared to a structured assessment by a mental
health professional (Spitzer et al., 1994).
2.3. Procedure
After responding to advertisement in newspapers, one parent
rst underwent a brief telephone interview to establish if basic
inclusion criteria were fullled (criteria a, c-e) and no exclusion
criteria were met (criteria a, c, e). Thereafter at least one parent
completed an internet-based screening instrument, DAWBA (to
screen for psychiatric comorbidity). The child and both parents
(except the cases where there only was one parent) then completed
CDI, SCAS, Prime-MD, and QOLI online. All questionnaires and self-
report measures were in Swedish. For those not meeting exclusion
criteria (b, d, f), the child and at least one parent underwent a face-
to-face assessment using the Swedish translation of the Anxiety
Disorders Interview Schedule (child and parent version; ADIS C/P)
with a research assistant or clinical psychologist. Global functioning
using CGAS was rated after the interview was completed. The
research assistants were last-year students in the Swedish ve-year
clinical psychology program, with completed one-year training in
CBT. The other assessors were experienced psychologists with CBT-
training. If a principal diagnosis of generalised anxiety disorder,
panic disorder, separation anxiety, social phobia or specic phobia
was conrmed (criterion b) and the family had given informed
consent, participants were included in the study and asked to
complete an additional self-report measure corresponding to their
principal diagnoses (FSSC-R, PSWQ, SAI or SPAI; no additional
measure was administered to those with a principal diagnosis of
panic disorder). Included participants were instructed to keep any
psychotropic medication constant throughout the treatment
period. Participants were then randomised to the intervention or a
waitlist control.
At post-treatment the relevant parts of the ADIS interview (di-
agnoses that had been scored 4 on CSR at pre-treatment) were
administered over the telephone. We did not consider it feasible to
interview the younger children over the telephone, and therefore
all post-and follow-up interviews were conducted with only a
parent. The interviewing research assistant or clinical psychologist,
who was not blind to treatment condition or baseline diagnoses,
then completed a CSR and CGAS for each participant. Parents and
children also completed CSS C/P, SCAS C/P, QOLI-C and the self-
50 S. Vigerland et al. / Behaviour Research and Therapy 76 (2016) 47e56
report measure corresponding to their principal diagnosis online.
This procedure was repeated with the treatment group at three-
month follow-up (CSS was not included at this time point). Par-
ticipants in the waitlist group received the treatment after the post-
measurement was completed.
2.4. Treatment
The intervention was developed by the research group and has
been evaluated for specic phobia with promising results
(Vigerland et al., 2013). The program focused mainly on exposure
but also included psychoeducation, coping strategies and problem
solving skills (see Table 1 for an overview of the treatment content).
For this study, the psychoeducation and examples in the program
was adapted into ve slightly different versions to t each principal
anxiety disorder (generalised anxiety disorder, panic disorder,
separation anxiety, social phobia and specic phobia). Furthermore,
rationale and instruction for worry-time were included in the GAD-
version included and brief psychoeducation on social skills training
was added to the social phobia version.
Participants had access to the treatment platform for ten weeks
and treatment content was presented in 11 chapters or modules
(see Table 1). The content was presented in a varied manner with
reading material, lms, animations, illustrations, and exercises (see
Figs. 1 and 2 for an example). The treatment can be described as a
combined parent-child intervention with seven of the modules
aimed at the parent(s), containing information and instructions on
how to help their child, and four modules addressed to the child.
Parents were to work with their modules rst so that they would be
prepared to assist their child on the child directed modules. Child
directed content was a shorter version of the parent directed in-
formation, adapted to an appropriate level and including less text
and more animations. Throughout the treatment participants had
online contact with an assigned psychologist/CBT-therapist
through written messages and written feedback on worksheets.
Families received individually tailored replies from their therapist
within 48 h after submitting exercises and/or questions. Three
telephone calls were scheduled during treatment (at the beginning,
middle and end of treatment), and additional telephone calls were
conducted if it was deemed necessary in order to increase moti-
vation or problem solve during the exposure focused weeks. The
role of the psychologist was to answer questions and clarify treat-
ment content, increase motivation and to help solve problems if
necessary.
Parents and children worked through the modules at their own
pace. They were encouraged to cover psychoeducation and ratio-
nale for exposure for both parents and children during the rst two
weeks, and then to engage in exposure for the main part of the
Table 1
Overview of treatment content.
Week Module Content
1e2 1 Psychoeducation on emotions, fear and anxiety
2 Psychoeducation on anxiety disorders and CBT
3 Psychoeducation on goals and exposure hierarchies
4 An introduction to exposure, coping techniques (e.g. breathing and relaxation) and worry time/social skills training (only for GAD/Social phobia
programs)
5 An introduction to using a reward system
6 Preparation for managing obstacles
2e3 7 Psychoeducation on fear and anxiety
8 Psychoeducation on exposure, setting goals and creating exposure hierarchies
9 Planning exposures and coping techniques (e.g. breathing and relaxation)
4e9 None Families are instructed to work on their own with exposure exercises and to report their progress in the platform
10 10 Problem solving, maintenance plan
11 Summary, follow-up on goals, maintenance plan
treatment. No new modules were presented during the weeks
focussing on exposure but families were instructed to log in to the
platform and report their progress. Participants were then
instructed to complete the last modules covering maintenance of
treatment progress, continued improvement, and relapse preven-
tion during the nal treatment week.
2.4.1. Statistical analyses
The power calculation was originally conducted based on an
estimated between group effect size of 0.5 (based on March et al.
(2009), d 0.56). Allowing for approximately 15% dropout, it was
estimated that 120 participants were needed (Kazdin, 2010).
However, only 93 participants were included in the trial, which led
to 80% power to detect an effect size of 0.60, with an alpha-level of
0.05 (Kazdin, 2010).
All statistics were calculated using SPSS version 22. The differ-
ence in change between the treatment group and waitlist group
from pre-to post-treatment (i.e., the group * time interaction effect)
were analysed with hierarchical linear mixed-models (HLMM), that
used all available data without excluding cases with missing data
(Gueorguieva & Krystal, 2004). The inclusion of random intercept,
random slopes and covariance structures for repeated measures
were determined analytically using log likelihood ratio test.
Maintenance of improvement from post-treatment to follow-up for
the intervention group was tested with paired t-tests based on
observed data (i.e., no imputations of missing values were made).
Between-group effect sizes (Cohen's d) at post-treatment were
calculated based on the estimates obtained in the HLMM by
dividing the difference in slope (i.e. coefcient of time * group
interaction effect) by the observed pre-treatment standard devia-
tion for the whole group (Feingold, 2009). Within-group effect sizes
were based on observed values and calculated by dividing the
difference between the rst and second assessment with the
pooled within-group standard deviation (calculated using Equa-
tions (4.15) and (4.27) in Borenstein, Hedges, Higgins, & Rothstein,
2009).
Self-report measures were collected from both parents when
possible. When data from two parents were available the mean of
the parent ratings was used. Only the ICBT condition was assessed
at three-month follow-up, as participants in the waitlist condition
had been offered ICBT after completing the post-treatment
assessment.
Kappa coefcients (Cohen, 1960) and intraclass correlation co-
efcients (ICC; two-way random, average measure) was used to
calculate inter-rater reliability for composite ratings of anxiety
disorders at pre-treatment assessment, and for investigation of
agreement between child, parent and composite ratings on anxiety
disorders. The k values are evaluated as having poor reliability
Directed
at
Parent
Child
Parent
Child
 S. Vigerland et al. / Behaviour Research and Therapy 76 (2016) 47e56 51

Fig. 1. Screenshot of an animated psychoeducation lm from the program.

Fig. 2. Screenshot of an animated drag-and-drop exercise from the program.

when k < 0.40; fair when k is 0.40e0.59; good when k is 0.60e0.74;
and excellent when k is > 0.74 (Mannuzza et al. 1989).
3. Results
3.1. Descriptives
Participants were 93 children, aged 8e12 years, and 182 parents.
Ninety per cent of participants lived in Stockholm County. De-
mographic variables are presented in Table 2. In general, parents
were well educated with the majority having completed a univer-
sity degree. About a third of the children had a principal diagnosis
of separation anxiety or specic phobia, respectively and approxi-
mately 20% had a principal GAD diagnosis. Seventy per cent of the
children had more than one diagnosis, the average number of di-
agnoses being 2.3 (SD 1.3, Mdn 2). At pre-treatment, child
ratings of depression and parent ratings of their own anxiety and
depression symptoms were low (CDI; M 6.6, SD 4.9 and PRIME-
MD; M 5.5, SD 2.5, respectively).
Inter-rater reliability at pre-treatment was good (k 0.65,
p < 0.01) for anxiety disorders (i.e., presence yes or no) and
excellent for CSR scores of anxiety disorders (ICC 0.77). Overall,
CSR-scores on anxiety disorders from the child and parent inter-
view showed moderate correlations (r(91) 0.43e0.49, p < 0.01),
while there was a stronger correlation between anxiety disorder
CSR-scores from the parent interview and the composite scores
(r(91) 0.77e0.83, p < 0.01). Agreement on presence of anxiety
disorders between parent interview and composite scores were
good to very good (k 0.72e0.82, p < 0.01). Agreement on presence
of anxiety disorders was lower between child- and parent-based
interviews (k 0.37e0.46, p < 0.01), and child and composite in-
terviews (k 0.56e0.68, p < 0.01).
3.2. Missing data
CSR ratings were missing for three participants (3%) at post-
treatment and seven participants (15%) at three-month follow-up.
Self-assessments were completed by 73 children (78%) and 76
parents (82%) at post-treatment and 35 (76%) parents and children
in the treatment group at three-month follow-up (see Fig. 3). No
52 S. Vigerland et al. / Behaviour Research and Therapy 76 (2016) 47e56

Table 2
Demographic and clinical information.

Total sample (N 93) ICBT (n 46) WL (n 47) p-value

Children females, n (%) 51.0 (55) 26 (57) 25 (54) 0.747

Age of children, M (SD) 10.1 (1.7) 10.3 (1.2) 9.9 (1.1) 0.109
Age of parents, M (SD) 43.1 (5.3) 42.8 (5.0) 43.4 (5.6) 0.483
Range 31e60 31e58 33e60
Parental education level, n (%) 0.377
9 years 6 (3) 5 (5.4) 1 (1.1)
12 years 34 (19) 19 (20.7) 15 (16.3)
Vocational education 14 (8) 6 (6.5) 8 (8.7)
Current university studies 6 (3) 2 (2.2) 4 (4.3)
University degree 124 (67) 60 (65.2) 64 (69.6)
Number of diagnoses, n (%) 0.648
1 30 (32) 15 (33) 15 (32)
2 27 (29) 14 (30) 13 (28)
3 21 (22) 10 (22) 11 (23)
4 8 (9) 4 (9) 4 (9)
>4 7 (8) 3 (7) 4 (9)
CDI score at baseline, M (SD) 6.6 (4.8) 6.8 (5.2) 6.3 (4.6) 0.632
Diagnoses, n (%) 0.918
GAD 20 (22) 11 (24) 9 (19)
Panic disorder 5 (5) 2 (4) 3 (6)
Separation anxiety 30 (32) 16 (35) 14 (30)
Social anxiety disorder 9 (10) 4 (9) 5 (11)
Specic phobia 29 (31) 13 (28) 16 (34)

Note: CDI Children's Depression Inventory; GAD Generalised anxiety disorder; ICBT Internet-delivered cognitive behavioural therapy; WL waitlist control.

signicant differences were found on pre- or post-treatment scores
on CSR or SCAS (child and parent reported), age, gender, CDI,
principal diagnosis or number of diagnoses between children who
did and children who did not complete self-report measures at
post-treatment. Between families where at least one parent
completed measures at post-treatment and families where none of
the parents completed measures at post-treatment there were
signicant differences on the number of completed modules (with
families where at least one parent completed measures at post-
treatment completing more modules on average) (t (44) 2.04,
p 0.048). Furthermore, there was a signicant difference between
families who did and did not participate in the ADIS interview at
three-month follow-up, with parents not participating having
higher post scores on the CSR and the SCAS-P.
3.3. Treatment outcomes
Pre- and post-treatment assessments for both groups including
interaction effects and between-group effects are presented in
Table 3 together with the follow-up assessments for the treatment
group. Within-group effects sizes are presented for both groups in
Table 4.
The pre-to post-treatment change on the CSR, the primary
outcome, was signicantly larger in the treatment group than in the
waitlist group at post-treatment, with a large between-group effect
size of d 1.66. CSR in the treatment group showed additional
signicant decrease between post-treatment and 3-month follow-
up (t (38) 4.72, p < 0.001). The within-group effect size between
pre-treatment and follow-up was large (d 1.84) (see Table 4).
Nine (20%) participants in the treatment group no longer met
criteria for their principal diagnosis at post-treatment compared to
three (7%) in the waitlist group. These proportions did not signi-
cantly differ (x2 (1) 3.60, p 0.058). At three-month follow-up
the number had increased to 23 (50%) in the treatment group.
The pre-to post-treatment change on CGAS-ratings was signif-
icantly higher, indicating higher functioning, in the treatment
group than in the waitlist group at post-treatment. The between-
group effect size was small (d 0.45). An additional signicant
increase in ratings of functioning was seen between post-treatment
and three-month follow-up in the treatment group (t (38) 5.14,
p < 0.001) and the within-group effect size between pre-treatment
and three-month follow-up was large (d 1.24).
Parent-reported child anxiety, measured with SCAS-P, showed a
signicantly greater pre-to post-treatment change in the treatment
group compared to the waitlist group, with a small between-group
effect size (d 0.45). At three-month follow-up, the within-group
effect size, compared to pre-treatment had increased from d 0.66
to d 0.91 and there was a signicant decrease in ratings (t
(27) 2.18, p < 0.038). There were no signicant differences on pre-
to post-treatment changes in SCAS-C between treatment and
waitlist group, although ratings dropped in both groups (within-
group effect size d 0.51 and d 0.56 in the treatment and waitlist
group, respectively). At three-month follow-up ratings had drop-
ped, compared to post-treatment, in the treatment group. How-
ever, the change was not statistically signicant (t (24) 1.93,
p < 0.065). No signicant group differences were found on pre-to
post-treatment changes on QOLI and no signicant difference was
seen between post-treatment and follow-up (t (24) 0.56,
p < 0.582) in the treatment group.
Diagnosis specic measures (child and parent versions of FSSC-
R, PSWQ, SAI or SPAI) did not reveal any signicant group differ-
ences on pre-to post-treatment changes. Due to a large amount of
missing data on these measures, test of maintenance of improve-
ment between post-treatment and FU was not conducted for the
specic measures. Detailed results are presented in Table S1
(available online).
3.4. Treatment satisfaction
Of the 46 families randomised to ICBT, 32 children and 62 par-
ents completed the CSS. Overall, children and parents were
moderately satised with the treatment with a mean rating of 3.67
and 3.78 respectively. Parents rated item I would recommend this
treatment highest, with 86% indicating Agree or Very much
agree and item My child is not more scared than other children
his/her age the lowest, with only 31% indicating Agree or Very
much agree. Children rated item This treatment was effective the
highest with 82% responding Agree or Very much agree, and
 S. Vigerland et al. / Behaviour Research and Therapy 76 (2016) 47e56 53

to 24,408 (M 9666, SD 5663, Mdn 9084), which corresponds

Telephone
interview Withdrew n=5
to approximately 0.5e11 pages (M 4.4, SD 2.6, Mdn 4.13).
n=170 Linear regression showed that post-treatment CSR scores were not
signicantly predicted by number of completed modules
Excluded n=29
Did not fulfill inclusion criteria: (Beta 0.169, t 1.12, p 0.268) or number of characters entered
a=6, b=9 and d=1 in the platform (Beta 0.04, t 0.262, p 0.795). Although fam-
Fulfilled exclusion criteria: ilies were instructed to log in to the platform and report completed
a=6, c=4 and e=3 exposure tasks, no families followed this recommendation in a
systematic way. No participants in the treatment group that
Screening with
DAWBA Withdrew/di d not answer n=11 participated in follow-up assessment had sought help for the
n=136 child's anxiety problems elsewhere.

Excluded n=3
4. Discussion
Did not fulfill inclusion criteria:
b=3 The aim of the present study was to evaluate the efcacy of ICBT
for children with anxiety disorders. On the primary outcome
measure, we found a large effect size in favour of the treatment
Self report
questionnaires Withdrew/di d not answer n=11 group at post-treatment and effects were maintained at three-
n=122 month follow-up. There was a small but signicant interaction ef-
fect of time and treatment on parent-rated anxiety at post-
Excluded n=4
Fulfilled exclusion criteria:
treatment, but no signicant effects were observed on child-
b=4 ratings of anxiety.
CSR ratings showed a signicantly larger decrease in the treat-
Diagnostic Withdrew/di d not participate ment group compared to the waitlist group at post-treatment, but
Interview n=6
the mean CSR rating did not fall below four, which is the cut off for
n=107
Excluded n=8 clinical level of impairment and severity, until three-month follow-
Did not fulfill inclusion criteria: up. The large between-groups effect on CSR was comparable to
b=5 March et al. (2009) although they reported a larger decrease of
Fulfilled exclusion criteria:
e=3 absolute CSR scores at post-treatment and follow-up. March et al.
Randomized (2009) had a longer follow-up time and also included booster
N=93 sessions after one and three months, which could account for the
greater improvement (although only 41% of children completed the
booster sessions).
The proportion of participants no longer meeting criteria for
Waitlist ICBT their principal diagnosis did not signicantly differ between groups
n=47 n=46 at post-treatment, the proportions being 20% and 7% in the treat-
ment and waitlist group, respectively. This was somewhat lower
than what was seen in March et al. (2009) where the corresponding
Post-treatment Post-treatment proportions were 30% and 10%. March et al. also reported that 75%
Primary outcome n=45 Primary outcome n=45 Families that did not no longer met criteria for their primary diagnosis at 6-month
Child report n=42 Child report n=31 participate at all n=0 follow-up. A review by James et al. (2013) found average remis-
Parent report n=43 Parent report n=33
sion rates for anxiety disorders after CBT to be 56%. Consistent with
ndings in Vigerland et al. (2013), the treatment in the present
study seems to have a delay in treatment effect, reaching compa-
Three-month rable levels of remission (50%) at three-month follow-up. As this
follow-up effect was uncontrolled, it needs to be interpreted cautiously. Also,
Families that did not
Primary outcome n=39 there was a larger proportion of missing data at three-month
participate at all n=6
Child report n=35
Parent report n=35
follow-up (15%), and analyses showed that participants with
missing follow-up data had signicantly higher CSR and SCAS-P
Fig. 3. Participation ow through the study. scores at post treatment than those who completed the follow-up
interview. This suggests that results on the CSR and on diagnostic
status in the observed sample may be a slight overestimation.
item After the treatment my fears don't affect my life at all the On the CGAS-ratings of global functioning, the small effect size
lowest, with only 44% responding Agree or Very much agree. found in this study at post-treatment is considerably lower than in
Results are presented in detail in Table S2 (available online). March et al. (2009) and the absolute difference between pre- and
post-treatment scores was small.
3.5. Treatment compliance and further treatment seeking No signicant treatment effect on child ratings of anxiety, SCAS-
C, was found in the present study. This is in line with March et al.
Not all families completed all modules in the recommended (2009), who reported no signicant treatment effects for levels of
pace. The mean number of completed modules was 9.7 (SD 1.8; child rated anxiety symptoms. Furthermore, other studies of CBT
range 4e11) and 83% completed the rst nine modules. Twenty- for children with anxiety disorders have not found any signicant
nine participants (63%) completed at least one of the two mainte- between group treatment effect on the SCAS-C, and it may be that
nance modules. Four families (9%) did not reach the modules that the properties of the measure are not optimal for treatment eval-
were intended for both children and parents (starting with module uation (Spence et al., 2011; Spence, Holmes, March, & Lipp, 2006;
7). Number of characters written in the platform ranged from 1440 Wergeland et al., 2014). On SCAS-P, a small signicant treatment
54 S. Vigerland et al. / Behaviour Research and Therapy 76 (2016) 47e56

Table 3
Treatment outcomes - Observed means and standard deviations for treatment- and waitlist group at pre-treatment, post-treatment and follow-up (for the treatment group).
Between-groups effect sizes based on the estimates obtained in the hierarchical linear mixed models are presented together with the coefcient (Dslope) for the interaction
effect between time and study group.

Pre Post FU Between-group differences

n M SD n M SD n M SD In pre-to-post change [95% CI]

CSR
ICBT 46 5.7 0.7 45 4.4 1.2 39 3.3 1.6 Dslope 1.16 [0.77, 1.55]
WL 47 5.6 0.7 45 5.4 1.1 Effect size d 1.66 [1.18, 2.14]
CGASa
ICBT 46 55.2 4.7 45 58.1 4.4 39 61.3 5.1 Dslope 2.31 [1.43, 3.19]
WL 47 57.5 5.4 45 57.9 5.6 Effect size d 0.45 [0.03, 0.87]
SCAS-C
ICBT 46 35.9 13.7 31 29.0 13.6 35 25.5 16.3 Dslope 0.06 [4.45, 4.57]
WL 47 34.4 13.3 42 27.1 12.9 Effect size d <0.01 [0.47, 0.46]
SCAS-P
ICBT 46 32.5 9.3 33 25.7 11.2 35 22.7 12.0 Dslope 4.29 [7.81, 0.77]
WL 47 28.6 9.3 43 25.7 10.2 Effect size d 0.45 [0.01,0.9]
QOLIa
ICBT 46 4.1 1.1 31 3.7 1.4 35 3.9 1.1 Dslope 0.29 [0.75, 0.17]
WL 47 4.0 1.1 42 4.0 1.1 Effect size d 0.26 [0.72, 0.27]

Note: CDI Children's Depression Inventory; CGAS Children's Global Assessment Scale; CSR Clinician Severity Rating; GAD Generalised anxiety disorder;
ICBT Internet-delivered cognitive behavioural therapy; SCAS C/P Spence Children's Anxiety Scale; QOLI Quality of Life Inventory; WL waitlist control.
a
Higher scores indicate higher functioning. Sign of effect sizes have been changed so that positive scores indicate improvement.

Table 4 to the present study, with regard to the inclusion of children with
Within-group effect sizes. diagnosed anxiety disorders and the use of a CBT-program with non
Within-group effect sizes [95% CI] face-to-face therapist support, they were aimed at adolescents and
Pre-post Pre-FU Post-FU
are therefore not completely comparable. Even so, results are
largely similar. Spence et al. (2011) and Wuthrich et al. (2012) both
CSR
showed a signicant between-group effect in favour of the treat-
ICBT 1.32 [0.88, 1.76] 1.84 [1.19, 2.49] 0.70 [0.41,0.99]
WL 0.19 [0.03, 0.42] ment group, compared to waitlist, and a large within-group effect
CGASa size on CSR ratings in the treatment group. Proportion of partici-
ICBT 0.63 [0.43, 0.83] 1.24 [0.84, 1.65] 0.67 [0.42,0.91] pants free of their principal anxiety disorder at post-treatment was
WL 0.06 [0.0, 0.12]
34% and 41% for Spence et al. and Wuthrich et al., respectively.
SCAS-C
ICBT 0.51 [0.18, 0.84] 0.63 [0.13, 1.14] 0.18 [0.05,0.40] Spence et al. did not nd signicantly larger improvements for ICBT
WL 0.56 [0.35, 0.77] on SCAS-C or SCAS-P compared to waitlist control, while Wuthrich
SCAS-P et al. found signicantly larger decreases in the treatment group
ICBT 0.66 [0.27, 1.05] 0.91 [0.39, 1.43] 0.26 [0.02,0.50] from pre-to post-treatment on SCAS-C and SCAS-P, with large
WL 0.29 [0.11, 0.47]
within-group effect sizes. Khanna and Kendall's (2010) evaluation
QOLIa
ICBT 0.30 [0.03, 0.57] 0.21 [0.15, 0.56] 0.12 [0.42,0.17] of computer-assisted CBT, including several therapist-led exposure
WL 0.02 [0.28, 0.33] sessions, also showed results in favour of computer-assisted CBT on
Note: CDI Children's Depression Inventory; CGAS Children's Global Assessment CSR ratings compared to a computerised control intervention, but
Scale; CSR Clinician Severity Rating; GAD Generalised anxiety disorder; no signicant differences on child ratings of anxiety.
ICBT Internet-delivered cognitive behavioural therapy; SCAS C/P Spence Chil- In summary, treatment outcomes in the present study were not
dren's Anxiety Scale; QOLI Quality of Life Inventory; WL waitlist control. substantially different from those in other, similar studies. The
a
Higher scores indicate higher functioning. Sign of effect sizes have been changed
somewhat weaker treatment outcomes seen in this study could be
so that positive scores indicate improvement.
associated with poor treatment engagement. Although a majority
of families completed all the psychoeducation there are no avail-
able data on number of exposures or amount of time that families
effect (d 0.44) was found at post-treatment, which is comparable
worked with the treatment outside of the computer. The experi-
to March et al. (2009) where the corresponding effect size was
ence of the psychologists in the study was that families understood
d 0.31.
the concept of exposure, but did not practice exposure tasks to the
Diagnosis specic measures showed no signicant effects at
extent that was necessary.
post-treatment. This could partly be explained by a lack of power,
Although the sample was relatively high-functioning with
as there were few patients (9e29) in each diagnostic group and
baseline CSR- and CGAS-scores indicating less severity than seen in
quite a large number of missing data on these measures (11%e43%)
for instance March et al. (2009) and Kendall et al. (2010), parent
at post-treatment. On the child versions of the FSSC-R there was an
ratings on SCAS-P showed levels of anxiety symptom consistent
increase in the scores (indicating more fear) in the treatment group.
with a clinical sample (Nauta et al., 2004). Moreover, a majority of
Questions on treatment satisfaction focused largely on
included children fullled criteria for at least two anxiety disorders,
perceived treatment outcome and a majority of participants in the
indicating a clinical, if not a highly severe, sample.
treatment group rated the treatment as helpful. Indeed, 86 per cent
of parents indicated they would recommend the treatment to a
friend with similar problems. Overall, parents and children were 4.1. Limitations
neutral or positive, which is in accordance with the moderate
treatment effects found at post-treatment. The study has a number of limitations, which should be taken
While Spence et al. (2011) and Wuthrich et al. (2012) are similar into consideration when interpreting the results. First of all, the
 S. Vigerland et al. / Behaviour Research and Therapy 76 (2016) 47e56
primary outcome measure was not rated by blind assessors, but by
assessors that were aware of treatment allocation, and involved in
the project. It cannot be ruled out that this can have affected the
results. Also, CSR scores at post-treatment and follow-up were
based solely on parent information whereas they were based on
child- and parent report at pre-treatment. Limited resources
necessitated telephone interviews at post-treatment and follow-
up, and only parents were interviewed as some children were
considered too young to participate in a telephone interview. At
pre-treatment, diagnoses based only on parent interviews showed
good to excellent agreement with composite diagnoses, while
agreement between child based diagnoses and composite di-
agnoses was lower. This is in line with previous research that has
shown that clinicians are more inuenced by parent report when
assigning diagnoses (Grills & Ollendick, 2003, Rapee, Barrett,
Dadds, & Evans, 1994). Together with results from Lyneham and
Rapee (2005), suggesting that relying only on information from
the parent can be sufcient in determining presence or absence of
diagnoses, it suggests that results in this study might not have been
substantially different if child-report had been included in post-
treatment assessment. However, the results should be interpreted
cautiously.
Secondly, we used a waitlist-control and not an active control
group, and thus we did not control for effects of expectancy of
improvement or attention from a caregiver. In addition, there was
no control group at three-month follow-up and it cannot be ruled
out that the further improvement that was found in the treatment
group was due to passage of time or natural recovery rather than an
effect of treatment.
Furthermore, there are no valid data on treatment compliance.
The data that are available on number of completed stages might
not provide a fair description of participants activity. Unfortu-
nately, we have no data on amount of exposure, which we believe
to be the component that contributes mostly to positive treatment
outcome. Also, we believe that some participants would have
beneted from a more regular contact with a supportive therapist.
Future trials should include a more regular interaction with the
treatment platform and contact with the therapist as a way of
increasing amount of exposure exercises, improving measurement
of treatment compliance and possibly improving effects.
Finally, this study does not include data on therapist time,
number of additional phone calls, length of phone calls or other
therapist variables. This should be addressed in future studies as it
could be important with regard to generalisability and cost-
effectiveness.
5. Conclusions
Although this study has some limitations, it supports previous
ndings that ICBT with therapist support can reduce clinician- and
parent-rated symptoms of anxiety. However, there is still need to
better understand secondary outcomes, including child-ratings of
anxiety and quality of life, and to get more reliable estimates of
follow-up outcomes. Future studies should include blind assessors,
a larger sample and an active control group. If ICBT for children with
anxiety disorders can prove its efcacy and effectiveness in future
studies, it could be a way of making CBT more available for this
population.
Acknowledgements
We are indebted to the families that took part in the study, and
m, Jenny Meyer, Sara Rose
to Mari Ljungstro
n, Kristina Loftsson and
Ola Enstrom who helped us with the assessments. Financial sup-
port was provided from the Stockholm County Council (HNS 9576)
55
and the regional agreement on medical training and clinical
research between Stockholm County Council and the Karolinska
Institutet (ALF 20110278 and 20120070), the Claes Groschinsky
foundation (SF11 147). These results were reported in part at the
annual conferences of the Anxiety and Depression Association of
America, European Society for Child and Adolescent Psychiatry and
World Congress of Behavioural and Cognitive Therapies in 2013.
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.brat.2015.11.006.
References
Albano, A., & Silverman, W. (1996). Clinical manual for anxiety disorders interview
schedule for children for DSM-IV (Child version). New York: Psychological
Corporation.
American Psychiatric Association. (2000). Diagnostic and statistical manual of mental
disorders (4th ed.). Washington, DC: Author.
Andersson, G. (2014). The Internet and CBT: A clinical guide. Boca Raton: CRC Press.
Andersson, G., Cuijpers, P., Carlbring, P., Riper, H., & Hedman, E. (2014). Guided
Internet-based vs. face-to-face cognitive behavior therapy for psychiatric and
somatic disorders: a systematic review and meta-analysis. World Psychiatry:
Ofcial Journal of the World Psychiatric Association, 13(3), 288e295. http://
dx.doi.org/10.1002/wps.20151.
Beidel, D., Turner, S., & Morris, T. (1995). A new inventory to assess childhood social
anxiety and phobia: the Social Phobia and Anxiety Inventory for Children.
Psychological Assessment, 7(1), 73e79.
Bird, H., Canino, G., & Rubio-Stipec, M. (1987). Further measures of the psycho-
metric properties of the Children's Global Assessment Scale. Archives of General
Psychiatry, 44, 821e824.
Borenstein, M., Hedges, L., Higgins, J., & Rothstein, H. (2009). Introduction to meta-
analysis. Chichester, UK: John Wiley & Sons, Ltd. http://dx.doi.org/10.1002/
9780470743386.
Chavira, D., Stein, M., Bailey, K., & Stein, M. (2005). Child anxiety in primary care:
prevalent but untreated. Depression and Anxiety, 20(4), 155e164. http://
dx.doi.org/10.1002/da.20039.
Chorpita, B., Tracey, S., Brown, T., Collica, T., & Barlow, D. (1997). Assessment of
worry in children and adolescents: an adaptation of the Penn State Worry
Questionnaire. Behaviour Research and Therapy, 35(6), 569e581.
Cohen, J. (1960). A coefcient of agreement for nominal scales. Educational and
Psychological Measurement, 20(1), 37e46. http://doi.org/10.1177/
001316446002000104.
Costello, E., Angold, A., & Keeler, G. (1999). Adolescent outcomes of childhood
disorders: the consequences of severity and impairment. Journal of the Amer-
ican Academy of Child & Adolescent Psychiatry, 38(2), 121e128. http://dx.doi.org/
10.1097/00004583-199902000-00010.
Costello, J., He, J., Sampson, N., Kessler, R., & Merikangas, K. (2014). Services for
adolescents with psychiatric disorders: 12-month data from the National Co-
morbidity Survey-Adolescent. Psychiatric Services, 65(3), 359e366. http://
dx.doi.org/10.1176/appi.ps.201100518.
Dadds, M., & Barrett, P. (2001). Practitioner review: psychological management of
anxiety disorders in childhood. Journal of Child Psychology and Psychiatry, and
Allied Disciplines, 42(8), 999e1011.
Donovan, C. L., & March, S. (2014). Online CBT for preschool anxiety disorders: a
randomised control trial. Behaviour Research and Therapy, 58(c), 24e35. http://
doi.org/10.1016/j.brat.2014.05.001.
Feingold, A. (2009). Effect sizes for growth-modeling analysis for controlled clinical
trials in the same metric as for classical analysis. Psychological Methods, 14(1),
43e53.
Fichter, M., Quadieg, N., Fischer, U., & Kohlboeck, G. (2009). Twenty-ve-year
course and outcome in anxiety and depression in the Upper Bavarian Longi-
tudinal Community Study. Acta Psychiatrica Scandinavica, 122(1), 75e85. http://
dx.doi.org/10.1111/j.1600-0447.2009.01512.x.
Frisch, M., Cornell, J., Villanueva, M., & Retzlaff, P. (1992). Clinical validation of the
Quality of Life Inventory: a measure of life satisfaction for use in treatment
planning and outcome assessment. Psychological Assessment, 4(1), 92e101.
Ginsburg, G., La Greca, A., & Silverman, W. (1998). Social anxiety in children with
anxiety disorders: relation with social and emotional functioning. Journal of
Abnormal Child Psychology, 26(3), 175e185.
Goodman, R., Ford, T., Richards, H., Gatward, R., & Meltzer, H. (2000). The devel-
opment and well-being Assessment: description and initial validation of an
integrated assessment of child and adolescent psychopathology. Journal of Child
Psychology and Psychiatry, and Allied Disciplines, 41(5), 645e655.
Grills, A. E., & Ollendick, T. H. (2003). Multiple informant agreement and the anxiety
disorders interview schedule for parents and children (Vol. 42(1), pp. 30e40).
http://doi.org/10.1097/00004583-200301000-00008.
Gueorguieva, R., & Krystal, J. (2004). Move over ANOVA: progress in analyzing
repeated-measures data and its reection in papers published in the archives of
56 S. Vigerland et al. / Behaviour Research and Therapy 76 (2016) 47e56
general psychiatry. Archives of general Psyciatry, 61(3), 310e317.
Hedman, E., Ljo
tsson, B., & Lindefors, N. (2012). Cognitive behavior therapy via the
Internet: a systematic review of applications, clinical efcacy and cost-
eeffectiveness. Expert Review of Pharmacoeconomics & Outcomes Research, 12(6),
745e764. http://dx.doi.org/10.1586/erp.12.67.
Higa, C., Fernandez, S., Nakamura, B., Chorpita, B., & Daleiden, E. (2006). Parental
assessment of childhood social phobia: psychometric properties of the social
phobia and anxiety inventory for children-parent report. Journal of Clinical Child
and Adolescent Psychology, 35(4), 590e597. http://dx.doi.org/10.1207/
s15374424jccp3504_11.
In-Albon, T., Meyer, A. H., & Schneider, S. (2013). Separation anxiety avoidance
inventory-child and parent version: psychometric properties and clinical utility
in a clinical and a school sample. Child Psychiatry and Human Development, 44,
689e687.
Ivarsson, T., Svalander, P., & Litlere, O. (2006). The Children's Depression Inventory
(CDI) as measure of depression in Swedish adolescents. A normative study.
Nordic Journal of Psychiatry, 60(3), 220e226. http://dx.doi.org/10.1080/
08039480600636395.
James, A., James, G., Cowdrey, F., Soler, A., & Choke, A. (2013). Cognitive behavioural
therapy for anxiety disorders in children and adolescents (p. CD004690). http://
dx.doi.org/10.1002/14651858.CD004690.pub3. The Cochrane Database of Sys-
tematic Reviews, 6.
Kazdin, A. (2010). Research design in clinical psychology (4th ed.). Needham Heights,
MA: Allyn & Bacon.
Kendall, P. C., Compton, S. N., Walkup, J. T., Birmaher, B., Albano, A. M., Sherrill, J.,
et al. (2010). Clinical characteristics of anxiety disordered youth. Journal of
Anxiety Disorders, 24(3), 360e365. http://doi.org/10.1016/j.janxdis.2010.01.009.
Khanna, M., & Kendall, P. (2010). Computer-assisted cognitive behavioral therapy for
child anxiety: results of a randomised clinical trial. Journal of Consulting and
Clinical Psychology, 78(5), 737e745. http://dx.doi.org/10.1037/a0019739.
Kovacs, M. (1985). The Children's Depression, Inventory (CDI). Psychopharmacology
bulletin, 21(4), 995e998.
Lindner, P., Andersson, G., Ost, L., & Carlbring, P. (2013). Validation of the internet-
administered quality of life inventory (QOLI) in different psychiatric conditions.
Cognitive Behaviour Therapy, 42(4), 315e327. http://dx.doi.org/10.1080/
16506073.2013.806584.
Lundh, A., Kowalski, J., Sundberg, C., Gumpert, C., & Lande
n, M. (2010). Children's
Global Assessment Scale (CGAS) in a naturalistic clinical setting: Inter-rater
reliability and comparison with expert ratings. Psychiatry Research, 177(1e2),
206e210. http://dx.doi.org/10.1016/j.psychres.2010.02.006.
Lyneham, H., & Rapee, R. (2005). Agreement between telephone and in-person
delivery of a structured interview for anxiety disorders in children. Journal of
the American Academy of Child and Adolescent Psychiatry, 44(3), 274e282.
Mannuzza, S., Fyer, A. J., Martin, L. Y., Gallops, M. S., Endicott, J., Gorman, J., et al.
(1989). Reliability of anxiety assessment. I. Diagnostic agreement. Archives of
General Psychiatry, 46, 1093e1101.
March, S., Spence, S., & Donovan, C. (2009). The efcacy of an internet-based
cognitive-behavioral therapy intervention for child anxiety disorders. Journal
of Pediatric Psychology, 34(5), 474e487. http://dx.doi.org/10.1093/jpepsy/jsn099.
Nauta, M., Scholing, A., Rapee, R., Abbott, M., Spence, S., & Waters, A. (2004).
A parent-report measure of children's anxiety: psychometric properties and
comparison with child-report in a clinic and normal sample. Behaviour Research
and Therapy, 42(7), 813e839. http://dx.doi.org/10.1016/S0005-7967(03)00200-
6.
Ollendick, T. (1983). Reliability and validity of the revised fear surgery schedule for
children (FSSC-R). Behaviour Research and Therapy, 21(6), 685e692. http://
dx.doi.org/10.1002/da.20344.
Ollendick, T. (2010). Personal communication.
Paunovi

c, N., & Ost, L. G. (2004). Clinical validation of the Swedish version of the
quality of life inventory in crime victims with posttraumatic stress disorder and
a nonclinical sample. Journal of Psychopathology and Behavioral Assessment,
26(1), 15e21.
Pennant, M. E., Loucas, C. E., Whittington, C., Creswell, C., Fonagy, P., Fuggle, P., et al.
(2015). Computerised therapies for anxiety and depression in children and
young people: a systematic review and meta-analysis. Behaviour Research and
Therapy, 67, 1e18. http://doi.org/10.1016/j.brat.2015.01.009.
Pine, D., & Klein, R. (2008). Anxiety disorders. Rutter's child and adolescent psychiatry
(pp. 628e647). Oxford, UK: Blackwell Publishing Ltd. http://dx.doi.org/10.1002/
9781444300895.ch39.
Rapee, R. M., Barrett, P. M., Dadds, M. R., & Evans, L. (1994). Reliability of the DSM-
III-R childhood anxiety disorders using structured interview: interrater and
parent-child agreement. Journal of the American Academy of Child & Adolescent
Psychiatry, 33(7), 984e992. http://doi.org/10.1097/00004583-199409000-
00008.
Reyes-Portillo, J. A., Mufson, L., Greenhill, L. L., Gould, M. S., Fisher, P. W., Tarlow, N.,
et al. (2014). Web-based interventions for youth internalizing problems: a
systematic review. Journal of the American Academy of Child & Adolescent Psy-
chiatry, 1e22. http://doi.org/10.1016/j.jaac.2014.09.005.
Richardson, T., Stallard, P., & Velleman, S. (2010). Computerised cognitive behav-
ioural therapy for the prevention and treatment of depression and anxiety in
children and adolescents: a systematic review. Clinical Child and Family Psy-
chology Review, 13(3), 275e290. http://dx.doi.org/10.1007/s10567-010-0069-9.
Shaffer, D., Gould, M., Brasic, J., Ambrosini, P., Fisher, P., Bird, H., et al. (1983).
A children's global assessment scale (CGAS). Archives of General Psychiatry,
40(11), 1228e1231.
Silverman, W., Saavedra, M., & Pina, A. (2001). Test-retest reliability of anxiety
symptoms and diagnoses with the anxiety disorders interview schedule for
DSM-IV: child and parent versions. Journal of the American Academy of Child &
Adolescent Psychiatry, 40(8), 937e944. http://dx.doi.org/10.1097/00004583-
200108000-00016.
Spence, S. (1998). A measure of symptoms among children. Behaviour Research and
Therapy, 36, 545e566.
Spence, S., Barrett, P., & Turner, C. (2003). Psychometric properties of the Spence
Children's Anxiety Scale with young adolescents. Journal of Anxiety Disorders,
17(6), 605e625. http://dx.doi.org/10.1016/S0887-6185(02)00236-0.
Spence, S., Donovan, C., March, S., Gamble, A., Anderson, R., Prosser, S., et al. (2011).
A randomised controlled trial of online versus clinic-based CBT for adolescent
anxiety. Journal of Consulting and Clinical Psychology, 79(5), 629e642. http://
dx.doi.org/10.1037/a0024512.
Spence, S., Holmes, J., March, S., & Lipp, O. (2006). The feasibility and outcome of
clinic plus Internet delivery of cognitive-behavior therapy for childhood anxi-
ety. Journal of Consulting and Clinical Psychology, 74(3), 614e621. http://
dx.doi.org/10.1037/0022-006X.74.3.614.
Spitzer, R., Williams, J., Kroenke, K., Linzer, M., deGruy, F., Hahn, S., et al. (1994).
Utility of a new procedure for diagnosing mental disorders in primary care. The
PRIME-MD 1000 study. JAMA : the Journal of the American Medical Association,
272(22), 1749e1756.
Stallard, P., Richardson, T., Velleman, S., & Attwood, M. (2011). Computerized CBT
(think, feel, do) for depression and anxiety in children and adolescents: out-
comes and feedback from a pilot randomised controlled trial. Behavioural and
Cognitive Psychotherapy, 39(03), 273e284. http://doi.org/10.1017/
S135246581000086X.
Vigerland, S., Thulin, U., Ljo

tsson, B., Svirsky, L., Ost, L., Lindefors, N., et al. (2013).
Cognitive Behaviour Therapy, 42(4), 303e314. http://dx.doi.org/10.1080/
16506073.2013.844201.
Wergeland, G., Fjermestad, K., Marin, C., Haugland, B., Bjaastad, J., Oeding, K., et al.
(2014). An effectiveness study of individual vs. group cognitive behavioral
therapy for anxiety disorders in youth. Behaviour Research and Therapy, 57, 1e12.
http://dx.doi.org/10.1016/j.brat.2014.03.007.
Wood, J., Piacentini, J., Bergman, L., McCracken, J., & Barrios, V. (2002). Concurrent
validity of the anxiety disorders section of the anxiety disorders interview
schedule for DSM-IV: child and parent versions. Journal of Clinical Child &
Adolescent Psychology, 31(3), 335e342. http://dx.doi.org/10.1207/
S15374424JCCP3103_05.
Wuthrich, V. M., Rapee, R. M., Cunningham, M. J., Lyneham, H. J., Hudson, J. L., &
Schniering, C. A. (2012 Mar 1). A randomized controlled trial of the cool teens
CD-ROM computerized program for adolescent anxiety. Journal of the American
Academy of Child & Adolescent Psychiatry, 51(3), 261e270.