Computer-Assisted Cognitive Behavioral Therapy for Child Anxiety:
Results of a Randomized Clinical Trial
Muniya S. Khanna
University of Pennsylvania
Philip C. Kendall
Objective: This study examined the feasibility, acceptability, and effects of Camp Cope-A-Lot (CCAL),
a computer-assisted cognitive behavioral therapy (CBT) for anxiety in youth. Method: Children (49; 33
males) ages 7–13 (M ? 10.1 ? 1.6; 83.7% Caucasian, 14.2% African American, 2% Hispanic) with a
principal anxiety disorder were randomly assigned to (a) CCAL, (b) individual CBT (ICBT), or (c) a
computer-assisted education, support, and attention (CESA) condition. All therapists were from the
community (school or counseling psychologists, clinical psychologist) or were PsyD or PhD trainees with
no experience or training in CBT for child anxiety. Independent diagnostic interviews and self-report
measures were completed at pre- and posttreatment and 3-month follow-up. Results: At posttreatment,
ICBT or CCAL children showed significantly better gains than CESA children; 70%, 81%, and 19%,
respectively, no longer met criteria for their principal anxiety diagnosis. Gains were maintained at
follow-up, with no significant differences between ICBT and CCAL. Parents and children rated all
treatments acceptable, with CCAL and ICBT children rating higher satisfaction than CESA children.
Conclusions: Findings support the feasibility, acceptability and beneficial effects of CCAL for anxious
youth. Discussion considers the potential of computer-assisted treatments in the dissemination of
empirically supported treatments.
Keywords: child anxiety, cognitive behavioral therapy, computer, computer-based treatment
Evidence supports cognitive-behavioral therapy (CBT) for anx-
iety in youth (e.g., Kendall, Hudson, Gosch, Flannery-Schroeder,
& Suveg, 2008; Walkup et al., 2008), but CBT remains rarely
implemented in community settings (Goisman, Warshaw, &
Keller, 1999). One barrier is the lack of a CBT-trained workforce
(Weisz, Hawley, & Jensen-Doss, 2004). Given the high prevalence
(10% to 20%; Costello, Egger, & Angold, 2004) and morbidity
associated with youth anxiety (Achenbach, Howell, McConaughy,
& Stanger, 1995), the dissemination of CBT is a priority (Emslie,
2008); computer-assisted CBT, if effective, could broaden its
availability (Greist, 2008).
Computer-based (stand-alone or self-help) and computer-assisted
(therapist guided, with contact) CBT also offers practical advantages.
Costs may be reduced (McCrone et al., 2004; Newman, 2000; Wright
et al., 2005), and standardization and adherence may be improved.
Programs can be self-paced and can facilitate review of material.
Record keeping and data collection are also easier (Greist, 2008;
Marks, Cavanagh, & Gega, 2007). Preliminary evidence supports
computer-based and computer-assisted interventions with adults (see
Griffiths & Christensen, 2006), and the National Institute for Clinical
Excellence (NICE; 2006) deemed computerized CBT an acceptable
option for adult depression and anxiety. Fewer evaluations exist of
computer-based programs for youth.
Spence and colleagues (Spence, Holmes, March, & Lipp, 2006)
delivered CBT via the Internet in a group format for anxious children
and a waitlist. CLIN-NET delivers eight of 16 sessions via the
Internet. Both treatments outperformed the waitlist, and anxiety re-
ductions were comparable to previous group CBT outcomes. The
Internet-assisted content was acceptable to families, with minimal
dropout and high compliance. Results were promising; however, the
therapists were psychologists trained in CBT for child anxiety, thus
limiting generalizability to untrained clinicians. March, Spence, and
Donovan (2008) evaluated BRAVE, an Internet-based (minimal ther-
apist contact via phone or e-mail) therapy for anxious children (ages
7–13). Children receiving BRAVE, compared with waitlist, showed
small posttreatment reductions in anxiety and increases in function-
ing, but the outcomes were less than in previous studies. Further, only
33% of children and 60% of parents completed all sessions. By
6-month follow-up, 62% of children and 72.3% of parents had com-
pleted all sessions, and outcomes improved and were similar to those
found in previous trials.
The present study evaluated the feasibility, acceptability, and
effects of a computer-assisted CBT, Camp Cope-A-Lot (CCAL;
Kendall & Khanna, 2008), for youth anxiety compared with indi-
vidual CBT (ICBT) and a computer-linked education, support, and
attention (CESA) condition. CCAL is a computer-assisted program
implemented with a mental health provider (“coach”) to ensure
monitoring of symptoms, alliance, compliance, and the integrity of
the empirically supported CBT (see Spek et al., 2007). The pro-
Muniya S. Khanna, Department of Psychiatry, Child and Adolescent
OCD, Tic, Trich & Anxiety Group, University of Pennsylvania; Philip C.
Kendall, Department of Psychology, Temple University.
The authors receive royalties from sales of materials related to the
treatment of anxiety in youth, including Camp Cope-A-Lot. This study was
supported in part by Grants MH067481 and MH084321 from the National
Institute of Mental Health.
Correspondence concerning this article should be addressed to Muniya
S. Khanna, University of Pennsylvania, Department of Psychiatry, 3535
Market Street, Suite 600, Philadelphia, PA 19104. E-mail: muniya@
Journal of Consulting and Clinical Psychology
2010, Vol. 78, No. 5, 737–745
© 2010 American Psychological Association
gram guides the coach and does not require previous CBT training.
It was hypothesized that CCAL would be acceptable to children
and feasible for novice therapists to implement in their setting. We
also hypothesized that CCAL and ICBT would produce significant
reductions in anxiety in comparison to the CESA condition and
that posttreatment gains would remain at follow-up.
Participants were children (ages 7–13) who met criteria in the
Diagnostic and Statistical Manual of Mental Disorders (4th ed.;
DSM–IV) for a principal anxiety disorder (i.e., separation anxiety,
social phobia, generalized anxiety disorder, specific phobia, or
panic disorder). Exclusion criteria included1(a) full-scale IQ be-
low 80, (b) psychotic symptoms, (c) use of antianxiety or depres-
sant medications or ongoing psychotherapy, (d) and lack of flu-
ency in English. Children were referred from schools or by
practitioners or parents. Recruitment began in September 2006,
and all follow-ups were completed by July 2008. A 20-min tele-
phone screen was followed for initial inclusion or exclusion. As
noted in the CONSORT flowchart (see Figure 1), 55 of 76 children
completed the intake assessment. One parent provided written
informed consent, and children provided written assent (protocol
approved by Temple University Institutional Review Board). De-
termination of eligibility was confirmed by results of intake, and
the study coordinator delivered group assignments to participants
based on a random number generating program.
Forty-nine (33 male) children ages 7–13 (M ? 10.1 ? 1.6) met
criteria and were randomly assigned to ICBT (N ? 17), CCAL
(N ? 16), or CESA (N ? 16), and to a therapist. Therapist
assignment was restricted using block randomization by location
so that children were assigned to a provider in their area.2Forty-
one (83.7%) were Caucasian; seven (14.2%) were African Amer-
ican; and one (2%) was Hispanic. Twenty-eight (57.1%) had
principal generalized anxiety disorder, eight (16.3%) had SP,
seven (14.3%) had separation anxiety, four (8.1%) had specific
phobia, and two (4%) had a principal diagnosis of panic disorder.
Twenty-six (53%) met diagnostic criteria for a secondary diagno-
sis (N ? 15 other internalizing; N ? 8 attention-deficit/
hyperactivity disorder; N ? 2 oppositional defiant disorder; N ? 1
tic disorder). One participant lost contact (CESA ? 1), one with-
drew from the study due to distance/time burden (ICBT ? 1), and
two were withdrawn due to worsening symptoms (CESA ? 2).
The 45 remaining participants completed all 12 sessions within 15
weeks of their first session.
Therapists (N ? 16; 12 female, 75%) were volunteers, including
five school psychologists (two EdS degree holders, one licensed
professional counselor, and two EdS candidates), 10 clinical psy-
chology doctoral candidates (six PhD candidates and four PsyD
candidates), and one clinical psychologist who reported no expe-
rience in CBT for child anxiety. Therapists ranged in experience
from 0 to 11 years (M ? 1.5 ? 2.6); nine had ? 6 months’ and 2
had ? 10 years’ experience. Of the 16 therapists, four reported that
they had read the Coping Cat manual (Kendall & Hedtke, 2006)
and/or attended a workshop but had never implemented the treat-
ment. The remaining eight therapists reported never having read
the Coping Cat manual. Therapists were randomly assigned to
provide CCAL or ICBT, and all provided CESA as well. Thera-
pists received (a) a study orientation and (b) the CCAL computer
software or the Coping Cat manual,3respectively, and the CESA
manual and software. So that ICBT could function as an adequate
benchmark, therapists randomized to ICBT also received a full-day
workshop on the implementation of ICBT, including an overview
of the principles of CBT for anxiety, training in conducting ses-
sions on problem solving, and strategies for effective exposure
tasks. ICBT therapists also participated in weekly supervision of
the implementation of the ICBT protocol via conference calls with
experienced CBT licensed clinical psychologists (Muniya Khanna
and Philip C. Kendall). CCAL therapists participated in a separate
weekly supervision call, which focused on patient safety monitor-
ing and did not include feedback on treatment implementation.
CESA cases were discussed on both calls, with a focus on safety
monitoring and on minimizing bleeding between treatment proto-
Anxiety Disorders Interview Schedule for Children–Parent
Version (ADIS-P; Silverman & Albano, 1997).
structured interview assesses symptoms and severity for DSM–IV
diagnoses and permits diagnoses of comorbidities. Good interrater
and retest reliability (Silverman & Eisen, 1992) have been re-
ported. Parents and children were interviewed together for this
study, following current recommendations to inform endorsement
of a diagnosis and to deal with discrepant reports (Grills & Ollen-
Children’s Global Assessment Scale (CGAS; Shaffer et al.,
Rated by independent evaluators (IEs) following ADIS,
the CGAS reflects general functioning. The CGAS has retest
1A participant was also excluded if he or she had missed more than 50%
of school days in the preceding 2 months. Home schooling did not require
exclusion. Children who met exclusionary criteria were referred as needed.
Participants receiving psychopharmacology (other than antianxiety or an-
tidepressant medications) were included if they were on a 2-month stable
2There were four geographic locations, with patients randomized by a
random number generator to condition within blocks for geographic loca-
tion. This was done to ensure that participants had an equal chance of
getting each treatment in their local area.
3A modified version of the Coping Cat therapist manual was used. The
16-session version was modified, keeping the core components but imple-
menting them in 12 sessions. Parent sessions were included as part of
Sessions 4 and 9, rather than in stand-alone sessions, and review activities
4The parent and child were interviewed together for this study. If there
was disagreement between parent and child in response to an item, we
followed current recommended guidelines, including giving consideration
to external validators of impairment and treating discrepant information
with an “OR” rule—that is, using both perspectives to inform endorsement
of the presence of a symptom or diagnosis (Grills & Ollendick, 2002).
reliability (.69–.95) and sensitivity to levels of impairment (Shaf-
fer et al., 1983).
Multidimensional Anxiety Scale for Children (MASC;
March, Parker, Sullivan, Stallings, & Conners, 1997).
MASC is a 39-item self-report inventory. Retest reliability (mean
intraclass correlation) is excellent over 3 weeks and 3 months (.93
and .78, respectively). Evidence of acceptable convergent and
discriminant validity has been provided (March & Albano, 1998).
Children’s Depression Inventory (CDI; Kovacs, 1981).
27-item CDI has high internal consistency and moderate retest
reliability, and it correlates with measures of related constructs
(Kazdin, French, Unis, Esveldt-Dawson, & Sherick, 1983).
Computer Experience Questionnaire (CEQ).
veloped for this project, assessed the child user’s experience with
and comfort using computers. The CEQ is a four-item, 5-point
scale that children completed at pretreatment.
Therapeutic Alliance, Adherence, and Flexibility.
item, 5-point scale was adapted from previous measures (Chu &
Kendall, 1999) used in previous trials (e.g., Walkup et al., 2008).
All sessions were audiotaped. Two raters (Muniya Khanna and
The CEQ, de-
Consort flowchart of participants.
Philip C. Kendall) independently reviewed 20% of the tapes (110
tapes, 50-min sessions), randomly selected to ensure representa-
tion of age and sex of patient and session number. A score of 3
represents a therapist who was adequately adherent and flexible
and achieved an adequate (“good”) therapeutic alliance.
Client Evaluation of Services (CSQ-8; Larsen, Attkisson,
Hargreaves, & Nguyen, 1979).
satisfaction and has good retest reliability, internal consistency,
and sensitivity to treatment (Nguyen, Attkisson, & Stegner, 1983).
Nguyen et al. (1983) found the CSQ-8 discriminated groups with-
out ceiling effects found in other measures.
The CSQ-8 measures client
IE training on the ADIS-C/P followed recommended guide-
lines.5The IEs reached interrater agreement at the outset (ADIS,
? ? .95, within 1 point on diagnoses in the clinical picture), and
no retraining was required, as interrater agreement was ? .85 for
the duration of the study. Training and reliability on the CGAS was
conducted simultaneously, with interrater agreement established
(CGAS, ? ? .91) at the outset of the study. IEs were blind to
participant treatment condition.
Camp Cope-A-Lot (Kendall & Khanna, 2008).
12-session computer-assisted intervention for anxious children
(ages 7–13 years). Based on the empirically supported Coping Cat
program (Kendall & Hedtke, 2006), CCAL uses Flash animation,
audio, photographs, videos, schematics, a reward system, text, and
a fun cartoon characters to guide the user through the program (see
Figures 2 and 3). CCAL consists of twelve 35-min “levels” and
optional video game rewards. The first six levels, which the user
completes independently, are skill building; the remaining six
levels, to be completed with the assistance of the therapist (the
coach), consist of exposure tasks and rehearsal in the specific
anxiety-arousing situations chosen for each child. Two parent
sessions are conducted by the coach while participant children
work independently on Levels 3 and 7.
The goal was to develop a state-of-the art program that takes
advantage of a multimedia platform to ensure effective and stan-
dardized delivery of CBT content (e.g., psychoeducation, cogni-
tive restructuring, relaxation training, principles of exposure,
homework), while preserving the benefits of face-to-face treat-
ment. Unlike other computer-based treatments, CCAL immerses
the child in an interactive learning environment. The participant
(user) learns experientially rather than by reading pages of online
text or audio/video instruction. CCAL has features that can be
individualized (e.g., theme music, program pace, type of exposure
tasks, video games). The program eliminates the need for specialty
training and reduces required contact hours but allows for moni-
toring of symptoms and building a therapeutic alliance and sup-
ports compliance (e.g., adequate exposure tasks), thereby maxi-
mizing the integrity of the empirically supported CBT (see Khanna
& Kendall, 2008, for a detailed description of the CCAL program).
Individual Cognitive Behavioral Therapy (ICBT).
pants randomized to ICBT received twelve 50-min individual CBT
sessions over 12 weeks. CBT was a 12-session version of the
Coping Cat program (Kendall & Hedtke, 2006), shortened to be
CCAL is a
the same length and contact time as CCAL. The first six sessions
teach skills to the child, and the second six provide opportunities
to practice new skills in exposure tasks both within and outside the
office. The treatment includes strategies such as exposure tasks,
relaxation training, and cognitive retraining and homework.
Computer-Assisted Education, Support, and Attention
CESA controlled for therapist contact, education about
anxiety, computer interaction, expectations, maturation, and the
passage of time. Therapists organized the sessions to include 30
min of education and support and 20 min for the child to use the
computer. The computer access involved a variety of age-
appropriate video games (e.g., Pac-Man, Simon Says, etc.) that
increased in degree of difficulty and intensity with success. A
12-session education, support, and attention manual (as in Kendall
et al., 2008) provided content that included psychoeducation about
anxiety and review homework tasks.
A priori power analyses confirmed that the sample (N ? 49) was
adequate (.81) for the primary analyses to be able to detect mod-
erate to large effects, such as those reported in Silverman et al.
(1999), a study comparing CBT for anxiety to an attention control.
Analyses were conducted using an intent-to-treat sample, where
scores were pulled forward from last assessment period and re-
placed missing values at posttreatment.6Preliminary analyses re-
vealed no significant pretreatment differences across conditions
(see Table 1; i.e., CCAL, ICBT, CESA) on age, gender, race, level
of computer experience (based on CEQ), pretreatment primary
diagnosis, or pretreatment diagnostic severity (i.e., ADIS, MASC,
CGAS, and CDI score).
Child diagnostic status.
by examining participants’ pretreatment principal diagnoses by
condition that were no longer present at posttreatment (i.e., clini-
cian severity rating [CSR] ? 4). Analyses indicated that 70%,
81%, and 19% (ICBT, CCAL, and CESA, respectively) of prin-
cipal diagnoses were no longer present at posttreatment. Both
ICBT–CESA differences and CCAL–CESA differences were sig-
nificant. No significant differences were found for gender or age
(with children categorized developmentally as 7–9 years and
Diagnostic outcomes were analyzed
5Research assistants watched and rated “gold standard” videotapes until
they matched those of the gold standard rater on four tapes. Trainees were
then supervised by an expert in two mock interviews. The IEs then
observed two real interviews and finally administered the interview to two
actual participants and received feedback and supervision. The expert
trainer conducted reliability checks randomly (twice). One IE was a mas-
ter’s level graduate student in clinical psychology. One IE held a B.A.
(postgraduate) with two years of experience conducting phone screens and
attending ADIS supervision in the Child and Adolescent Anxiety Disorders
Clinic. All completed reliability training and reached reliability with the
diagnostic supervisor and with each other.
6Completer analyses were equivalent to those using the intent-to-treat
sample and thus were not interpreted separately.
Camp Cope-A-Lot: sample interactivities.
10–13 years). Twenty-one (43%) participants continued to meet
diagnostic criteria for a secondary diagnosis (N ? 10, other inter-
nalizing; N ? 8, attention-deficit/hyperactivity disorder; N ? 2,
oppositional defiant disorder; N ? 1, tic disorder).
Clinician-rated severity and global assessment of function-
IE ratings of severity (CSR) and global functioning (CGAS)
showed significant change over time (see Table 1), with significant
time effects on CSR (F1, 46? 135.6, p ? .000, ?2? .75) and
CGAS (F2, 46? 141, p ? .000, ?2? .77). Significant Time ?
Condition interaction effects were evident for CSR (F2, 46? 10.5,
p ? .000, ?2? .31) and CGAS (F2, 46? 10.9, p ? .000, ?2?
.38). Children in ICBT and CCAL conditions showed significantly
greater improvement than those in CESA.
Results of analyses of variance showed no
significant condition effect for the CDI or MASC (see Table 1).
However, main effects for time were evident for the MASC
(F2, 46? 10.7, p ? .000, ?2? .39), and CDI (F2, 45? 5.6, p ?
.05, ?2? .12). No significant Time ? Condition effect was found.
Only CCAL (N ? 12) and ICBT participants (N ? 14) were
involved in statistical analyses, because nonresponders (N ? 10) to
the CESA treatment were offered ICBT or CCAL or appropriate
referral and thus were not available. At follow-up, there were four
dropouts from CCAL and two from ICBT. The families were
either not interested in participating in the diagnostic interview
(N ? 2) or nonresponsive to calls (N ? 4). Significant effects for
time across three assessment periods were found (see Table 1) on
CSR (F1, 26? 85.4, p ? .000, ?2? .93), CGAS (F1, 26? 20.9,
p ? .000, ?2? .78), and MASC (F2, 26? 14.6, p ? .000, ?2?
.57). No significant condition effects or Time ? Condition inter-
action effects were found in CSR, CGAS scores, or child-rated
CDI and MASC scores between children in CCAL and ICBT
conditions at 3-month follow-up.
Therapist Adherence, Flexibility,
and Therapeutic Alliance
There were significant therapist adherence differences across
conditions (see Table 2). Ratings indicated significantly greater
therapist adherence to protocol for CCAL, t(40) ? 2.1, p ? .05,
and CESA, t(33) ? 3.2, p ? .01, than for ICBT (F2, 39? 2.6, p ?
.05). However, ratings suggest greater therapist flexibility to meet
the needs of the patient in the ICBT condition (F2, 39? 10.9, p ?
.01) than in CCAL, t(38) ? 3.5, p ? .01, and CESA, t(33) ? 2.6,
p ? .05. There were no significant differences in ratings of the
therapeutic alliance across the treatments.
There were significant differences across conditions in child-
reported satisfaction (see Table 3). CCAL children reported sig-
nificant higher rates of satisfaction on the CSQ-8 than those in
CESA (F2, 45? 6.1, p ? .05), as did children in the ICBT
condition (F2, 45? 3.36, p ? .05). There was no difference in
child-reported satisfaction between ICBT and CCAL. No signifi-
cant difference was found across conditions in parent ratings of
satisfaction, though means were higher for parents of children in
ICBT and CCAL than parents of children in CESA.
Camp Cope-A-Lot: sample characters.
The computer-assisted approach was found to be acceptable to
children and parents and feasible for implementation by providers
with no CBT training. Findings also support effectiveness for
anxiety reduction. Though preliminary, outcomes are encouraging:
Children reached significantly greater treatment gains in the ICBT
and CCAL treatments than in the comparison condition, with more
than twice the number no longer meeting criteria for a principal
anxiety disorder. Also, ICBT and CCAL children showed signif-
icantly greater change in anxiety severity and global functioning,
based on IE ratings, than CESA children. Gains were maintained
at 3-month follow-up, with continued improvement in anxiety
severity and global functioning.
Although active and control treatments differed on both diag-
nostic and continuous measures of anxiety, child self-report of
anxiety did not show posttreatment differences between condi-
tions—all children reported a reduction in anxiety over time. This
finding may reflect that children benefited from the educational
components that were consistent across all conditions or from the
therapeutic relationship. Some improvements in anxiety from ESA
treatments are not uncommon (e.g., Silverman et al, 1999; Kendall
et al., 2008). It may also be that parents, who were not blind to
condition, influenced IE ratings of anxiety severity at posttreat-
Our results found that CCAL had higher ratings of therapist
adherence than ICBT. Having a proportion of the treatment sys-
tematically delivered via computer likely contributed to greater
adherence in novice therapists. Establishing adequate adherence is
not only an important precursor to wide-scale dissemination
(Schoenwald & Hoagwood, 2001) but also may reduce problems
associated with therapist deviations using manual-based treatments
Ratings of therapist flexibility (within fidelity) were greater in
ICBT than in CCAL and CESA, likely due to the delivery of a
Means, Standard Deviations, and Group Differences for Measures of Disorder at Pretreatment, Posttreatment, and 3-Month
Follow-Up Across Conditions
Pretreatment Posttreatment 3-month follow-up
(N ? 17)
(N ? 16)
(N ? 16)
(N ? 17)
(N ? 16)
(N ? 16)
(N ? 14)
(N ? 12)
CSR (95% CI)
CGAS rating (95% CI)
MASC total score (95% CI)
CDI total score (95% CI)
support, and attention control; CSR ? Clinical Severity Rating (based on ADIS-C/P); CGAS ? Clinical Global Assessment Scale; MASC ?
Multidimensional Anxiety Scale for Children; CDI ? Children’s Depression Inventory.
?p ? .05.
ICBT ? individual cognitive behavioral therapy; CCAL ? Camp Cope-A-Lot (computer-assisted therapy); CESA ? computer-assisted education,
??p ? .01.
Means and Standard Deviations (95% CI) of the Therapist
Adherence, Flexibility, and Therapeutic Alliance for the Three
Cope-A-Lot (computer-assisted therapy); CESA ? computer-assisted ed-
ucation, support, and attention control.
?p ? .05.
ICBT ? individual cognitive behavioral therapy; CCAL ? Camp
??p ? .01.
Means and Standard Deviations (95% CI) of Patient
Satisfaction: Child and Parent Ratings
PatientICBT CCAL CESA
Cope-A-Lot (computer-assisted therapy); CESA ? computer-assisted ed-
ucation, support, and attention control.
?p ? .05.
ICBT ? individual cognitive behavioral therapy; CCAL ? Camp
proportion of the content via computer in the CCAL and CESA
conditions. Although both manuals and computers have been
accused of limiting therapist flexibility (e.g., Eifert, Evans, &
McKendrick, 1990), some evidence suggests that the degree of
therapist flexibility is not predictive of outcome status (Chu &
Kendall, 2009). In contrast to concerns voiced in the literature
about computer-based work, therapeutic alliance did not suffer as
a result of delivery via computer (no significant alliance differ-
ences across conditions). In CCAL and CESA, therapist involve-
ment guided 50% of the treatment, which seems to have been
sufficient for a therapeutic alliance. Related to this, therapist
contact and alliance may account for the compliance (100%) in
CCAL compared with compliance in computer-based treatments
with minimal therapist contact (e.g., Spek et al., 2007).
Though both parents and children rated all three treatments as
acceptable, only children rated having more satisfaction with ICBT
and CCAL than with CESA. It may be that because the child
participants had access to the treatment content and delivery ap-
proach, they had greater variability in their ratings, whereas par-
ents’ ratings may reflect a positive experience with the therapists
(i.e., “quality of service”) in all conditions.
Future trials should include (a) longer term follow-up, (b) suf-
ficient samples to examine mediators and moderators of outcome
and to reliably detect small effects, (c) parent self-report measures,
and (d) a more diverse sample, including data regarding socioeco-
nomic status; ours was a primarily Caucasian sample, and socio-
economic status data are not available. An additional limitation of
this study is that all therapists provided CESA, and there was no
random therapist assignment for training to provide CESA. There-
fore, future studies would also benefit from a larger sample of
novice therapists to be able to achieve equal therapist random
assignment to group. Questions also remain regarding differences
in the effectiveness of the interventions, given that no differences
were found in child self-reported symptoms of anxiety and depres-
sion across conditions. Also, it would be of interest to investigate
whom CCAL would be most likely to benefit, particularly with
regard to the type and severity of disorder. Large-scale effective-
ness research is needed to determine the cost-effectiveness of
computer-assisted approaches, the extent to which they will be
adopted by community therapists with sustained use and adherence
to the treatment protocol, and any barriers to implementation and
As March (2009) noted, “more often than not, interventions will
be streamed over the Internet for reasons of uniformity and stan-
dardization, ease of delivery, and cost-effectiveness” (p. 174).
Though many patients will require specialized treatment, there is a
place within stepped care for empirically supported computer-
A significant proportion of children continued to meet diagnos-
tic criteria for a secondary diagnosis; although there were gains,
mean global functioning scores did not return to optimal function-
ing, and more treatment for nontarget issues would be warranted.
Though questions remain and further evaluation research is
needed, the computer-assisted approach holds promise for the
dissemination of evidence-based treatment of child anxiety.
Achenbach, T. M., Howell, C. T., McConaughy, S. H., & Stanger, C.
(1995). Six-year predictors of problems in a national sample of children
and youth: I. Cross-informant syndromes. Journal of the American
Academy of Child & Adolescent Psychiatry, 34, 336–347.
Chu, B. C., & Kendall, P. C. (1999). Therapist flexibility scoring manual:
Coping Cat protocol adherence checklist—Flexibility Scale. Unpub-
lished manuscript, Department of Clinical Psychology, Rutgers Univer-
Chu, B. C., & Kendall, P. C. (2009). Therapist responsiveness to child
engagement: Flexibility within manual-based CBT for anxious youth.
Journal of Clinical Psychology, 65, 736–754.
Costello, E. J., Egger, H. L., & Angold, A. (2004). Developmental epide-
miology of anxiety disorders. In T. H. Ollendick & J. S. March (Eds.),
Phobic and anxiety disorders in children and adolescents (pp. 61–91).
New York, NY: Oxford University Press.
Eifert, G. H., Evans, I. M., & McKendrick, V. G. (1990). Matching
treatments to client problems not diagnostic labels: A case for paradig-
matic behavior therapy. Journal of Behavior Therapy & Experimental
Psychiatry, 21, 163–172.
Emslie, G. J. (2008). Pediatric anxiety: Underrecognized and undertreated.
New England Journal of Medicine, 359, 26.
Goisman, R. M., Warshaw, M. G., & Keller, M. B. (1999). Psychosocial
treatment prescriptions for generalized anxiety disorder, panic disorder,
and social phobia, 1991–1996. American Journal of Psychiatry, 156,
Greist, J. H. (2008). A promising debut for computerized therapies [Edi-
torial]. American Journal of Psychiatry, 165, 793–795.
Griffiths, K. M., & Christensen, H. (2006). Review of randomized con-
trolled trials of Internet interventions for mental health disorders and
related conditions. Clinical Psychologist, 10, 16–29.
Grills, A. E., & Ollendick, T. H. (2002). Issues in parent–child agreement:
The case of structured diagnostic interviews. Clinical Child and Family
Psychology Review, 5, 57–83.
Kazdin, A., French, N., Unis, A., Esveldt-Dawson, K., & Sherick, R.
(1983). Hopelessness, depression, and suicidal intent among inpatient
children. Journal of Consulting and Clinical Psychology, 51, 504–510.
Kendall, P. C., & Hedtke, K. (2006). Cognitive-behavioral therapy for
anxious children: Therapist manual (3rd ed.). Ardmore, PA: Workbook.
Available from http://workbookpublishing.com
Kendall, P. C., Hudson, J., Gosch, E., Flannery-Schroeder, E., & Suveg, C.
(2008). Cognitive-behavioral therapy for anxiety disordered youth: A
randomized clinical trial of child and family modalities. Journal of
Consulting and Clinical Psychology, 76, 282–297.
Kendall, P. C., & Khanna, M. S. (2008). Camp Cope-A-Lot: The Coping
Cat DVD [DVD]. Ardmore, PA: Workbook. Available from http://
Khanna, M. S., & Kendall, P. C. (2008). Computer assisted CBT for child
anxiety: Development of the Coping Cat CD-ROM. Cognitive and
Behavioral Practice, 15, 159–165.
Kovacs, M. (1981). Rating scales to assess depression in school aged
children. Acta Paedopsychiatrica, 46, 305–315.
Larsen, D. L., Attkisson, C. C., Hargreaves, W. A., & Nguyen, T. D.
(1979). Assessment of client/patient satisfaction: Development of a
general scale. Evaluation and Program Planning, 4, 197–207.
March, J. S. (2009). The future of psychotherapy for mentally ill children
and adolescents. Journal of Child Psychology and Psychiatry, 50, 170–
March, J. S., & Albano, A. M. (1998). Advances in the assessment of
pediatric anxiety disorders. Advances in Clinical Child Psychology, 20,
March, J. S., Parker, J., Sullivan, K., Stallings, P., & Conners, C. (1997). Download full-text
The Multidimensional Anxiety Scale for Children (MASC): Factor
structure, reliability and validity. Journal of the American Academy of
Child and Adolescent Psychiatry, 36, 554–565.
March, S., Spence, S. H., & Donovan, C. L. (2008). The efficacy of an
Internet-based cognitive-behavioral therapy intervention for child anxi-
ety disorders. Journal of Pediatric Psychology, 34, 474–487.
Marks, I. M., Cavanagh, K., & Gega, L. (2007). Hands-on help: Computer-
aided psychotherapy. London, England: Psychology Press.
McCrone, P., Knapp, M., Proudfoot, J., Ryden, C., Cavanagh, K., Shapiro,
D., . . . Tylee, A. (2004). Cost-effectiveness of computerized cognitive-
behavioral therapy for anxiety and depression in primary care: Random-
ized controlled trial. British Journal of Psychiatry, 185, 55–62.
National Institute for Clinical Excellence. (2006). Computerized cognitive
behaviour therapy for depression and anxiety [Technology Appraisal
97]. Retrieved from http://www.nice.org.uk/nicemedia/live/11568/
Newman, M. G. (2000). Recommendations for a cost-offset model of
psychotherapy allocation using generalized anxiety disorder as an ex-
ample. Journal of Consulting and Clinical Psychology, 68, 549–555.
Nguyen, T. D., Attkisson, C. C., & Stegner, B. (1983). Assessment of
patient satisfaction: Development and refinement of a service evaluation
questionnaire. Evaluation and Program Planning, 6, 299–314.
Schoenwald, S. K., & Hoagwood, K. (2001). Effectiveness and dissemi-
nation research: Their mutual roles in improving mental health services
for children and adolescents. Emotional & Behavioral Disorders in
Youth, 3, 3–4, 18–20.
Shaffer, D., Gould, M., Brasic, J., Ambrosini, P., Fisher, P., Bird, &
Aluwahlia, S. (1983). A Children’s Global Assessment Scale. Archives
of General Psychiatry, 40, 1228–1231.
Silverman, W., & Albano, A. M. (1997). The Anxiety Disorders Interview
Schedule for Children (DSM–IV). San Antonio, TX: Psychological Corp.
Silverman, W., & Eisen, A. (1992). Age differences in the reliability of
parent and child reports of child anxious symptomatology using a
structured interview. Journal of the American Academy of Child and
Adolescent Psychiatry, 31, 117–124.
Silverman, W., Kurtines, W. M., Ginsburg, G. S., Weems, C. F., Lumpkin,
P., White, C., & Hicks, D. (1999). Treating anxiety disorders in children
with group cognitive-behavioral therapy: A randomized clinical trial.
Journal of Consulting and Clinical Psychology, 67, 995–1003.
Spek, V., Cuijpers, P., NyklAicek, I., Riper, H., Keyzer, H., & Pop, V.
(2007). Internet-based cognitive-behaviour therapy for symptoms of
depression and anxiety: A meta-analysis. Psychological Medicine, 37,
Spence, S. H., Holmes, J., March, S., & Lipp, O. (2006). The feasibility and
outcome of clinic plus Internet delivery of cognitive-behavioral therapy
for childhood anxiety. Journal of Consulting and Clinical Psychology,
Walkup, J. T., Albano, A. M., Piacentini, J., Birmaher, B., Compton, S. N.,
Sherrill, J. T., . . . Kendall, P. C. (2008). Cognitive behavioral therapy,
sertraline, or a combination in childhood anxiety. New England Journal
of Medicine, 359, 1–14.
Weisz, J. R., Hawley, K., & Jensen-Doss, A. (2004). Empirically tested
psychotherapies for youth internalizing and externalizing problems and
disorders. Child and Adolescent Psychiatric Clinics of North America,
Wright, J. H., Wright, A. S., Albano, A. M., Basco, M. R., Goldsmith, J.,
Raffield, T., & Otto, M. W. (2005). Computer-assisted cognitive therapy
for depression: Maintaining efficacy while reducing therapist time. The
American Journal of Psychiatry, 162, 1158–1164.
Received June 17, 2009
Revision received January 25, 2010
Accepted March 23, 2010 ?