Content uploaded by Maree J. Abbott
Author content
All content in this area was uploaded by Maree J. Abbott
Content may be subject to copyright.
Bibliotherapy for Children With Anxiety Disorders Using Written
Materials for Parents: A Randomized Controlled Trial
Ronald M. Rapee, Maree J. Abbott, and Heidi J. Lyneham
Macquarie University
The current trial examined the value of modifying empirically validated treatment for childhood anxiety
for application via written materials for parents of anxious children. Two hundred sixty-seven clinically
anxious children ages 6 –12 years and their parents were randomly allocated to standard group treatment,
waitlist, or a bibliotherapy version of treatment for childhood anxiety. In general, parent bibliotherapy
demonstrated benefit for children relative to waitlist but was not as efficacious as standard group
treatment. Relative to waitlist, use of written materials for parents with no therapist contact resulted in
around 15% more children being free of an anxiety disorder diagnosis after 12 and 24 weeks. These
results have implications for the dissemination and efficient delivery of empirically validated treatment
for childhood anxiety.
Keywords: child anxiety, anxiety disorders, childhood, treatment, self-help
Anxiety disorders in childhood have begun to be recognized as
a serious mental health problem. Epidemiological studies have
repeatedly shown anxiety disorders in childhood to be one of the
most prevalent forms of mental disorder (Anderson, Williams,
McGee, & Silva, 1987; Canino et al., 2004). Combined with this
high prevalence, the moderate to severe life impact of anxiety
(Strauss, Frame, & Forehand, 1987) means that childhood anxiety
provides a serious burden for public health. Furthermore, several
studies have shown that anxiety disorders in childhood provide
risk markers for a range of later adolescent and adult psychopa-
thology (Last, Perrin, Hersen, & Kazdin, 1996; Pine, Cohen,
Gurley, Brook, & Ma, 1998; Weissman et al., 1999).
On the positive side, the past decade has seen an increased focus
on treatment for broad-based childhood anxiety disorders, and
treatment effects have been positive (Allen & Rapee, 2005). These
positive treatment results for childhood anxiety have generally
been achieved through a traditional therapist–client treatment
model typically involving 10–20 face-to-face therapeutic sessions.
Although this model is highly efficacious, it is also resource
intensive. This means that therapeutic costs are high and that many
treatment settings have long waiting lists.
One of the most widely researched alternatives to traditional
therapy is self-help. Individuals can help themselves with the
assistance of video materials, audiotapes, or computerized pro-
grams, but by far the most widely developed and researched
delivery is via printed materials (bibliotherapy). Among adults,
bibliotherapy has been applied to the successful management of
depression, eating disorders, and a variety of anxiety disorders
(Loeb, Wilson, Gilbert, & Labouvie, 2000; McKendree-Smith,
Floyd, & Scogin, 2003; Newman, Erickson, Preworski, & Dzus,
2003). In fact, self-help through bibliotherapy in adults has shown
good outcomes, in many cases as efficacious as standard therapist-
conducted treatment (Mains & Scogin, 2003). One of the obvious
difficulties for self-help lies in the maintenance of motivation to
change throughout a lengthy program. As a result, some biblio-
therapy programs rely on at least minimal involvement from a
therapist, and some evidence has demonstrated that the degree of
improvement in anxiety disorders is related to the amount of
therapist involvement (Marrs, 1995).
One potential characteristic of childhood disorders that may
make them especially amenable to bibliotherapy is the fact that
children live under the care and guidance of an adult, most com-
monly a parent. In this situation, bibliotherapy programs for chil-
dren can make use of a possible motivating factor that exists in the
child’s daily environment by utilizing the parent’s desire for his or
her child to change. The potential advantages of targeting parents
in bibliotherapy for children’s problems are that (a) parents have
personal knowledge of the child across several areas of function-
ing, (b) children usually have established trust and rapport with
their parents, (c) parents have some degree of distance from the
child’s distress, and (d) parents are more broadly present and
available in a child’s life than is a therapist. On the other hand,
conducting therapy for a child through his or her parents can be
filled with dangers: (a) The parents may in fact not be very distant
from the problems and in some cases may be part of the cause
(J. L. Hudson & Rapee, 2004), (b) the parent– child relationship
may be distressed and could undermine trust and rapport, and (c)
parents typically lack the professional education and knowledge of
a therapist. Thus, whether or not parents are capable of helping
Ronald M. Rapee, Maree J. Abbott, and Heidi J. Lyneham, Department of
Psychology, Macquarie University, Sydney, New South Wales, Australia.
Maree J. Abbott is now at the Department of Psychology, University of
Sydney, Sydney, New South Wales, Australia.
We thank Alan Taylor for help with statistical analyses. We also thank
Leigh Carpenter, Joanne McDonald, Sarah Perini, and Lexine Stapinski for
assistance with project management and Jonathan Gaston for clinical
supervision. This research was supported by a grant from the Australian
Rotary Health Research Fund.
Correspondence concerning this article should be addressed to Ronald
M. Rapee, Department of Psychology, Macquarie University, Sydney
NSW 2109, Australia. E-mail: ron.rapee@mq.edu.au
Journal of Consulting and Clinical Psychology Copyright 2006 by the American Psychological Association
2006, Vol. 74, No. 3, 436 – 444 0022-006X/06/$12.00 DOI: 10.1037/0022-006X.74.3.436
436
their children overcome emotional disorders via written materials
is an empirical question.
Despite the promise of bibliotherapy for adult mental disorders
and the potential advantages for childhood problems, there has
been surprisingly little research using this method with children.
Some work has indicated the value of written instructions for
parents for childhood oppositional behavior (Connell, Sanders, &
Markie-Dadds, 1997; Long, Rickert, & Ashcraft, 1993), headache
(Griffiths & Martin, 1996), enuresis (van Londen, van Londen-
Barentsen, van Son, & Mulder, 1995), and social skills in intel-
lectually delayed children (A. M. Hudson et al., 2003). To date, no
studies have examined the value of parent-delivered material for
children’s internalizing problems such as anxiety. This is surpris-
ing given the high prevalence and impact of these disorders
(Canino et al., 2004; Strauss et al., 1987).
The aim of the current study was to examine the impact of using
parents as therapists for their own child in a trial of bibliotherapy
materials for parents of children with anxiety disorders. Being the
first such study, its aim was simply to address the question of
whether bibliotherapy for childhood anxiety was more efficacious
than no treatment and as efficacious as standard therapist-led
treatment. We focused on prepubertal children because of previous
indications that parent involvement in child anxiety treatment is
especially valuable for younger children (Barrett, Dadds, & Rapee,
1996). Furthermore, given the limited theoretical differences be-
tween childhood anxiety disorders, the high levels of comorbidity
between disorders, and the fact that many treatment trials for
childhood anxiety combine several anxiety disorders, this study
included a variety of children’s anxiety disorders using the same,
adaptable program. Finally, we were aware of a serious limitation
to external validity in many self-help trials for adult problems that
include additional motivators such as reminder telephone calls,
regular mail contact, or homework mailings. Therefore, to maxi-
mize the real-world relevance of the trial, we chose to have no
contact between the researchers and self-help participants during
the active treatment phase, thereby providing a very conservative
test of self-help.
Method
Participants
Participants for the study were 267 children meeting Diagnostic and
Statistical Manual of Mental Disorders (4th ed. [DSM–IV]; American
Psychiatric Association, 1994) criteria for an anxiety disorder. Because
demonstration of differences between active treatments requires good
power, the intention was to recruit the maximum sample size possible, with
a minimum of 53 participants per condition, to allow a power of .8 to detect
a maximum effect-size difference of f ⫽ .25 (Faul & Erdfelder, 1992).
Participants were included if they were in Years 1 through 6 at school (ages
6 –12 years), they met criteria for an anxiety disorder as their principal
(most interfering) disorder, and their parent or parents were able to read a
standard, English-language newspaper. To maximize external validity,
children with comorbid nonanxiety disorders were not excluded unless
these disorders demanded immediate attention (e.g., severe school nonat-
tendance, suicidal risk). Children on medication were included if the
medication had been stable for the previous month.
Diagnoses were assigned by graduate students in clinical psychology or
qualified clinical psychologists following structured interview with the
Anxiety Disorders Interview Schedule for DSM–IV, Parent and Child
Versions (ADIS-CP; Silverman & Albano, 1996). Interviewers received
training to criterion, and research from our clinic overlapping with the
current sample has demonstrated interrater agreement of
⫽ 1.00 for an
overall diagnosis of anxiety disorder and ranging from .68 to .93 across the
major anxiety disorders (Lyneham, Abbott, & Rapee, 2004). Children met
criteria for the following principal diagnoses: generalized anxiety disorder
(N ⫽ 103), social phobia (N ⫽ 64), separation anxiety disorder (N ⫽ 51),
specific phobia (N ⫽ 33), obsessive– compulsive disorder (N ⫽ 13), and
panic disorder (N ⫽ 3). The main comorbid diagnostic groups included
anxiety disorder (N ⫽ 219; 82.0%), externalizing disorder (N ⫽ 72;
27.0%), and mood disorder (N ⫽ 23; 8.6%).
Measures
Structured interview. As described above, all children and their parents
were interviewed by a clinician using the ADIS-CP. The interview was
repeated at posttreatment and 3-month follow-up. Repeated interviews
were conducted by clinicians who were masked to the child’s allocated
treatment condition but who were told the child’s pretreatment diagnoses.
This was done to ensure that clinicians completed a measure of diagnostic
severity on each diagnosis even if the child no longer met criteria. Thus, the
interview provided data on diagnostic status as well as clinician-rated
severity.
Child reports. To measure symptoms of anxiety, children completed
the Spence Children’s Anxiety Scale (SCAS; Spence, 1998). This self-
report measure contains 38 anxiety items that all load on a single higher
order scale, with a range from 0 to 114. Internal consistency (.92) and
6-month retest reliability (.60) for the total scale are good (Spence, 1998).
Finally, children completed the Children’s Automatic Thoughts Scale
(CATS; Schniering & Rapee, 2002), a measure of children’s negative
thoughts and beliefs. The measure contains four subscales: social threat
(
␣
⫽ .85), physical threat (
␣
⫽ .92), failure and loss (
␣
⫽ .92), and
hostility (
␣
⫽ .85). The measure also has good retest reliability over 3
months (.68–.77), and the various subscales each discriminate between
relevant forms of child psychopathology (Schniering & Rapee, 2002). For
the current study, total scores were used to provide a measure of general
negative thinking. This measure has a total of 40 items with a range from
0 to 160.
Parent reports. Parents completed the parent version of the SCAS
(SCASp; Nauta, Scholing, Rapee, Abbott, & Spence, 2004). This measure
contains items parallel to those of the child version but relates to questions
about “my child” rather than self. Scores range from 0 to 114, and internal
consistency (.89) is good (Nauta et al., 2004). The measure distinguishes
clinically anxious from nonclinical groups of children (Ms 31.8 and 14.2,
respectively; Nauta et al., 2004). Data from the mother and father SCASp,
when available, were combined to form a single measure of parent-rated
anxiety symptoms. When two parents did not provide data, results are
based on a single parent, most often the mother.
To assess related and comorbid symptoms, parents also completed the
Child Behavior Checklist (CBCL; Achenbach, 1991), which was scored
simply as internalizing (CBCL-int) and externalizing (CBCL-ext) sub-
scales. Internal consistency for the two subscales is reported to be strong
(.89 –.93), as is 1-week retest reliability (.87–.95), and the scales have been
shown to differentiate referred and nonreferred children (Achenbach,
1991). Again, when available, the two parent scores on the CBCL were
combined to form single measures of parent internalizing and parent
externalizing, and t scores are reported.
Treatment Conditions
Group treatment. Group treatment was based on the Cool Kids Pro-
gram, a nine-session cognitive– behavioral program for the management of
broad-based childhood anxiety disorders (Rapee & Wignall, 2002). Previ-
ous research has shown the program to be efficacious and to provide results
comparable with other empirically validated programs for the management
437
BIBLIOTHERAPY FOR CHILDHOOD ANXIETY
of child anxiety (Rapee, 2000, 2003). Therefore, it provided a valuable
gold standard against which to compare bibliotherapy. Parents and children
attend all nine sessions of the program on a weekly basis over 12 weeks
(the final few sessions are biweekly) and cover recognition of emotion and
anxiety, realistic thinking, child management strategies, exposure to feared
cues, and additional skills such as assertiveness and dealing with teasing.
Each session lasts for approximately 2 hours and is conducted in groups of
around seven families. The program is manualized, and both child and
parents receive written summaries, worksheets, and guides for home prac-
tice during sessions. In the present study, groups were conducted by pairs
of therapists who were mostly graduate students in clinical psychology,
with at least one having had previous experience conducting Cool Kids
groups.
Bibliotherapy. Parents of participants allocated to bibliotherapy were
informed that they were to conduct treatment at home with their own child
using self-help materials. Each parent was provided with a copy of the
commercially available consumer book Helping Your Anxious Child: A
Step-by-Step Guide (Rapee, Spence, Cobham, & Wignall, 2000), which
describes anxiety management skills and ways of introducing them to and
implementing them with children. In addition, children were provided with
a workbook that contained the summaries and worksheets referred to in the
parent book and used within the group program. The materials for biblio-
therapy were developed to contain the strategies and information that
characterize most empirically supported treatments for child anxiety.
Hence, the techniques covered in bibliotherapy more or less paralleled
those covered in the standard group program. Nevertheless, we were
working with a commercially available product, and hence, the bibliother-
apy program was not completely parallel to the Cool Kids group program.
Examples of some differences are a section on relaxation in the commercial
book that was not included in the group program and attention to dealing
with teasing and bullying in the group program that was not included in the
commercial book.
Parents were given a cover letter explaining that they were to work
through the materials with their child and would be seen again for assess-
ment in 3 months’ time. The letter informed parents that they were to work
through the program at their own pace but provided them with a suggested
timetable for completing the program within the 3 months. To evaluate
pure bibliotherapy, no further contact with a therapist or the researchers
was to be initiated. If a parent initiated contact during the treatment period,
he or she was encouraged by research staff to continue to implement the
skills and was reminded of the timing of the next assessment. If a parent
reported a severe deterioration in his or her child (e.g., school refusal,
suicidal ideation), research staff referred the case to the unit’s clinical
supervisor to determine the most appropriate course of action.
Waitlist. Participants in waitlist were simply told that they had been
randomly assigned to wait for treatment and that they would be recontacted
for additional assessment in 3 months’ time, after which they would be
offered the next available treatment group.
Procedure
Recruitment for the study began in November 1999 and ended in August
2002. Final follow-up concluded in early 2003. Participant flow is shown
in Figure 1. Participants contacted the Macquarie University Anxiety
Research Unit (Sydney, New South Wales, Australia) following referral
from school counselors, general practitioners, or mental health profession-
als or by word of mouth and were screened with a brief telephone
conversation to determine whether the stated problem appeared reasonably
appropriate. They were then sent the questionnaire measures to complete at
home and return at the pretreatment structured diagnostic interview. Those
who met criteria for inclusion were allocated to a research condition and
provided the appropriate materials, scheduled into a group, or informed
when their waitlist period would end. Randomization occurred in blocks of
eight to allow allocation to group treatment based on a predetermined
random number schedule known only to the study coordinator (Maree J.
Abbott). After the initial 12 weeks, all children were once again given the
initial questionnaires and were reassessed using the ADIS-CP by inter-
viewers who were masked to group allocation. No adverse events due to
participation in treatment were noted. Participants in waitlist were then
offered group treatment. Participants in bibliotherapy and group treatments
were asked to continue practicing the skills for a further 3 months, at which
time they were again reassessed. After the 3-month follow-up period,
bibliotherapy participants were offered group treatment if they felt they
needed it.
Parents signed informed-consent forms, and children provided verbal
assent. The methods were approved by the Macquarie University Human
Ethics Committee.
Data Analysis
Data from all three conditions were available only at pre- and posttreat-
ment points because, for ethical reasons, waitlist participants were offered
group treatment after the end of the waiting period. Thus, the three
experimental groups were compared at pre- and posttreatment using
mixed-model analyses. Two types of analyses were conducted. Treatment
completer analyses included only those participants who returned data at
both time points and, in the group condition, completed at least seven
sessions. An equivalent to the seven-session rule for people in bibliother-
apy was not possible, and therefore, all bibliotherapy participants who
returned data were included in treatment completer analyses. Hence, results
for this condition are necessarily somewhat conservative. Intention-to-treat
analyses included all participants who were allocated to a condition (aside
from 7 participants who did not return any data at pretreatment) and used
267 Children with DSM-IV
anxiety disorders
Group Treatment
N = 90
Bibliotherapy
N = 90
Waitlist
N = 87
12-week
(post)
Assessment
N = 76
12-week
(post)
Assessment
N = 61
12-week
(post)
Assessment
N = 75
Dropout
N = 14
Dropout
N = 29
Dropout
N = 2
Dropout
N = 0
Referred for
Active
Treatment
24-week
(follow-up)
Assessment
N = 74
24-week
(follow-up)
Assessment
N = 61
Dropout
N = 12
Random Allocation
Figure 1. Flow diagram of participants through the study. DSM-IV ⫽
Diagnostic and Statistical Manual of Mental Disorders (4th ed.).
438
RAPEE, ABBOTT, AND LYNEHAM
the last-point-carried-forward method to deal with missing data. Both
methods of analyzing data address somewhat different questions.
In addition, the two active conditions, bibliotherapy and group treatment,
were compared across the three time points for which they were expected
to return data: pretreatment, posttreatment, and 3-month follow-up. Mixed-
model analyses were used, and again, both completer and intention-to-treat
analyses were conducted.
Results
Pretreatment Comparisons
Of the 267 children, 90 were allocated to group treatment, 90 to
bibliotherapy, and 87 to waitlist. Participants allocated to the three
experimental groups were compared on both demographic vari-
ables and pretreatment measures of psychopathology. On demo-
graphic variables, there were no significant differences between
groups on child’s age, parent marital status, percentage on a low
family income, number of siblings, or use of medication (all ps ⬎
.10). Child’s sex differed significantly between groups,
2
(2, N ⫽
267) ⫽ 11.13, p ⬍ .01, with the group treatment condition having
a greater proportion of female children. Therefore, all analyses
described below were repeated with sex included as a covariate. In
no case was sex a significant covariate, and hence, it is not
described further. Demographic data are presented in Table 1. On
measures of psychopathology, there was no significant difference
between groups on their principal diagnosis, severity of principal
diagnosis, number of comorbid diagnoses, or scores on the SCAS,
CATS, SCASp, CBCL-int, or CBCL-ext (all ps ⬎ .10; see
Table 2).
Analyses of Dropout Data
A total of 55 participants (20.6%) failed to return posttreatment
data or attended fewer than seven group treatment sessions.
Among these participants, 12 (13.8%) were from waitlist, 29
(32.2%) were from bibliotherapy, and 14 (15.6%) were from group
treatment. There was a significant difference between conditions in
the proportion of those who did not return posttreatment data,
2
(2,
N ⫽ 267) ⫽ 11.30, p ⬍ .01.
Participants who dropped out at posttreatment were compared
with those who did not drop out on demographic and psychopa-
thology measures. Those who dropped out had a significantly
greater number of comorbid diagnoses (M dropout ⫽ 2.2, SD ⫽
1.3; M nondropout ⫽ 1.8, SD ⫽ 1.3), F(1, 265) ⫽ 4.71, p ⬍ .05,
and scored significantly higher on several measures of psychopa-
thology: CATS (M dropout ⫽ 48.9, SD ⫽ 33.8; M nondropout ⫽
34.3, SD ⫽ 26.7), F(1, 254) ⫽ 11.18, p ⫽ .001; CBCL-ext (M
dropout ⫽ 57.8, SD ⫽ 9.1; M nondropout ⫽ 54.4, SD ⫽ 9.7), F(1,
262) ⫽ 5.29, p ⬍ .05; and SCAS (M dropout ⫽ 39.6, SD ⫽ 18.6;
M nondropout ⫽ 31.8, SD ⫽ 18.2), F(1, 246) ⫽ 7.64, p ⬍ .01.
Several other measures did not differ significantly between groups,
including child’s age, child’s sex, child’s medication use, parents’
marital status, number of siblings, SCASp, and CBCL-int.
Among participants in the bibliotherapy and group treatments,
45 (25.0%) failed to return any data at 3-month follow-up. Among
these participants, 29 (32.2%) were from bibliotherapy, and 16
(17.7%) were from group treatment,
2
(1, N ⫽ 180) ⫽ 5.01, p ⬍
.05.
Pre–Post Comparisons Between Conditions
Changes across time from pre- to posttreatment (completion of
first 12 weeks) were compared between the three conditions using
repeated measures mixed-model analyses. As described above,
both completer and intention-to-treat analyses were conducted.
The proportion of children in each condition who no longer met
criteria for any DSM–IV anxiety disorder is shown in Figures 2 and
3. There were significant differences between groups based on
both the completer sample,
2
(2, N ⫽ 212) ⫽ 51.79, p ⬍ .001, and
the intention-to-treat sample,
2
(2, N ⫽ 267) ⫽ 47.88, p ⬍ .001.
Post hoc analyses indicated that for the completer sample, biblio-
therapy resulted in significantly more children free of an anxiety
disorder (25.9%) than for waitlist (6.7%),
2
(1, N ⫽ 136) ⫽ 8.62,
p ⬍ .005, but less than for children in group treatment (61.1%),
2
(1, N ⫽ 137) ⫽ 15.31, p ⬍ .001. A similar pattern was apparent
for the intention-to-treat sample: bibliotherapy (17.8%) versus
waitlist (5.7%),
2
(1, N ⫽ 177) ⫽ 6.12, p ⬍ .05; bibliotherapy
versus group treatment (48.9%),
2
(1, N ⫽ 180) ⫽ 19.6, p ⬍ .001.
The clinician-rated global severity scale for the principal anxiety
disorder diagnosis also showed a significant Condition ⫻ Time
interaction based on both completer, F(2, 208.8) ⫽ 67.14, p ⬍
.001, partial
2
⫽ .372, and intention-to-treat samples, F(2, 264) ⫽
44.77, p ⬍ .001, partial
2
⫽ .253. Comparisons of the change
across time for each condition using the intention-to-treat sample
showed that children in bibliotherapy improved more than those on
waitlist, t(264) ⫽ 2.14, p ⬍ .05, but not as much as those in group
treatment, t(264) ⫽ 6.95, p ⬍ .001. Children in group treatment
improved significantly more than those on waitlist, t(264) ⫽ 9.03,
p ⬍ .001. Similar patterns were shown in the completer sample:
bibliotherapy versus waitlist, t(208.1) ⫽ 3.54, p ⬍ .001; biblio-
therapy versus group, t(210.3) ⫽ 7.19, p ⬍ .001; group versus
waitlist, t(208) ⫽ 11.41, p ⬍ .001. Data are presented in Tables 2
and 3.
Parent report of child anxiety (SCASp) demonstrated a signifi-
cant difference between groups based on both intention-to-treat,
F(2, 262.6) ⫽ 8.20, p ⫽ .001, partial
2
⫽ .058, and completer
samples, F(2, 179.3) ⫽ 6.75, p ⫽ .001, partial
2
⫽ .072. Com
-
parisons of change across time for each condition showed slightly
different patterns based on analysis sample. For intention-to-treat
analyses, bibliotherapy failed to differ significantly from waitlist,
t(262.2) ⫽ 0.98, p ⬎ .32, but was significantly different from
group treatment, t(262.8) ⫽ 2.93, p ⬍ .01. Group treatment also
Table 1
Demographic Data Across the Three Conditions (SDs in
Parentheses)
Demographic
Waitlist
(N ⫽ 87)
Bibliotherapy
(N ⫽ 90)
Group
treatment
(N ⫽ 90)
Child age (months) 114.1 (19.1) 114.7 (18.1) 113.7 (20.4)
Child sex (% female) 29.9 35.6 53.3
Marital status (% married) 85.1 87.8 86.7
Low family income
(% below $30,000) 16.3 10.3 17.0
Using medication (%) 17.2 16.7 24.4
Number of siblings 1.4 (0.9) 1.4 (0.7) 1.4 (0.9)
Number of comorbid
diagnoses 1.9 (1.2) 1.9 (1.4) 1.9 (1.4)
439
BIBLIOTHERAPY FOR CHILDHOOD ANXIETY
showed a greater decrease across time than waitlist, t(262.8) ⫽
3.89, p ⬍ .001. In contrast, completer analyses indicated that
bibliotherapy showed a significantly greater decrease over time
than waitlist, t(180.7) ⫽ 2.02, p ⬍ .05, but did not differ signifi-
cantly from group treatment, t(179.9) ⫽ 1.36, p ⬎ .17. Group
treatment again showed a significantly greater decrease across
time than waitlist, t(177.5) ⫽ 3.66, p ⬍ .001.
Parent report of internalizing behavior (CBCL-int) showed a
significant Condition ⫻ Time interaction based on both intention-
to-treat, F(2, 262.1) ⫽ 9.41, p ⬍ .001, partial
2
⫽ .066, and
completer samples, F(2, 188.6) ⫽ 7.75, p ⫽ .001, partial
2
⫽
.084. Comparisons across time for each condition showed slightly
different patterns based on analysis sample. Intention-to-treat anal-
yses showed that children in bibliotherapy failed to improve sig-
nificantly more than those on waitlist, t(261.4) ⫽ 0.53, p ⫽ .60,
and improved significantly less than children in group treatment,
t(262.5) ⫽ 3.49, p ⫽ .001. Children in group treatment improved
significantly more than those on waitlist, t(262.5) ⫽ 3.98, p ⬍
.001. In contrast, completer analyses indicated that bibliotherapy
showed a significantly greater decrease over time than waitlist,
t(189.7) ⫽ 2.08, p ⬍ .05, but did not differ significantly from
group treatment, t(189.2) ⫽ 1.55, p ⬎ .12. Group treatment again
showed a significantly greater decrease across time than waitlist,
t(187.2) ⫽ 3.93, p ⬍ .001.
Parent report of externalizing behavior (CBCL-ext) showed a
significant Condition ⫻ Time interaction based on both intention-
to-treat, F(2, 261.7) ⫽ 6.03, p ⬍ .01, partial
2
⫽ .043, and
completer samples, F(2, 183.4) ⫽ 5.56, p ⫽ .005, partial
2
⫽
.060. Comparisons across time for each condition showed slightly
different patterns based on analysis sample. Intention-to-treat anal-
yses showed that children in bibliotherapy failed to improve sig-
nificantly more than those on waitlist, t(261.3) ⫽ 1.26, p ⬎ .20,
and improved significantly less than children in group treatment,
t(261.9) ⫽ 2.20, p ⬍ .05. Children in group treatment improved
significantly more than those on waitlist, t(261.9) ⫽ 3.43, p ⫽
.001. In contrast, completer analyses indicated that children in
bibliotherapy showed a significantly greater decrease over time
than those on waitlist, t(184.1) ⫽ 2.25, p ⬍ .05, but did not differ
significantly from children in group treatment, t(183.8) ⫽ 0.72,
p ⬎ .47. Group treatment again showed a significantly greater
decrease across time than waitlist, t(182.4) ⫽ 3.23, p ⫽ .001.
Children’s reports of anxious symptoms (SCAS) failed to show
a significant Condition ⫻ Time interaction based on either
intention-to-treat, F(2, 243.3) ⫽ 0.15, p ⬎ .86, partial
2
⬍ .01, or
completer samples, F(2, 179.3) ⫽ 0.82, p ⬎ .44, partial
2
⫽ .011.
As seen in Tables 2 and 3, children reported a marked decrease in
anxiety in all three conditions, including waitlist.
Finally, children’s reports of negative beliefs (CATS) also failed
to show a significant Condition ⫻ Time interaction based on either
the intention-to-treat sample, F(2, 249.4) ⫽ 0.15, p ⬎ .85, partial
2
⬍ .01, or the completer sample, F(2, 187.8) ⫽ 0.96, p ⬎ .38,
partial
2
⬍ .01.
Maintenance Effects for Active Treatment Conditions
The proportion of participants in each condition who no longer
met criteria for any anxiety disorder differed significantly at
3-month follow-up based on both intention-to-treat (bibliother-
apy ⫽ 18.9%; group ⫽ 61.1%),
2
(1, N ⫽ 180) ⫽ 33.43, p ⬍ .001,
and completer samples (bibliotherapy ⫽ 17.9%; group ⫽ 73.6%),
2
(1, N ⫽ 128) ⫽ 39.18, p ⬍ .001 (see Figures 2 and 3).
Participants in group treatment and bibliotherapy were com-
pared across time from preintervention to 3-month follow-up using
mixed-model analyses. Data are presented in Tables 2 and 3.
Waitlist Bibliothera
py
Grou
p
treatment
0
10
20
30
40
50
60
70
80
%
Posttreatment
Follow-up
Figure 2. Percentage (including standard error) of children in each con-
dition who no longer met criteria for any anxiety disorder at posttreatment
and 3-month follow-up based on completer analyses.
Table 2
Mean Pretreatment, Posttreatment, and Follow-Up Data Across the Three Conditions for All Participants (Intention to Treat) With
Last Data Carried Forward (SDs in Parentheses)
Measure
Waitlist Bibliotherapy Group treatment
Pre Post Pre Post Follow-up Pre Post Follow-up
Diagnostic severity 6.5 (0.9) 5.8 (1.6) 6.4 (1.0) 5.2 (1.8) 5.0 (1.8) 6.5 (1.0) 3.4 (2.6) 2.8 (2.5)
SCASp 30.1 (13.4) 27.7 (13.8) 31.1 (14.2) 27.2 (15.4) 25.9 (15.7) 32.0 (13.0) 23.7 (13.6) 22.3 (14.3)
CBCL-int 68.4 (7.7) 65.1 (8.8) 68.4 (8.3) 64.6 (10.3) 63.4 (11.0) 67.7 (8.3) 60.3 (9.7) 58.7 (11.3)
CBCL-ext 55.1 (9.5) 53.9 (10.5) 55.1 (9.7) 52.8 (10.5) 51.7 (10.6) 55.2 (9.8) 50.9 (9.3) 49.6 (10.3)
SCAS 33.2 (18.0) 25.5 (15.9) 34.2 (18.2) 28.1 (20.1) 25.4 (19.4) 32.9 (19.6) 25.6 (16.7) 23.8 (17.2)
CATS 36.9 (28.7) 26.8 (24.2) 35.7 (26.2) 18.9 (18.6) 18.6 (19.1) 39.5 (31.7) 25.0 (20.7) 19.0 (21.2)
Note. Pre ⫽ pretreatment; Post ⫽ posttreatment; SCASp ⫽ Spence Children’s Anxiety Scale, Parent Version; CBCL-int ⫽ Child Behavior Checklist,
internalizing scale; CBCL-ext ⫽ Child Behavior Checklist, externalizing scale; SCAS ⫽ Spence Children’s Anxiety Scale; CATS ⫽ Children’s Automatic
Thoughts Scale.
440
RAPEE, ABBOTT, AND LYNEHAM
Analyses based on both intention-to-treat and completer analyses
produced similar patterns, and hence, only intention-to-treat anal-
yses, being the more conservative, are reported.
Clinician rating of diagnostic severity showed a significant
decrease across time, F(2, 356) ⫽ 177.00, p ⬍ .001, partial
2
⫽
.499, and a significant Condition ⫻ Time interaction, F(2, 356) ⫽
36.77, p ⬍ .001, partial
2
⫽ .171. The two conditions differed
significantly in their change from pretreatment to follow-up,
t(356) ⫽ 8.03, p ⬍ .001, but not in their change from posttreatment
to follow-up, t(356) ⫽ 1.40, p ⬎ .16.
Parent rating of anxiety (SCASp) showed a similar significant
effect of time, F(2, 354.2) ⫽ 69.31, p ⬍ .001, partial
2
⫽ .281,
and a Condition ⫻ Time interaction, F(2, 354.2) ⫽ 6.97, p ⫽ .001,
partial
2
⫽ .038. Bibliotherapy and group treatment differed
significantly in their change from pretreatment to follow-up,
t(354.3) ⫽ 3.27, p ⫽ .001, but not in their change from posttreat-
ment to follow-up, t(354.1) ⫽ 0.06, p ⬎ .95.
Parent report of internalizing symptoms (CBCL-int) showed a
significant decrease across time, F(2, 353.5) ⫽ 90.68, p ⬍ .001,
partial
2
⫽ .338, and a Condition ⫻ Time interaction, F(2,
353.5) ⫽ 7.72, p ⫽ .001, partial
2
⫽ .041. The conditions differed
significantly in their change from pretreatment to follow-up,
t(353.6) ⫽ 3.55, p ⫽ .001, but not in their change from posttreat-
ment to follow-up, t(353.2) ⫽ 0.31, p ⬎ .75.
Parent report of externalizing symptoms (CBCL-ext) also
showed a significant effect of time, F(2, 353.3) ⫽ 59.68, p ⬍ .001,
partial
2
⫽ .252, and a significant Condition ⫻ Time interaction,
F(2, 353.3) ⫽ 3.54, p ⬍ .05, partial
2
⫽ .019. The conditions
differed significantly in their change from pretreatment to follow-
up, t(353.3) ⫽ 2.45, p ⬍ .05, but not in their change from
posttreatment to follow-up, t(353.1) ⫽ 0.32, p ⬎ .74.
Children’s report of anxiety symptoms (SCAS) showed a sig-
nificant effect of time, F(2, 322.7) ⫽ 42.77, p ⬍ .001, partial
2
⫽
.216, but failed to show a significant Condition ⫻ Time interac-
tion, F(2, 322.7) ⫽ 0.06, p ⬎ .94, partial
2
⬍ .01.
Finally, children’s report of negative beliefs (CATS) showed a
significant effect of time, F(2, 327.2) ⫽ 26.17, p ⬍ .001, partial
2
⫽ .138, but failed to show a significant Condition ⫻ Time
interaction, F(2, 327.2) ⫽ 0.15, p ⬎ .86, partial
2
⬍ .01.
Discussion
Overall, the results of the current trial demonstrate that children
whose parents received bibliotherapy with no therapist contact
improved somewhat more than children on waitlist after 12 weeks
and that these results maintained up to 3 months. However, the
specific pattern of results differs slightly depending on the type of
analysis and the reporting source. On the basis of structured
clinical interviews, bibliotherapy was significantly better than no
treatment according to both completer and intention-to-treat anal-
yses. In contrast, parent reports indicate a significant difference
between bibliotherapy and waitlist according to completer, but not
according to intention-to-treat, analyses. Finally, children in all
three groups reported significant and marked change over time, but
differences between groups were not significant. The results also
show that standard cognitive– behavioral group treatment with a
therapist resulted in greater change than bibliotherapy according to
both clinician and parent reports. Therefore, these results do not
suggest a replacement of traditional models of therapy but do
suggest a potential alternate model of treatment delivery under
appropriate circumstances.
The data indicate the usual disagreement between sources of
reporting (Cole, Truglio, & Peeke, 1997). The clearest differences
Waitlist Bibliotherapy Group treatment
0
10
20
30
40
50
60
70
%
Posttreatment
Follow-up
Figure 3. Percentage (including standard error) of children in each con-
dition who no longer met criteria for any anxiety disorder at posttreatment
and 3-month follow-up based on intention-to-treat analyses.
Table 3
Mean Pretreatment, Posttreatment, and Follow-Up Data Across the Three Conditions for Participants Who Completed Each
Assessment (SDs in Parentheses)
Measure
Waitlist Bibliotherapy Group treatment
Pre Post Pre Post Follow-up Pre Post Follow-up
Diagnostic severity 6.5 (0.9) 5.8 (1.7) 6.4 (1.1) 4.7 (1.8) 4.8 (1.8) 6.4 (0.9) 2.5 (2.2) 2.0 (2.1)
SCASp 30.6 (13.7) 26.6 (13.1) 29.6 (14.2) 21.0 (12.2) 16.1 (10.8) 30.8 (13.0) 20.2 (11.2) 16.9 (10.8)
CBCL-int 68.6 (7.4) 64.3 (8.8) 68.2 (8.8) 60.6 (10.8) 56.7 (12.1) 67.2 (8.6) 58.3 (9.3) 54.8 (10.9)
CBCL-ext 54.7 (9.1) 52.7 (10.1) 54.3 (10.4) 48.9 (9.9) 46.5 (10.6) 54.3 (9.7) 48.8 (8.2) 45.9 (9.2)
SCAS 32.6 (18.4) 21.9 (14.7) 32.1 (16.6) 19.1 (15.0) 15.5 (11.6) 30.7 (19.5) 21.4 (14.5) 17.6 (14.5)
CATS 34.5 (26.8) 27.1 (24.3) 33.6 (24.6) 18.9 (18.6) 18.6 (19.1) 34.8 (28.5) 25.0 (20.7) 19.0 (21.2)
Note. Pre ⫽ pretreatment; Post ⫽ posttreatment; SCASp ⫽ Spence Children’s Anxiety Scale, Parent Version; CBCL-int ⫽ Child Behavior Checklist,
internalizing scale; CBCL-ext ⫽ Child Behavior Checklist, externalizing scale; SCAS ⫽ Spence Children’s Anxiety Scale; CATS ⫽ Children’s Automatic
Thoughts Scale.
441
BIBLIOTHERAPY FOR CHILDHOOD ANXIETY
between conditions are reflected in the ratings made by therapists
and based on interview with parents and child. Although every
attempt was made to keep interviewers masked to treatment con-
dition, in reality this is quite difficult, and it is possible that
interview ratings were influenced by interviewers’ preconceptions
of outcome. Nevertheless, the advantages of interview ratings
include the potential to combine all relevant information, to adhere
to clear diagnostic criteria, to incorporate information about con-
text and interference, and to provide a relative benchmark based on
experience with a variety of cases. Overall, the results based on
interview criteria indicate that relative to the passage of time,
around 15% more children were free of an anxiety disorder simply
through the use of psychotherapeutic materials written for their
parents.
In general, the pattern of results from interview was reflected for
the most part in parent questionnaire reports. However, the differ-
ence between bibliotherapy and waitlist failed to reach statistical
significance based on intent-to-treat analyses whereby participants
who failed to return data had their previous data carried forward to
the next point. This is an especially conservative method of miss-
ing data imputation. However, in the absence of widely agreed
methods for handling missing data (Tabachnick & Fidell, 2001),
we decided to utilize this conservative approach. The different
pattern of significance between intent-to-treat and completer anal-
yses suggests that although bibliotherapy appears to be of value for
those who implement it completely, it may not be of significant
value across a general population to which it is applied. In other
words, bibliotherapy for anxious children may be efficacious when
used appropriately, but it may not be effective because of limits in
people’s ability or willingness to implement it. This point is of
relevance to all modes of treatment but may be especially relevant
in the case of self-help strategies such as bibliotherapy where
factors such as understanding, correct implementation, and moti-
vation become particularly salient because of the lack of a thera-
pist. Combining self-help with reduced amounts of therapist input
may be one way in which to maximize the strengths of each
approach (Marrs, 1995; Newman et al., 2003). In addition, future
research into identification of factors that influence the implemen-
tation and success of self-help will be vital to translating efficacy
into effectiveness.
Surprisingly, self-report data from children did not indicate
differences between conditions but showed a marked reduction in
symptoms across all groups, including waitlist. Childhood anxiety
is not generally considered to be a transient condition, and previ-
ous treatment outcome studies have usually indicated relatively
small changes across time (Barrett et al., 1996; Kendall, 1994).
Therefore, the reasons for the greater change across time in child
report relative to other sources of information in this study are
unclear. Some authors have reported the tendency of anxious
children to “fake good” (Kendall, 1993), and it is likely that they
also become bored with repeatedly completing the same measures.
Most self-report measures are not validated at the lower age groups
included in this trial. Consequently, the reports from young chil-
dren such as those used in this study need to be considered with
some care. However, children are the ultimate customer, and
hence, their perceptions of change need to be considered in any
evaluation of outcome. Clearly, the children in this trial reported
marked improvement across the 24 weeks, but whether these
effects were due to intervention or simply to the passage of time
remains uncertain.
On the negative side, bibliotherapy resulted in greater dropout
from participation than did traditional group therapy or waiting for
treatment. This finding highlights an important caveat to the use of
bibliotherapy. Clearly bibliotherapy, or perhaps simply the concept
of bibliotherapy, is not suitable for some people. Patients seeking
psychotherapy often expect guidance and advice from an expert
who assumes some degree of responsibility and control (Furnham
& Wardley, 1990). Thus, individuals with such expectations who
are then confronted with an independent and active model of
therapeutic change may be disheartened. It is possible that an
alternate recruitment strategy, self-selection, or a more active
marketing of the positive side of bibliotherapy could attract a
greater proportion of individuals who are suited to this treatment
model. In the current study, we found that treatment dropouts had
slightly more severe symptomatology than completers. Further
post hoc examination of this issue demonstrated that this greater
severity was not specific to those who dropped out of bibliotherapy
but characterized dropouts across all conditions. Future research
aimed at identifying sufferers who can derive the greatest benefit
from self-help would be of tremendous value for the streamlining
and cost reduction of mental health resources (Baillie & Rapee,
2004).
Several additional limitations to the current study should be
considered. First, the structured interviews were not technically
administered in exactly the way prescribed by Silverman and
Albano (1996). Although the interview instructions require inter-
viewers to make separate diagnoses based on parent report and
child report and then to combine these diagnoses, our interviewers
created combined diagnoses based on their separate interviews
with parents and children without the intervening step of produc-
ing actual diagnoses based on each separate interview. It is ex-
tremely unlikely that this would have systematically influenced the
results, but this possibility cannot be excluded. More important to
note is that the lack of separate diagnoses meant that there was no
independent validation for the self-reports from parents and chil-
dren. Given the lack of significant differences shown on child
report, it may be argued that bibliotherapy produced changes only
in parent perceptions rather than in actual anxiety. However, this is
unlikely given that (a) child report in fact indicated a marked
reduction in anxiety, but not significantly more than the surprising
decrease in anxiety reported by waitlist children, and (b) an iden-
tical pattern of results was shown in the group treatment. An
additional limitation is the fact that parents did not complete data
on compliance or preference for bibliotherapy. Such data would be
important to more fully understanding the implications and bene-
fits of self-help and should be included in any future studies.
Perhaps the main limitation to the study is the fact that the sample
for the study came from a traditional, specialist anxiety clinic. This
was necessary to allow a properly controlled scientific design and
a group treatment comparison. However, this recruitment means
that we cannot be certain whether those families who do not seek
traditional forms of therapy would benefit from bibliotherapy.
Clearly, it would be useful to specifically target these families in
future research.
Given the suggestion that bibliotherapy for child anxiety ap-
pears to be of value at least for those families who successfully
implement it, some important clinical implications can be high-
442
RAPEE, ABBOTT, AND LYNEHAM
lighted. Most obviously, written materials could be used to reduce
the waiting lists of busy therapeutic services. In particular, if it was
made clear to clients that bibliotherapy materials were being
offered as an interim measure while waiting for an available
therapist, the current results suggest that up to 20% would no
longer require the services of a therapist 12 weeks later. Although
not directly addressed in the current study, it is also possible that
using bibliotherapy materials might allow a reduction in the
amount of required therapist time and hence allow a stretching of
limited professional resources. Similarly, other research has sug-
gested that augmenting bibliotherapy for childhood anxiety with
alternate methods of therapist contact (such as telephone and
Internet) can provide treatment to populations whose physical
location makes reaching traditional therapeutic services impossible
(Lyneham & Rapee, 2005).
The current study also raises important issues for public health.
Epidemiological surveys have shown that only a small proportion
of children in need reach appropriate sources of help (Canino et al.,
2004; Farmer, Stangl, Burns, Costello, & Angold, 1999). Per-
ceived stigma and a desire to handle problems themselves have
been shown to characterize anxious adults who do not seek help
(Issakidis & Andrews, 2002), and it is likely that similar attitudes
are held by parents of anxious children. The traditional treatment
model is also limited by a lack of physical and professional
resources available to implement the empirically supported treat-
ments (Kendall & Choudhury, 2003; Miller et al., 2003). Conse-
quently, if it were possible to increase the proportion of sufferers
who seek help, there would not exist sufficient resources in the
mental health sector to provide the necessary assistance.
Bibliotherapy has the potential to address all of these issues. The
promising findings reported here provide the first step toward
exploring these broader public health implications by showing that
bibliotherapy has the potential to reduce childhood anxiety.
Clearly, future studies need to focus on methods to attract those
anxious children whose families are reluctant to seek traditional
treatment and to determine whether bibliotherapy can provide an
effective means of reducing the impact of anxiety for these chil-
dren while minimizing stigma and allowing greater flexibility.
Increased public education campaigns or training for professionals
who work with children can help to increase the recognition of
anxiety and its effects, as well as methods for managing it. By
including bibliotherapeutic strategies among the treatment options,
it may be possible to attract a greater proportion of families who
would usually avoid traditional therapeutic services.
References
Achenbach, T. M. (1991). Manual for the Teacher’s Report Form and
1991 Profile. Burlington: University of Vermont, Department of
Psychiatry.
Allen, J. L., & Rapee, R. M. (2005). Anxiety disorders. In P. Graham (Ed.),
Cognitive behaviour therapy for children and families (2nd. ed., pp.
300 –319). Cambridge, England: Cambridge University Press.
American Psychiatric Association. (1994). Diagnostic and statistical man-
ual of mental disorders (4th ed.). Washington, DC: Author.
Anderson, J. C., Williams, S., McGee, R., & Silva, P. A. (1987). DSM–III
disorders in preadolescent children. Prevalence in a large sample from a
general population. Archives of General Psychiatry, 44, 69 –76.
Baillie, A. J., & Rapee, R. M. (2004). Predicting who benefits from
psychoeducation and self help for panic attacks. Behaviour Research
and Therapy, 42, 513–527.
Barrett, P. M., Dadds, M. R., & Rapee, R. M. (1996). Family treatment of
childhood anxiety: A controlled trial. Journal of Consulting and Clinical
Psychology, 64, 333–342.
Canino, G., Shrout, P. E., Rubio-Stipec, M., Bird, H. R., Bravo, M., &
Ramirez, R. (2004). The DSM–IV rates of child and adolescent disorders
in Puerto Rico. Archives of General Psychiatry, 61, 85–93.
Cole, D. A., Truglio, R., & Peeke, L. (1997). Relation between symptoms
of anxiety and depression in children: A multitrait–multimethod–
multigroup assessment. Journal of Consulting and Clinical Psychology,
65, 110 –119.
Connell, S., Sanders, M. R., & Markie-Dadds, C. (1997). Self-directed
behavioral family intervention for parents of oppositional children in
rural and remote areas. Behavior Modification, 21, 379 – 408.
Farmer, E. M., Stangl, D. K., Burns, B. J., Costello, E. J., & Angold, A.
(1999). Use, persistence, and intensity: Patterns of care for children’s
mental health across one year. Community Mental Health Journal, 35,
31– 46.
Faul, F., & Erdfelder, E. (1992). GPOWER: A priori, post-hoc, and
compromise power analyses for MS-DOS [Computer program]. Bonn,
Germany: Bonn University.
Furnham, A., & Wardley, Z. (1990). Lay theories of psychotherapy: I.
Attitudes toward, and beliefs about, psychotherapy and therapists. Jour-
nal of Clinical Psychology, 46, 878 – 890.
Griffiths, J. D., & Martin, P. R. (1996). Clinical- versus home-based
treatment formats for children with chronic headache. British Journal of
Health Psychology, 1, 151–166.
Hudson, A. M., Matthews, J. M., Gavidia-Payne, S. T., Cameron, C. A.,
Mildon, R. L., Radler, G. A., et al. (2003). Evaluation of an intervention
system for parents of children with intellectual disability and challenging
behaviour. Journal of Intellectual Disability Research, 47, 238 –249.
Hudson, J. L., & Rapee, R. M. (2004). From anxious temperament to
disorder: An etiological model of generalized anxiety disorder. In R. G.
Heimberg, C. L. Turk, & D. S. Mennin (Eds.), Generalized anxiety
disorder: Advances in research and practice (pp. 51–76). New York:
Guilford Press.
Issakidis, C., & Andrews, G. (2002). Service utilisation for anxiety in an
Australian community sample. Social Psychiatry and Psychiatric Epi-
demiology, 37, 153–163.
Kendall, P. C. (1993). Cognitive–behavioral therapies with youth: Guiding
theory, current status, and emerging developments. Journal of Consult-
ing and Clinical Psychology, 61, 235–247.
Kendall, P. C. (1994). Treating anxiety disorders in children: Results of a
randomized clinical trial. Journal of Consulting and Clinical Psychol-
ogy, 62, 100 –110.
Kendall, P. C., & Choudhury, M. S. (2003). Children and adolescents in
cognitive– behavioral therapy: Some past efforts and current advances,
and the challenges in our future. Cognitive Therapy and Research, 27,
89 –104.
Last, C. G., Perrin, S., Hersen, M., & Kazdin, A. E. (1996). A prospective
study of childhood anxiety disorders. Journal of the American Academy
of Child and Adolescent Psychiatry, 35, 1502–1510.
Loeb, K. L., Wilson, G., Gilbert, J. S., & Labouvie, E. (2000). Guided and
unguided self-help for binge eating. Behaviour Research and Therapy,
38, 259 –272.
Long, N., Rickert, V. I., & Ashcraft, E. W. (1993). Bibliotherapy as an
adjunct to stimulant medication in the treatment of attention-deficit
hyperactivity disorder. Journal of Pediatric Health Care, 7, 82– 88.
Lyneham, H., Abbott, M., & Rapee, R. M. (2004). [Interrater agreement on
childhood anxiety disorders using structured interview]. Unpublished
data.
Lyneham, H., & Rapee, R. M. (2005). Evaluation of therapist supported
parent-implemented CBT for anxiety disorders in rural children. Manu-
script submitted for publication.
Mains, J. A., & Scogin, F. R. (2003). The effectiveness of self-
443
BIBLIOTHERAPY FOR CHILDHOOD ANXIETY
administered treatments: A practice-friendly review of the research.
Journal of Clinical Psychology, 59, 237–246.
Marrs, R. W. (1995). A meta-analysis of bibliotherapy studies. American
Journal of Community Psychology, 23, 843– 870.
McKendree-Smith, N. L., Floyd, M., & Scogin, F. R. (2003). Self-
administered treatments for depression: A review. Journal of Clinical
Psychology, 59, 275–288.
Miller, T. W., Miller, J. M., Kraus, R. F., Kaak, O., Sprang, R., &
Veltkamp, L. J. (2003). Telehealth: A clinical application model for rural
consultation. Consulting Psychology Journal: Practice and Research,
55, 119 –127.
Nauta, M. H., Scholing, A., Rapee, R. M., Abbott, M., & Spence, S. H.
(2004). Development of 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, 813– 839.
Newman, M. G., Erickson, T., Preworski, A., & Dzus, E. (2003). Self-help
and minimal-contact therapies for anxiety disorders: Is human contact
necessary for therapeutic efficacy? Journal of Clinical Psychology, 59,
251–274.
Pine, D. S., Cohen, P., Gurley, D., Brook, J., & Ma, Y. (1998). The risk for
early-adulthood anxiety and depressive disorders in adolescents with
anxiety and depressive disorders. Archives of General Psychiatry, 55,
56 – 64.
Rapee, R. M. (2000). Group treatment of children with anxiety disorders:
Outcome and predictors of treatment response. Australian Journal of
Psychology, 52, 125–129.
Rapee, R. M. (2003). The influence of comorbidity on treatment outcome
for children and adolescents with anxiety disorders. Behaviour Research
and Therapy, 41, 105–112.
Rapee, R. M., Spence, S. H., Cobham, V. E., & Wignall, A. (2000).
Helping your anxious child: A step-by-step guide for parents. Oakland,
CA: New Harbinger.
Rapee, R. M., & Wignall, A. (2002). Cool Kids Program: Family Version
session guide. Sydney, Australia: Macquarie University Anxiety Re-
search Unit.
Schniering, C. A., & Rapee, R. M. (2002). Development and validation of
a measure of children’s automatic thoughts: The Children’s Automatic
Thoughts Scale. Behaviour Research and Therapy, 40, 1091–1109.
Silverman, W. K., & Albano, A. M. (1996). The Anxiety Disorders Inter-
view Schedule for Children–IV (Child and Parent Versions). San Anto-
nio: TX: Psychological Corporation.
Spence, S. H. (1998). A measure of anxiety symptoms among children.
Behaviour Research and Therapy, 36, 545–566.
Strauss, C. C., Frame, C. L., & Forehand, R. (1987). Psychosocial impair-
ment associated with anxiety in children. Journal of Clinical Child
Psychology, 16, 235–239.
Tabachnick, B. G., & Fidell, L. S. (2001). Using multivariate statistics (4th
ed.). Boston: Allyn & Bacon.
van Londen, A., van Londen-Barentsen, M. W., van Son, M. J., & Mulder,
G. A. (1995). Relapse rate and subsequent parental reaction after suc-
cessful treatment of children suffering from nocturnal enuresis: A 2
1
⁄
2
year follow-up of bibliotherapy. Behaviour Research and Therapy, 33,
309 –311.
Weissman, M. M., Wolk, S., Wickramaratne, P., Goldstein, R. B., Adams,
P., Greenwald, S., et al. (1999). Children with prepubertal-onset major
depressive disorder and anxiety grown up. Archives of General Psychi-
atry, 56, 794 – 801.
Received January 31, 2005
Revision received October 20, 2005
Accepted December 7, 2005 䡲
444
RAPEE, ABBOTT, AND LYNEHAM
