Child Psychiatry Hum Dev
DOI 10.1007/s10578-013-0410-x
ORIGINAL ARTICLE
Somatic Complaints in Anxious Youth
Sarah A. Crawley • Nicole E. Caporino • Boris Birmaher • Golda Ginsburg
John Piacentini • Anne Marie Albano • Joel Sherrill • Dara Sakolsky •
Scott N. Compton • Moira Rynn • James McCracken • Elizabeth Gosch •
Courtney Keeton • John March • John T. Walkup • Philip C. Kendall
•
Ó Springer Science+Business Media New York 2013
Abstract This study examined (a) demographic and clinical
characteristics associated with physical symptoms in anxietydisordered youth and (b) the impact of cognitive-behavioral
therapy (Coping Cat), medication (sertraline), their combination, and pill placebo on physical symptoms. Youth
(N = 488, ages 7–17 years) with a principal diagnosis of
generalized anxiety disorder, separation anxiety disorder, or
social phobia participated as part of a multi-site, randomized
controlled trial and received treatment delivered over
12 weeks. Diagnostic status, symptom severity, and impairment were assessed at baseline and week 12. The total number
and severity of physical symptoms was associated with age,
principal diagnosis, anxiety severity, impairment, and the
presence of comorbid internalizing disorders. Common
somatic complaints were headaches, stomachaches, head cold
or sniffles, sleeplessness, and feeling drowsy or too sleepy.
Physical symptoms decreased over the course of treatment,
S. A. Crawley N. E. Caporino P. C. Kendall (&)
Department of Psychology, Temple University, Weiss Hall, 1701
North 13th Street, Philadelphia, PA 19122, USA
e-mail: pkendall@temple.edu
B. Birmaher D. Sakolsky
Western Psychiatric Institute and Clinic, University of Pittsburgh
Medical Center, Pittsburgh, PA, USA
G. Ginsburg C. Keeton
Division of Child and Adolescent Psychiatry, The Johns Hopkins
University School of Medicine, Baltimore, MD, USA
J. Piacentini J. McCracken
Semel Institute for Neuroscience and Human Behavior,
University of California Los Angeles, Los Angeles, CA, USA
and were unrelated to treatment condition. Clinical implications and directions for future research are discussed (ClinicalTrials.gov number, NCT00052078).
Keywords
Treatment
Children Adolescents Anxiety
Introduction
Anxiety disorders occur in approximately 10 % of youth
[1] and are associated with impairment in family, social,
and academic functioning [2, 3]. If left untreated, anxiety
disorders run a chronic course [4–6] and are ‘‘gateway’’
disorders associated with an increased risk for depression,
substance use problems, and educational underachievement
[7, 8].
J. Sherrill
Division of Services and Intervention Research, National
Institute of Mental Health, Bethesda, MD, USA
S. N. Compton J. March
Department of Psychiatry and Behavioral Services, Duke
University Medical Center, Durham, NC, USA
E. Gosch
Department of Psychology, Philadelphia College of Osteopathic
Medicine, Philadelphia, PA, USA
J. T. Walkup
Division of Child and Adolescent Psychiatry, Weill Cornell
Medical College, New York, NY, USA
A. M. Albano M. Rynn
Department of Psychiatry, Columbia University Medical Center,
New York, NY, USA
123
Child Psychiatry Hum Dev
Somatic complaints are common among anxiety-disordered youth, with more than 50 % reporting at least one
somatic complaint [9–11]. These complaints include a
range of physical symptoms, such as headaches, stomachaches, muscle tension/pain, difficulty breathing, shaking,
pounding or racing heart, sweating, blushing, and fatigue.
Physical symptoms play a major role in the classification of
anxiety disorders. For example, somatic complaints are
required for the diagnosis of panic disorder, generalized
anxiety disorder (GAD), and posttraumatic stress disorder,
and are among the criteria possible for a diagnosis of
separation anxiety disorder (SAD) and acute stress disorder
[12]. Somatic complaints are not specific to one anxiety
disorder and are common in other mental health disorders
with diagnostic criteria that do not include somatic complaints (e.g., major depressive disorder) [11].
Research comparing the rate of somatic complaints
across disorders has yielded mixed results. Hofflich et al.
[11] found that somatic symptoms are equally common in
youth with principal diagnoses of GAD, SAD, and social
phobia (SoP) even though somatic symptoms are required
for a diagnosis of GAD only. In contrast, Ginsburg et al.
[10] found that somatic symptoms were more common in
youth with GAD as compared to SoP and SAD. In community samples, somatic complaints are more common
among adolescents than younger children [13]. Age differences in somatic complaints have been found in anxious
youth as well. For example, Ginsburg et al. [10] reported
that older youth (12–17 years) were more likely than
younger youth (6–11 years) with anxiety disorders to
endorse somatic complaints. Although at least one study
did not find age differences in somatic symptoms, this
could have been due to a relatively restricted age range
(7–14 years) [11].
Somatic complaints have been found to be more
common among girls than boys in community samples
[13]. However, with the exception of a study that sampled African American adolescents exclusively [14],
research has not supported sex differences in the overall
number of physical symptoms reported by youth with
anxiety disorders [10, 11, 15]. Two studies reported sex
differences in the rates with which specific physical
symptoms were endorsed [10, 16]. Somatic complaints
among youth with anxiety disorders have not been
related to race or family income [10, 11, 15]. However,
the reported findings were from relatively small samples
with limited diversity.
Somatic symptoms in youth with anxiety disorders may
be associated with greater psychopathology. Relative to
anxiety-disordered children without somatic complaints,
children with anxiety disorders and somatic complaints
have more severe anxiety and poorer global functioning
[10, 16]. They perform less well academically and are more
123
likely to refuse school [15, 17]. Also, anxiety-disordered
youth with somatic symptoms are more likely than their
counterparts without somatic symptoms to have comorbid
externalizing disorders [11] and depressive symptoms [11,
18]. Finally, among anxiety-disordered adolescents,
somatic symptoms are negatively associated with perceived competence across multiple domains [14].
Despite these relationships, few studies have examined
change in the number of somatic complaints following
anxiety treatment. Ginsburg et al. [10] reported decreased
somatic complaints following fluvoxamine treatment and
Storch et al. [16] reported reduced somatic symptoms
following CBT for obsessive–compulsive disorder (OCD).
Masia-Warner et al. [19] reported reductions in physical
symptoms following a CBT protocol modified to target
anxiety-related somatic complaints. To date, no studies
have compared the effect of CBT and medication, alone
and in combination, on the number or severity of physical
symptoms.
The current study examined somatic complaints in
children and adolescents who received CBT (Coping Cat),
medication (sertraline; SRT), their combination (CBT ?
SRT), or pill placebo [20]. We examined baseline levels of
somatic complaints in relation to demographic variables
(i.e., age; sex; race/ethnicity; socioeconomic status) and
presenting characteristics (i.e., principal diagnosis;
anxiety severity; functional impairment; comorbidities1),
and we examined treatment-related change in somatic
complaints and associations with improvement in anxiety. The number of general somatic complaints was
reported by youth using a symptom checklist whereas
the overall severity of anxiety-related somatic complaints was reported by parents and youth, and rated by
clinicians based on discussion with parents and children
together.
Method
Participants
Participants were 488 youth enrolled in the Child/Adolescent Anxiety Multimodal Study (CAMS [20]), a six-site
randomized controlled trial that compared the efficacy
of CBT, SRT, their combination, and pill placebo. Participants ranged in age from 7 to 17 years (M = 10.7,
SD = 2.8 years) and met criteria for a principal diagnosis
1
Youth with comorbid, secondary depression were initially included
in CAMS. However a decision was made to exclude youth with major
depressive disorder (MDD) early in the trial. Participants with
comorbid, secondary dysthymia were included throughout the trial.
Child Psychiatry Hum Dev
of SAD, GAD, and/or SoP based on composite rating
from the Anxiety Disorders Interview Schedule for DSMIV—Child and Parent Versions (ADIS-IV-C/P [21]).
Participants responded to announcements in local media
and from clinics, schools, primary care offices, mental
health centers, churches/temples, and community organizations. Youth with comorbid psychiatric disorders were
included in the trial so long as they were of lesser
severity than the target disorder and there was no active
suicidality (for methodology details, see Compton et al.
[22]).
Approximately 54 % of participants were male. The
sample was predominantly White (79 %), with 9 % of
participants identifying race as Black, 3 % as Asian, 1 %
as American Indian, \1 % as Native Hawaiian/Pacific
Islander, and 8 % as Other. With regard to ethnicity,
12 % of participants self-identified as Hispanic or Latino.
The majority of participants (75 %) were of middle to
high socioeconomic status, as indicated by a score of
40–66 on the Hollingshead Four Factor Index of Social
Status [23].2
Among exclusion criteria were the presence of an
unstable medical condition, IQ of less than 80, current
school refusal, and failure to respond to two adequate trials
of selective serotonin-reuptake inhibitors (SSRIs) or a trial
of CBT. Youth were also excluded if their psychiatric
condition made participation clinically inappropriate.
Clinical characteristics of the sample have been reported in
detail by Kendall et al. [24]. The flow of participants
through the current study is identical to that reported by
Walkup et al. [20].
Physical Symptoms Checklist (PSC)
The number of physical symptoms was assessed using the
PSC [27], a 46-item self-report measure of the extent to
which youth are bothered by general health problems over
the past week. Items are rated on a Likert scale ranging
from 0 (not at all) to 3 (very much); for this study, item
responses were dichotomized to indicate the presence (1, 2,
3) or absence (0) of each symptom and summed. Two
gynecological items/symptoms were omitted. In this sample, Cronbach’s alpha ranged from .89 to .91.
Pediatric Anxiety Rating Scale (PARS)
The PARS [28] is a clinician-rated measure of the presence
and severity of 51 anxiety symptoms in youth. Total scores
represent the severity and frequency of anxiety symptoms
as well as associated distress, avoidance, and interference
during the previous week. For the current study, severity of
physical symptoms of anxiety was assessed using an item
from the PARS as rated by the parent, child, and clinician.
The PARS has acceptable reliability and validity [28, 29].
Inter-rater reliability in CAMS was determined based on a
review of 10 % of videotaped assessments conducted at
pre- and post-treatment (Pearson’s r = .85).
Children’s Global Assessment Scale (CGAS)
The CGAS [30] is a clinician-rated measure of a child’s
global functioning. The CGAS demonstrates high retest
and inter-rater reliability, and discriminates between inpatients and outpatients [30, 31].
Measures
Clinical Global Impression Scales (CGI)
Anxiety Disorders Interview Schedule for DSM-IV-Child
and Parent Versions (ADIS-IV-C/P)
The ADIS-IV-C/P [21] is a semi-structured interview to
diagnose anxiety disorders and common comorbidities in
youth. For each disorder, a clinical severity rating (CSR)
is assigned using an 8-point scale (with CSRs C 4 indicating a clinical diagnosis). The disorder with the highest
CSR is identified as principal. The ADIS-IV-C/P has
solid psychometric properties [25, 26]. Based on a
review of 10 % of videotaped pre- and post-treatment
assessments, interrater reliability for diagnostic status
(intraclass correlation coefficient) in CAMS ranged from
.82 to .88.
2
Site differences in demographic variables have been discussed by
Walkup et al. [20].
The CGI [32] scales provide clinician ratings of global
severity (CGI-S) and improvement (CGI-I). The CGI-S
ranges from 1 (not at all ill) to 7 (extremely ill). The CGI-I
ranges from 1 (very much improved) to 7 (very much
worse), with scores of 1 (very much improved) or 2 (much
improved) used to designate treatment response. In this
study, the CGI-S and CGI-I ratings reflected severity and
impairment associated with GAD, SAD, and SoP.
Procedure
All procedures were approved by the site institutional
review board and the NIMH data and safety monitoring
board. CAMS used a three-gate entry procedure to ensure
a stable anxiety diagnosis (see [20, 22]). Informed consent
was obtained from all participants. As detailed in
Compton et al. [22], independent evaluators (IEs) were
123
Child Psychiatry Hum Dev
trained to reliability in the administration of clinicianrated measures. The ADIS-C/P and CGI-S were used to
determine diagnoses and anxiety severity. The PARS was
used to assess overall severity of anxiety-related physical
symptoms, as rated by parents, youth, and IEs. Number of
physical symptoms was reported by youth using the PSC,
which was completed approximately 1 week before the
start of treatment. Youth were randomly assigned to one
of the four treatment conditions: CBT, SRT, combination,
or placebo. Assessments were repeated within approximately 5 days of the last session. IEs were blind to
treatment condition.
Treatment
Cognitive-Behavioral Therapy
Youth randomized to CBT received 14, 60-min sessions
delivered over a 12-week period. CBT followed the Coping
Cat program [33, 34], which teaches youth to recognize
and manage anxious arousal. The Coping Cat was adapted
for the study: six sessions focused on anxiety-management
skills and eight sessions on exposure tasks. There were 12
individual child sessions and two parent sessions (without
the child present). Adolescents received the C.A.T. Project
[35], the teen version of the Coping Cat program. CBT was
delivered by experienced therapists who were certified in
the treatment protocols and received regular onsite and
cross-site supervision (see [22, 36]).
Medication
Youth randomized to medication received sertraline (SRT)
and eight, 30- to 60-min medication management sessions.
Psychiatrists provided education and guidance in addition
to symptom review and medication monitoring. Medication
was administered on a fixed-flexible schedule beginning
with 25 mg of sertraline per day and titrating up to 200 mg
by week 8. Participants were eligible for dose increases
through week 8 if they continued to be at least mildly ill
and experienced minimal side effects. Pill counts and
medication diaries were used to enhance and document
adherence to the medication regimen.
Combination
Youth randomized to the combination treatment received
both the 14-session CBT protocol and the SRT protocols.
Pharmacotherapy and CBT appointments, when possible,
were scheduled for the same day and location. Communication between CBT therapists and psychiatrists was
facilitated by weekly CAMS meetings.
123
Placebo
Participants assigned to pill placebo were treated with the
medication protocol (same as youth assigned to SRT).
Psychiatrists were blind to treatment condition.
Data Analytic Plan
Analyses were conducted using participants who completed all measures of interest, as well as with the intentto-treat sample. An analysis of variance (ANOVA) was
conducted to rule out pretreatment differences in somatic
complaints by treatment condition. Correlational analyses
and t tests assessed demographic differences in somatic
complaints; based on results, age was covaried in further
analyses. Analyses of covariance (ANCOVA) examined
differences in somatic complaints by principal diagnosis
and presence/absence of individual anxiety disorder
diagnoses (SAD, GAD, SoP). Correlations evaluated
relationships between physical symptoms and anxiety
severity, impairment, and treatment-related improvement.
Change in physical symptoms from pre- to post-treatment
was examined using a paired-samples t test. An
ANCOVA examined differences between treatment
responders and non-responders in posttreatment number
and severity of physical symptoms, controlling for pretreatment symptoms. ANCOVA assessed the relationship
between physical symptoms and the presence of comorbid
disorders (0, 1, or C2). An alpha level of .01 was used for
all analyses. Cohen’s d is reported as an index of effect
size.
Results
The frequencies with which physical symptoms were
endorsed at pretreatment are presented in Table 1. Almost
all participants (95 %) endorsed at least one physical
symptom on the PSC. The most common symptoms
reported by youth at pretreatment were headaches (50 %),
trouble sleeping (48 %), stomach pain or aches (47 %),
head cold or sniffles (40 %), restlessness or uncomfortable
urge to move (35 %), sleeplessness (34 %), feeling drowsy
or too sleepy (34 %), and nightmares or very strange
dreams (34 %). Means and standard deviations for the
number and severity of physical symptoms at pretreatment
are presented in Table 2. Child, parent, and clinician ratings of the severity of anxiety-related physical symptoms
were highly correlated (Pearson’s r C .80); given that the
clinician rating was based on parent and child report, it was
used for all subsequent analyses.
Child Psychiatry Hum Dev
Table 1 Frequency of physical symptoms by principal diagnosis at pretreatment
Pretreatment symptom
Principal diagnosis
SAD
(n = 16)
SoP
(n = 55)
GAD
(n = 33)
SAD SoP
(n = 33)
SAD GAD
(n = 39)
SoP GAD
(n = 137)
SAD SoP GAD
(n = 175)
Total
(N = 488)
Trouble sleeping (%)
37.5
32.5
45.5
44.7
60.4
48.3
50.3
48.0
Restless or uncomfortable
urge to move
37.5
17.5
27.3
23.7
35.4
33.1
45.3
35.1
Feeling drowsy or too
sleepy
25.0
55.0
39.4
31.6
47.9
58.1
50.6
33.6
Sleeplessness
12.5
22.5
36.4
27.0
29.2
32.0
42.2
33.6
6.2
15.0
36.4
21.6
37.5
35.1
40.4
33.5
50.0
40.0
25.0
34.2
41.7
39.2
44.7
40.4
Sleep disturbance
Nightmares or very
strange dreams
Upper respiratory
Head cold or sniffles
Allergies
6.2
17.5
42.4
23.7
31.2
35.6
37.9
33.0
Sore throat
12.5
12.5
30.3
28.9
35.4
23.6
30.0
26.5
Dry mouth
6.2
12.5
15.2
18.4
27.1
21.6
31.2
23.4
25.0
12.5
15.2
16.2
37.5
28.4
29.4
26.3
0.0
12.5
15.2
10.5
2.1
8.1
14.9
10.5
Coughing or wheezing
Fever
Pain
Headache
33.3
32.5
39.4
39.5
58.3
54.5
55.3
50.4
Stomach pain or ache
50.0
30.0
45.5
47.4
58.3
43.9
51.6
47.3
Muscle aches or cramps
25.0
25.0
18.2
18.9
25.0
30.4
27.3
26.5
Joint pain
12.5
12.8
15.2
7.9
22.9
16.2
14.9
15.3
Racing or pounding heart
18.8
22.5
18.2
13.2
33.3
26.4
23.0
23.8
Chest pain
Dizziness/faintness
12.5
6.2
17.5
17.5
12.1
12.1
15.8
7.9
22.9
12.5
20.3
21.5
23.6
54.5
20.2
19.0
Tremor, trembling, or
shakiness
12.5
17.5
9.1
8.1
25.0
18.9
22.4
18.8
Excessive sweating
6.2
20.0
9.1
18.9
16.7
13.5
17.4
15.5
Difficulty breathing
6.2
7.5
15.2
10.8
18.8
15.5
15.5
14.5
Heart skips beats
12.5
10.0
3.0
10.5
20.8
8.1
12.4
11.0
Numbness or tingling in
arms or legs
12.5
7.5
6.1
8.1
18.8
9.5
18.0
12.8
0.0
5.0
3.0
2.7
4.2
9.5
13.1
8.5
25.0
10.0
15.2
15.8
29.2
16.2
29.2
21.5
Feeling bloated or gassy
0.0
10.3
15.2
5.3
18.8
21.6
14.9
15.7
Constipation
0.0
2.5
9.1
5.3
14.6
13.2
11.5
14.3
Panic/cardiac
Can’t hear well
Elimination/gastric
Nausea/vomiting
Frequent urination
0.0
5.0
9.1
13.5
12.5
12.8
18.0
13.3
Diarrhea
6.2
7.5
17.2
7.9
12.5
10.1
15.5
12.2
Heartburn
0.0
5.0
6.1
7.9
8.3
12.2
11.2
9.7
Pain with urination
0.0
2.5
3.0
2.6
12.5
3.4
5.0
4.5
Skin
Dry skin
12.5
22.5
12.1
34.2
41.7
29.5
29.8
28.9
Acne
18.8
37.5
27.3
18.4
16.7
24.2
17.5
21.9
Skin rash
0.0
7.5
6.1
15.8
12.5
12.8
16.8
13.0
Hives
6.2
2.5
3.0
2.6
6.2
0.7
3.7
2.9
123
Child Psychiatry Hum Dev
Table 1 continued
Pretreatment symptom
Principal diagnosis
SAD
(n = 16)
SoP
(n = 55)
GAD
(n = 33)
SAD SoP
(n = 33)
SAD GAD
(n = 39)
SoP GAD
(n = 137)
SAD SoP GAD
(n = 175)
Total
(N = 488)
Feeling flushed or warm
(%)
18.8
17.5
21.2
10.5
25.0
27.5
32.9
26.2
Feeling cold or chilled
12.5
22.5
18.2
15.8
22.9
28.2
28.6
25.2
Agitation/disinhibition
18.8
17.9
25.8
5.3
29.2
24.5
25.8
23.2
0.0
15.0
24.2
7.9
16.7
18.8
23.6
18.8
Other
Ringing in ears
Easy bruising
6.2
7.5
15.2
21.1
16.7
13.6
19.4
15.8
10.0
9.1
0.0
22.2
13.3
15.7
3.6
10.6
Dental problems
0.0
5.0
9.1
10.8
10.4
10.1
9.4
9.1
Hair loss or brittle hair
0.0
7.5
3.0
0.0
4.2
5.4
36.4
4.6
Swelling, water retention
0.0
5.1
3.0
2.6
2.1
2.7
5.6
3.7
Cold or canker sores
Table 2 Pre- and post-treatment means and standard deviations for
number and severity of physical symptoms (N = 488)
Table 3 Pretreatment means and standard deviations for the number
and severity of physical symptoms by principal diagnosis
Pretreatment
Posttreatment
M
SD
M
SD
PSC number of physical symptoms
9.22
7.25
4.98
5.69
SAD
PARS clinician-rated severity
2.35
1.41
0.91
1.20
SoP
55
7.03
6.70
1.77
1.40
PARS child-reported severity
PARS parent-reported severity
2.21
2.29
1.50
1.48
0.82
0.82
1.19
1.21
GAD
SAD SoP
33
33
7.97
7.06
5.81
6.18
2.30
1.83
1.47
1.32
SAD GAD
39
10.46
7.50
2.46
1.22
GAD SoP
137
9.50
6.70
2.48
1.41
SAD SoP GAD
175
10.33
1.62
2.54
1.37
PSC Physical Symptoms Checklist, PARS Pediatric Anxiety Rating
Scale (0–5)
Demographic Differences in Physical Symptoms
The number of physical symptoms endorsed by youth at
pretreatment was significantly related to age (r = .17,
p \ .01), with younger children reporting fewer symptoms
than older children. Age was also positively associated
with the severity of physical symptoms, based on clinician
rating (r = .17, p \ .01). Controlling for age, the number
and severity of physical symptoms at pretreatment did not
differ by SES, sex, or race/ethnicity. At posttreatment, the
number of physical symptoms endorsed by youth was
positively associated with age (r = .17, p \ .01), as was
the severity of physical symptoms (r = .28, p \ .01). The
number and severity of physical symptoms at posttreatment
was not related to SES, sex, or race/ethnicity.
Differences in Physical Symptoms by Principal
Diagnosis
Table 1 presents the frequency of individual physical symptoms by principal diagnosis at pretreatment. When controlling
for age, clinician-rated severity of physical symptoms differed
significantly based on principal diagnosis, F(6, 479) = 4.19,
123
n
Number
M
16
5.47
Severity
SD
M
SD
4.16
1.73
1.62
Number of physical symptoms was assessed using the Physical
Symptoms Checklist
Severity of physical symptoms was rated by clinicians using the
Pediatric Anxiety Rating Scale
0 = no symptoms, 1 = minimal, 2 = mild, 3 = moderate, 4 =
severe, 5 = extreme
p \ .01, as did number of symptoms reported, F(6,
450) = 4.07, p \ .01. Means and standard deviations for the
number and severity of physical symptoms by principal diagnosis are in Table 3. Analyses were repeated to determine
whether having a specific diagnosis of GAD, SAD, or SoP
(regardless of whether it was principal) was associated with the
report of physical symptoms. Diagnostic group significantly
predicted the number of physical symptoms endorsed by youth,
F(4, 453) = 9.23, p \ .01. Follow-up analyses revealed that
GAD (t = -3.80, p \ .01, d = -.36) and SAD (t =
-3.09, p \ .01, d = -.29) were significantly associated with
the number of physical symptoms endorsed by youth while SoP
was not (t = -.26, p = .80). GAD (t = -4.40, p \ .01,
d = -.41) and SAD (t = -2.50, p = .01, d = -.23) were
also significantly associated with clinician-rated severity of
physical symptoms while SoP was not (t = -.43, p = .67).
Child Psychiatry Hum Dev
Table 4 Posttreatment means and standard deviations for number and severity of physical symptoms by treatment condition
COMBO (n = 140)
SRT (n = 133)
CBT (n = 139)
Placebo (n = 76)
F
p
PSC number of physical symptoms
5.16 (5.66)
5.04 (5.69)
4.35 (5.39)
5.31 (5.65)
0.30
0.83
PARS clinician-rated severity
0.75 (1.10)
0.82 (1.15)
1.04 (1.32)
0.98 (1.27)
2.05
0.11
PARS child-reported severity
0.68 (1.11)
0.72 (1.10)
0.96 (1.34)
0.68 (1.10)
4.65
0.18
PARS parent-reported severity
0.73 (1.34)
0.64 (1.12)
0.94 (1.33)
0.94 (1.35)
1.90
0.13
PSC Physical Symptoms Checklist, PARS Pediatric Anxiety Rating Scale (0–5)
Relationship of Physical Symptoms to Impairment
Global functioning, as measured using the CGAS, was
negatively associated with the pretreatment number of
physical symptoms endorsed by youth (r = -.13, p \ .01)
and clinician-rated severity of physical symptoms (r =
-.34, p \ .01).
Change in Physical Symptoms by Treatment Condition
At baseline, there were no differences by treatment condition in the number or severity of somatic complaints
when controlling for age. The number of physical symptoms reported by youth decreased significantly from pre- to
post-treatment, t(390) = 13.15, p \ .01, d = 1.33, as did
clinician-rated severity of physical symptoms, t(437) =
20.01, p \ .01, d = 1.91.
The number of physical symptoms reported by youth at
posttreatment was not related to treatment condition when
controlling for age and pretreatment number of physical
symptoms (Table 4). Planned contrasts did not reveal significant differences in the number and severity of physical
symptoms between participants who received medication
and those who did not (youth-reported number of symptoms, t(413) = -.18, p = .86; clinician-rated severity,
t(434) = 2.06, p = .04), and between participants who
received CBT and those who did not (youth-reported
number of physical symptoms, t(413) = -.35, p = .73;
clinician-rated severity, t(434) = 2.06, p = .56).3
Relationship of Physical Symptoms to Anxiety Severity
and Improvement
The pretreatment number of physical symptoms endorsed by
youth was positively associated with pretreatment anxiety
severity, as measured by the CGI-S (r = .19, p \ .01), but
was unrelated to posttreatment anxiety severity (r = .03,
p = .57) and improvement, as measured by the CGI-I
(r = .01, p = .96). Pretreatment severity of physical symptoms, as rated by clinicians, was positively associated with
3
Findings did not differ when analyses were conducted using the
intent-to-treat sample with last observation carried forward versus
participants with complete data at baseline and week 12.
anxiety severity at pretreatment (r = .46, p \ .01) and
posttreatment (r = .15, p \ .01), but was unrelated to
improvement in anxiety (r = .07, p = .14). Posttreatment
number of physical symptoms endorsed by youth was positively associated with pretreatment anxiety severity (r = .20,
p \ .01) and posttreatment anxiety severity (r = .24,
p \ .01), and was related to CGI-I ratings reflecting less
improvement in anxiety (r = .16, p \ .01). Posttreatment
severity of physical symptoms, as rated by clinicians, was
also positively associated with pretreatment anxiety severity
(r = .21, p \ .01) and posttreatment anxiety severity
(r = .53, p \ .01), and was related to CGI-I ratings reflecting
less improvement in anxiety (r = .48, p \ .01).
Analyses were conducted to determine if the posttreatment number and severity of physical symptoms were related to treatment response (i.e., CGI-I ratings of much
improved or very much improved). Controlling for baseline
symptoms, treatment responders reported fewer physical
symptoms, F(1, 414) = 6.89, p \ .01, d = .26, and less
severe clinician-rated physical symptoms, F(1, 435) =
77.58, p \ .01, d = .84, than non-responders.3
Relationship of Physical Symptoms to Presence
of Comorbid Disorders
Controlling for age, neither the number nor the severity of
physical symptoms reported by participants differed by the
presence/number of pretreatment comorbid diagnoses.
When internalizing and externalizing comorbidities were
examined separately, there were significant pretreatment
differences in the number and severity of physical symptoms based on the presence/number (0, 1, C2) of comorbid
internalizing disorders (youth-reported number of physical
symptoms, F(2, 454) = 4.54, p = .01; clinician-rated
severity of physical symptoms, F(2, 483) = 4.54, p =
.01). Planned contrasts revealed that participants without
comorbid internalizing disorders had fewer and less severe
physical symptoms than participants with one or more
comorbid internalizing disorders (youth-reported number
of physical symptoms, t(454) = 2.95, p \ .01, d = .28;
clinician-rated severity of physical symptoms, t(483) =
2.98, p \ .01, d = .27). There were no significant differences in the number or severity of physical symptoms
between participants with one comorbid internalizing
123
Child Psychiatry Hum Dev
disorder and those with two or more comorbid internalizing
disorders. At pretreatment, there were no significant differences between participants with/without comorbid
externalizing disorders on the number or severity of
physical symptoms endorsed.
Controlling for age, the number of physical symptoms
endorsed by youth at posttreatment differed based on the
presence/number of comorbid diagnoses, F(2, 413) =
4.96, p \ .01. Planned contrasts revealed that participants
with no comorbid disorders endorsed significantly fewer
physical symptoms at posttreatment than participants with
one or more disorders, t(413) = -2.87, p \ .01, d =
-.28. Participants with one comorbid disorder did not
differ from participants with two or more comorbid disorders, t(413) = -.24, p = .81, suggesting that the number
of physical symptoms is related to the presence but not the
number of comorbid disorders. Clinician-rated severity of
physical symptoms did not differ based on presence/number of comorbid diagnoses.
When comorbid internalizing and externalizing disorders were examined separately, the presence/number of
comorbid internalizing disorders (0, 1, C2) was significantly associated with clinician-rated severity of physical
symptoms, F(2, 434) = 8.05, p \ .01, but not youthreported number of physical symptoms, F(2, 413) = 3.94,
p = .02, at posttreatment. Planned contrasts showed that
clinician-rated severity of physical symptoms was significantly greater for participants with at least one comorbid
internalizing disorder, t(434) = 3.89, p \ .01, d = .37, but
did not differ between participants with one versus two or
more comorbid internalizing diagnoses. At posttreatment,
there were no group differences in the number or severity
of physical symptoms based on the presence/number of
comorbid externalizing disorders.
Discussion
Consistent with previous reports [11, 16], over 95 % of
anxiety-disordered youth endorsed at least one somatic
symptom. The number and severity of physical symptoms
decreased over the course of treatment. Treatment type
(i.e., CBT; SRT; CBT ? SRT; pill placebo) did not differentially affect change in the number or severity of
physical symptoms reported. Improvement in anxiety
symptoms, regardless of treatment, was negatively associated with number and severity of physical symptoms at
posttreatment. Treatment responders reported fewer physical symptoms and less severe clinician-rated physical
symptoms than non-responders.
When controlling for age, there were no significant
differences in the number or severity of physical symptoms
across conditions at posttreatment. It is surprising that
123
participants who received active treatment did not have
fewer and less severe physical symptoms than participants
in the placebo condition. Prior studies that reported
reductions in physical symptoms following CBT [16, 19]
but did not include placebo groups may have found
symptom change partially due to the passage of time,
maturation, or repeated assessment. Although Ginsburg
et al. [10] found that pharmacological treatment was
superior to placebo for reducing physical symptoms of
anxiety, the duration of the trial was shorter than CAMS
(8 weeks vs. 12 weeks). It may be that active treatment
reduces the number and severity of some physical symptoms but, with increased knowledge and awareness,
increases participants’ reports of others. Because CBT
teaches participants to identify physical symptoms as signals to engage in coping, it’s possible that CBT results in
self-monitoring and reporting of these somatic signals
(symptoms). In the case of SSRI’s, increased reporting of
physical symptoms due to side effects (e.g., gastrointestinal
symptoms, nausea, insomnia) [37] might balance out anxiety symptom reduction.
Consistent with prior research [10, 14, 16], the number
and severity of physical symptoms were positively associated with anxiety severity and impairment. It may be the case
that more severe anxiety leads to hyperawareness of physical
symptoms, which triggers increases in physical symptoms
that signal anxiety, such as rapid heart rate. Or, it may be that
there is a relationship between physical symptoms and
anxiety severity because physical symptoms are characteristic of anxiety disorders such that youth with more severe
disorders report more severe physical symptoms.
Youth with GAD and SAD reported greater numbers of
and more severe physical symptoms than youth with SoP.
Previous studies examining the relationship between
diagnosis and somatic symptoms yielded mixed findings
[10, 11, 15]. Inconsistencies across studies could be
attributed to measurement issues; the PSC is more comprehensive than measures of physical complaints used in
previous studies and includes items that are not part of the
diagnostic criteria for anxiety disorders. Also, diagnostic
criteria for both GAD and SAD include the presence of
physical symptoms while the criteria for SoP do not. In
other words, it is possible that the presence of somatic
complaints in these samples is related more to screening
and diagnosis than to meaningful differences across these
groups. Alternatively, GAD and SAD may involve more
physiological reactivity than does SoP. Future research is
needed to clarify reasons for the differential distribution of
somatic complaints across diagnostic groups. For example,
studies that involve presenting children with mildly anxiety-provoking situations and examining differences in
physiological reactivity across diagnostic groups could
address this question.
Child Psychiatry Hum Dev
As in prior studies [11, 18], the presence of comorbid
internalizing disorders (but not externalizing disorders) was
related to the severity of somatic complaints at pre- and posttreatment. These findings are consistent with those of Hofflich et al. [11], in which children with comorbid anxiety and
depression reported more frequent somatic symptoms than
children with anxiety alone or children with anxiety and
externalizing symptoms. Internalizing disorders, including
mood and anxiety disorders, are characterized by the presence of physical symptoms including insomnia, restlessness, and fatigue, while the diagnostic criteria for
externalizing disorders do not include physical symptoms.
With regard to demographic variables, age was associated with number and severity of physical symptoms
before and after treatment. This finding is in line with
previous research that used samples spanning childhood
and adolescence [10, 13]. The finding that age was related
to physical symptoms suggests that either youth become
more aware their physical symptoms and more likely to
communicate about them with age, or youth actively
experience more physical symptoms with age.
There were no sex differences in the number of somatic
symptoms reported. This finding is consistent with diagnosed cases [10, 11], though inconsistent with community
samples [13, 14]. Given that parents seek treatment for
their children only when symptoms negatively impact
functioning, it may be that girls in community samples
report somatic complaints more frequently than boys but
that for a subset of girls, somatic symptoms do not impact
their general functioning.
Limitations
Potential limitations merit mention. First, the PARS and
the PSC each assess physical symptoms experienced over
the past week and may not reflect persistent symptoms.
Further, the PSC does not allow for anxiety-related somatic
complaints to be teased apart from symptoms associated
with illness or medication side effects. Although the PARS
is specific to anxiety, it may not be easy for parents and
youth to determine the source of physical symptoms.
Second, this sample excluded youth with a principal
diagnosis of MDD. Research is needed to examine relationships between physical and psychological symptoms as
well as treatment effects in youth with co-principal anxiety
and depressive disorders. Third, the CBT and combination
treatment conditions were not blinded; treatment expectancies may have influenced reports. Fourth, the extent to
which IEs considered change in physical symptoms while
assigning CGI-I ratings cannot be determined. Improvement in somatic complaints may have influenced assessment of treatment response. Finally, findings may not
generalize to non-treatment-seeking samples, as there have
been reports of differences in somatic symptoms between
community and diagnosed samples [9–11, 13, 14].
Summary
As in prior studies [10, 11], somatic complaints were
common among youth diagnosed with an anxiety disorder;
almost all participants endorsed at least one physical
symptom. The total number and severity of pretreatment
physical symptoms was associated with age (but not sex,
race/ethnicity, or SES), principal diagnosis, anxiety
severity, impairment, and the presence of comorbid internalizing (but not externalizing) disorders. Somatic complaints decreased over 12 weeks of CBT (Coping Cat),
medication (sertraline), their combination, or pill placebo.
Findings suggest that an evaluation for anxiety disorders is
warranted when youth present with frequent somatic
complaints (e.g., headaches, stomachaches, difficulty
sleeping, drowsiness, and sniffles) and that promoting
awareness of physical symptoms of anxiety may facilitate
accurate identification. Treatment of anxiety is critical
given impairments in functioning [2, 3, 38, 39]. The current
study suggests that treatment reduces reports of physical
symptoms, more so for treatment responders, but the type
of treatment does not differentially affect the number and
severity of physical symptoms experienced by youth.
Acknowledgments This research was supported by grants (U01
MH064089, to Dr. Walkup; U01 MH64092, to Dr. Albano; U01
MH64003, to Dr. Birmaher; U01 MH63747, to Dr. Kendall, U01
MH64107, to Dr. March; U01 MH64088, to Dr. Piacentini; and U01
MH064003, to Dr. Compton) from the National Institute of Mental
Health. Sertraline and matching placebo were supplied free of charge
by Pfizer. Views expressed within this article represent those of the
authors and are not intended to represent the position of NIMH, NIH,
or DHHS.
References
1. Costello EJ, Mustillo S, Erkanli A, Keeler G, Angold A (2003)
Prevalence and development of psychiatric disorders in childhood
and adolescence. Arch Gen Psychiatry 60(8):837–844
2. Ezpeleta L, Keeler G, Erkanli A, Costello EJ, Angold A (2001)
Epidemiology of psychiatric disability in childhood and adolescence. J Child Psychol Psychiatry 42(7):901–914
3. Van Ameringen M, Mancini C, Farvolden P (2003) The impact of
anxiety disorders on educational achievement. J Anxiety Disord
17(5):561–571
4. Bittner A, Egger HL, Erkanli A, Jane costello E, Foley DL,
Angold A (2007) What do childhood anxiety disorders predict?
J Child Psychol Psychiatry 48(12):1174–1183
5. Costello EJ, Angold A (eds) (1995) Epidemiology. Guilford, New York
6. Ferdinand RF, Verhurlst FC (1995) Psychopathology from adolescence into young adulthood: an 8-year follow-up study. Am J
Psychiatry 152:1586–1594
7. Swendsen J, Conway KP, Degenhardt L, Glantz M, Jin R, Merikangas KR et al (2010) Mental disorders as risk factors for
123
Child Psychiatry Hum Dev
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
substance use, abuse and dependence: results from the 10-year
follow-up of the National Comorbidity Survey. Addiction
105(6):1117–1128
Woodward LJ, Fergusson DM (2001) Life course outcomes of
young people with anxiety disorders in adolescence. J Am Acad
Child Adolesc Psychiatry 40(9):1086–1093
Beidel DC, Christ MAG, Long PJ (1991) Somatic complaints in
anxious children. J Abnorm Child Psychol 19:659–670
Ginsburg GS, Riddle MA, Davies M (2006) Somatic symptoms in
children and adolescents with anxiety disorders. J Am Acad Child
Adolesc Psychiatry 45(10):1179–1187
Hofflich SA, Hughes AA, Kendall PC (2006) Somatic complaints
and childhood anxiety disorders. Int J Clin Health Psychol
6:229–242
American Psychiatric Association (1994) Diagnostic and statistical manual of mental disorders: DSM-IV-TR (4th ed.). Author,
Washington
Garber J, Walker LS, Zeman J (1991) Somatization symptoms in
a community sample of children and adolescents: further validation of the Children’s Somatization Inventory. Psychol Assess
3(4):588–595
Kingery JN, Ginsburg GS, Alfano CA (2007) Somatic symptoms
and anxiety among African American adolescents. J Black Psychol 33(4):363–378
Last CG (1991) Somatic complaints in anxiety disordered children. J Anxiety Disord 5(2):125–138
Storch EA, Merlo LJ, Keeley ML, Grabill K, Milsom VA, Geffken GR et al (2008) Somatic symptoms in children and adolescents with obsessive-compulsive disorder: associations with
clinical characteristics and cognitive-behavioral therapy
response. Behav Cogn Psychother 36(03):283–297
Hughes AA, Lourea-Waddell B, Kendall PC (2008) Somatic
complaints in children with anxiety disorders and their unique
prediction of poorer academic performance. Child Psychiatry
Hum Dev 39(2):211–220
Bernstein GA, Massie ED, Thuras PD, Perwien AR, Borchardt CM,
Crosby RD (1997) Somatic symptoms in anxious-depressed school
refusers. J Am Acad Child Adolesc Psychiatry 36(5):661–668
Masia-Warner C, Colognori D, Kim RE, Reigada LC, Klein RG,
Browner-Elhanan KJ et al (2011) Cognitive-behavioral treatment
of persistent functional somatic complaints and pediatric anxiety:
an initial controlled trial. Depress Anxiety 28(7):551–559
Walkup JT, Albano AM, Piacentini J, Birmaher B, Compton SN,
Sherrill JT et al (2008) Cognitive behavioral therapy, sertraline,
or a combination in childhood anxiety. N Engl J Med
359(26):2753–2766
Silverman WK, Albano AM (1996) The anxiety disorders interview schedule for DSM-IV-child and parent versions. Graywind
Publications, San Antonio
Compton SN, Walkup JT, Albano AM, et al. Child/adolescent
anxiety multimodal study (CAMS): rationale, design, and methods. Child Adolesc Psychiatry Ment Health 4(1): 1–15
Hollingshead AB (1975). Four factor index of social status.
Unpublished manuscript. Department of Sociology, Yale
University
123
24. Kendall PC, Compton SN, Walkup JT, Birmaher B, Albano AM,
Sherrill J et al (2010) Clinical characteristics of anxiety disordered youth. J Anxiety Disord 24(3):360–365
25. Silverman WK, Nelles WB (1988) The anxiety disorders interview schedule for children. J Am Acad Child Adolesc Psychiatry
27(6):772–778
26. Silverman WK, Eisen AR (1992) Age differences in the reliability of parent and child reports of child anxious symptomatology using a structured interview. J Am Acad Child Adolesc
Psychiatry 31(1):117–124
27. Emslie G, Kratochvil C, Vitiello B, Silva S, Mayes T, McNulty S
et al (2006) Treatment for adolescents with depression study
(TADS): safety results. J Am Acad Child Adolesc Psychiatry
45(12):1440–1455
28. Research Units on Pediatric Psychopharmacology Anxiety Study
(2002) The pediatric anxiety rating scale (PARS): development
and psychometric properties. J Am Acad Child Adolesc Psychiatry 41:1061–1069
29. Ginsburg GS, Keeton CP, Drazdowski TK, Riddle MA (2011)
The utility of clinicians ratings of anxiety using the pediatric
anxiety rating scale (PARS). Child Youth Care Forum
40(2):93–105
30. Shaffer D (1983) A children’s global assessment scale (CGAS).
Arch Gen Psychiatry 40(11):1228–1231
31. Bird HR (1987) Further measures of the psychometric properties
of the children’s global assessment scale. Arch Gen Psychiatry
44(9):821–824
32. Guy W (1976) ECDEU assessment manual for psychopharmacology, revised. National Institute of Mental Health, Rockville
33. Kendall PC, Hedtke K (2006) Cognitive-behavioral therapy for
anxious children: therapist manual (3rd ed). Workbook Publishing, Ardmore
34. Kendall PC, Hedtke K (2006) Coping cat workbook (2nd ed.).
Workbook Publishing, Ardmore
35. Kendall PC, Choudhury M, Hudson J, Webb A (2002) The C.A.T
project workbook for the cognitive-behavioral treatment of anxious adolescents. Workbook Publishing, Ardmore
36. Podell JL, Kendall PC, Gosch EA, Compton SN, March JS,
Albano AM, Piacentini JC (2013) Therapist factors and outcomes
in CBT for anxiety in youth. Prof Psychol Res Pract 44(2):89
37. Safer DJ, Zito JM (2006) Treatment-emergent adverse effects
from selective serotonin reuptake inhibitors by age group: children versus adolescents. J Child Adolesc Psychopharmacol
16:159–169
38. Ialongo N, Edelsohn G, Werthamer-Larsson L, Crockett L, Kellam S (1995) The significance of self-reported anxious symptoms
in first grade children: prediction to anxious symptoms and
adaptive functioning in fifth grade. J Child Psychol Psychiatry
36(3):427–437
39. Strauss CC, Frame CL, Forehand R (1987) Psychosocial
impairment associated with anxiety in children. J Clin Child
Psychol 16(3):235–239