Journal of Pediatric Psychology, Vol. 28, No. 6, 2003, pp. 413-422
© 2003 Society of Pediatric Psychology
The Children's Somatization Inventory: Further Evidence for Its Reliability and Validity in a Pediatric and a Community Sample of Dutch Children and Adolescents
Department of Medical, Clinical, and Experimental Psychology, Maastricht University
All correspondence should be sent to Dr. Cor Meesters, Department of Medical, Clinical, and Experimental Psychology, Maastricht University, P.O. Box 616, 6200 MD Maastricht, The Netherlands. E-mail: c.meesters{at}dep.unimaas.nl
 |
Abstract |
|---|
 Top Abstract IntroductionMethodResultsDiscussionReferences |
|---|
Objective To examine the psychometric properties of the Children's Somatization Inventory (CSI) in the Netherlands. Method The CSI and a number of personality and psychopathology questionnaires were administered to Dutch schoolchildren (N = 479), children referred to a pediatric clinic (N = 63), and children's parents. Results Factor analysis yielded a number of factors that have also been found in previous research, viz., pain/weakness, gastrointestinal symptoms, and pseudoneurological symptoms. The reliability (internal consistency) of the CSI was satisfactory. Furthermore, support was obtained for the validity of the CSI. More specifically, the scale correlated in a theoretically meaningful way with child and parent reports of personality and psychopathology, and discriminated well between healthy and pediatric children. Finally, highly similar psychometric properties were obtained for the Parent version of the CSI (i.e., PCSI). Conclusion The Dutch version of the CSI seems to be a reliable and valid self-report measure for assessing somatization symptoms in children and adolescents.
Key words: somatization symptoms; Children's Somatization Inventory; children.
|
Introduction |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionReferences |
|---|
Somatization, which can be defined as "the tendency to experience and communicate somatic distress and symptoms unaccounted for by pathological findings" (Lipowski, 1988
; p. 1359) is regularly encountered in the pediatric
population. When children are referred with multiple somatic complaints that
cannot be fully explained by any known medical condition, the diagnosis of
somatization disorder should be considered. According to the latest edition of
the Diagnostic and Statistical Manual of Mental Disorders, Fourth
Edition-Text Revision (DSM-IV-TR;
American Psychiatric Association [APA],
2000), somatic complaints as seen in somatization disorder can be
allocated in four domains: (a) pain symptoms (e.g., headache, stomachache,
back pain), (b) gastrointestinal symptoms (e.g., nausea, vomiting, diarrhea),
(c) sexual symptoms (e.g., sexual indifference, erectile dysfunction,
irregular menses), and (d) pseudoneurological symptoms (e.g., conversion
symptoms such as impaired coordination, paralysis, loss of touch
sensation).
Recurrent complaints of somatic symptoms are common among children and
adolescents. For example, in a general population survey, Offord et al.
(1987) found recurrent
distressing somatic symptoms to be present in 11% of girls and 4% of boys aged
12 to 16 years. Similarly, Garber, Walker, and Zeman
(1991), who evaluated somatic
symptoms in a large sample of school-age children, found that more than half
of the children reported experiencing at least one serious somatic symptom,
with 15.2% endorsing four or more serious complaints. The most commonly
reported symptoms included headaches, fatigue, sore muscles, abdominal
distress, and back pain. It is generally assumed that somatic symptoms in
children and adolescents follow a predictable developmental course. In younger
children, complaints are monosymptomatic and mostly pertain to recurrent
abdominal pain or headaches (Garber,
Zeman, & Walker, 1990;
Perquin et al., 2000). As
children get older, the number of symptoms increases, with limb pain, aching
muscles, fatigue, and neurological symptoms becoming more prominent (Walker
& Greene, 1989,
1991).
Despite the relatively high prevalence of somatic symptoms, the diagnosis
of somatization disorder in children and adolescents is relatively rare. In
Garber et al.'s (1991)
community sample, for instance, only 1.1% of the children endorsed the
threshold of 13 symptoms required by the DSM, Third Edition-Revised
(DSM-III-R) (APA,
1987). In their review article, Fritz, Fritsch, and Hagino
(1997) argue that the low
prevalence rate of somatization disorder in children and adolescents is
probably due to the fact that the diagnostic criteria are tailored toward
adults. Even current diagnostic criteria
(APA, 2000) require the
individual to suffer from at least one sexual symptom, which is, of course,
not appropriate to the average prepubertal child. In the words of the Fritz et
al. study: "The rarity of the diagnosis of DSM somatization disorder in
children and adolescents probably reflects developmentally inappropriate
criteria more than the disorder arising de novo only in adulthood" (p.
1330).
More epidemiological studies are needed comparing the numbers, frequency
rates, and severity of somatic symptoms in community and pediatric samples in
order to provide an empirical basis for the somatization phenomenon in
children and adolescents (see Campo &
Fritsch, 1994; Fritz et al.,
1997). In such research, standardized assessment instruments are
essential if the results are to be meaningful across studies. A promising
instrument in this regard is the Children's Somatization Inventory (CSI;
Walker & Garber, 1992),
which intends to assess the occurrence of somatization symptoms in children
and adolescents. It can be used in children aged as young as 7 years, as it is
generally acknowledged that children from this age possess sufficient
knowledge about physical illness (e.g.,
Burbach & Peterson, 1986;
Charman & Chandiramani,
1995; Eiser, 1989;
Sigelman, Maddock, Epstein, &
Carpenter, 1993). The CSI includes 35 symptoms that were taken
from the DSM-III-R criteria for somatization disorder and the
somatization factor of the Hopkins Symptom Checklist
(Derogatis, Lipman, Rickels, Uhlenhuth,
& Covi, 1974). Children have to rate the extent to which they
experience each symptom on a 5-point scale ranging from 0 = not at
all to 4 = a whole lot. A total somatization score can be
computed by summing the scores across all items.
Previous research has provided initial support for the reliability and
validity of the CSI. In a first study, Walker, Garber, and Greene
(1991) examined the CSI in a
pediatric sample. These authors reported the CSI to have good internal
consistency, with a Cronbach's alpha of .88 for the total score. Furthermore,
the test-retest correlation (3 months) was .50 for a control group of children
who visited the pediatric clinic for a routine examination and .66 for a group
of children who were referred to the clinic for recurrent abdominal pain,
suggesting acceptable stability in symptom reporting. Evidence was also
obtained for the construct validity of the CSI. More specifically, the CSI
correlated significantly with other measures of emotional problems, such as
the internalizing scale of the Child Behavior Checklist (CBCL;
Achenbach & Edelbrock,
1983), the trait anxiety scale of the State-Trait Anxiety
Inventory for Children (STAIC;
Spielberger, 1973), and the
Children's Depression Inventory (CDI;
Kovacs, 1981). In contrast,
the CSI did not correlate significantly with a scale measuring behavioral
problems, like the CBCL externalizing scale. Finally, CSI scores were
predictive for children's school attendance. That is, the higher the CSI
score, the more frequent children were absent during 3 months following the
initial administration.
In a follow-up study, Garber et al.
(1991) investigated the
psychometric properties of the CSI in a community sample of 540 children and
adolescents. Factor analysis performed on these CSI data yielded four
interpretable factors, although it should be mentioned that various items did
not load convincingly on any of these factors. The factors were
pseudoneurological symptoms, cardiovascular symptoms, gastrointestinal
symptoms, and pain/weakness, and thus strongly resembled a number of the
somatization disorder categories as described in the DSM-III-R.
Furthermore, the CSI had good concurrent validity with another self-report
measure of somatic symptoms (i.e., the Pennebaker Inventory of Limbic
Languidness; Pennebaker,
1982). The construct validity of the CSI was also satisfactory in
this sample: Again, positive associations were found with the STAIC and the
CDI, even after removing the somatic items from these questionnaires, and a
negative connection was found with perceived competence as indexed by the
Self-Perception Profile for Children (SPPC;
Harter, 1985). Finally, the
parent-child agreement of the CSI appeared modest: That is, a small but
significant correlation was found with parents' ratings of their children's
somatization symptoms (see Garber, Van
Slyke, & Walker, 1998).
In another investigation, by Litcher et al.
(2001), the psychometric
properties of the CSI were examined in a sample of 600 10- to 12-year-old
Ukrainian children. Results were largely in keeping with those obtained in
earlier studies. That is, factor analysis yielded four similar factors, as
reported by Garber et al.
(1991), viz.,
pseudoneurological, cardiovascular, gastrointestinal, and pain/weakness
symptoms. Further, CSI scores were in a theoretically meaningful way related
to self-reports and parents' ratings of children's health and psychological
functioning (e.g., anxiety, depression, self-esteem).
A final study that should be mentioned is that by Walker, Garber, and
Greene (1993), who compared
CSI scores of children with recurrent abdominal complaints, children with
peptic disease, children with emotional disorders, and healthy children.
Results showed that children with recurrent abdominal complaints and peptic
disease displayed the highest CSI scores, healthy children displayed the
lowest CSI scores, whereas children with emotional disorders were in between.
Altogether, these results provided tentative evidence for the discriminant
validity of the CSI.
Altogether, research in the United States and Ukraine has provided evidence
for the reliability and validity of the CSI as an instrument for assessing
somatization symptoms in children and adolescents. The present research
further examined the psychometric properties of the CSI in the Netherlands.
Dutch schoolchildren (N = 479) and children referred to a pediatric
clinic (N = 63) were asked to complete the CSI in order to: (a)
investigate the factor structure of the Dutch CSI, (b) further examine the
reliability (internal consistency) of the scale, (c) establish the construct
validity of the CSI through its associations with measures of psychopathology
and personality, and (d) test the discriminant validity of the CSI by
comparing the scores of healthy and pediatric children. Finally, it is
important to note that we also obtained parent-rated somatization scores
(using a parent version of the CSI) for most children and adolescents, and so
it became possible (e) to assess the parent-child agreement of the CSI. In
keeping with previous studies (Garber et
al., 1991; Litcher et al.,
2001), it was expected that (a) factor analysis would yield four
factors, viz., pseudoneurological symptoms, cardiovascular symptoms,
gastrointestinal symptoms, and pain/weakness, (b) the internal consistency
would be adequate (i.e., Cronbach's alpha > .70), (c) the CSI would be
positively associated with anxiety, depression, and neuroticism and negatively
associated with self-perceived competence, (d) CSI scores would discriminate
healthy from pediatric children and adolescents, and (e) the parent-child
correspondence of the CSI would be moderate but significant.
|
Method |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionReferences |
|---|
Participants and Procedure
Eight-hundred eighty-two children and adolescents from six regular primary and secondary schools and their parents were invited to participate in the present study. More than half of the youths and their parents (N = 495; i.e., 56.1%) responded favorably to our invitation and completed the informed consent form. Youth participants were asked to complete the CSI and a number of other questionnaires at school during regular classes, with a teacher and a research assistant always being present in order to ensure independent and confidential responding and to provide assistance when necessary. Parent participants received the questionnaires via their children, completed them at home, and returned materials in a sealed envelope. Eventually, complete data sets of 479 children and adolescents (220 boys and 259 girls) were available. Mean age of the children was 12.79 years (SD = 1.39, range, 10-16 years). Percentages of participants from low, middle, and high socioeconomic backgrounds (based on educational levels of parents) were 14.4%, 64.4%, and 21.3%, respectively. Complete parent data were obtained for 438 children and adolescents.
The pediatric sample consisted of 63 consecutive patients (26 boys and 37 girls; mean age = 11.62, SD = 1.89, range, 8 -16 years) who were referred to the psychologist of the pediatric department of the Academic Hospital Maastricht, the Netherlands. These children and adolescents did not suffer from a diagnosable physical health condition, but all displayed somatic symptoms and were seen by the psychologist for evaluation of these complaints or treatment. Most children had pain symptoms (n = 33, of which 14 had headaches and 14 had pain in the stomach). Other frequently reported symptoms were fatigue (n = 10), constipation (n = 7), and nausea (n = 7). Children in the pediatric sample had a middle or high socioeconomic background (87.5% and 12.5%, respectively). After informed consent was obtained, all children completed the CSI as well as other questionnaires during the standard psychological assessment under supervision of a psychological test assistant. For a substantial proportion of the pediatric sample, parent data of the CSI and other scales were also available.
Questionnaires
The Dutch version of the CSI (Ghys
& Meesters, 1993) was employed. The instrument was translated
from English by a native Dutch speaker and independently translated back into
English. This procedure ensured that the translated items closely resembled
the original English items. The CSI contains 35 items that have to be rated on
a 5-point scale: 0 = not at all, 1 = a little, 2 =
somewhat, 3 = a lot, and 4 = a whole lot),
reflecting the extent to which the symptoms were experienced in the past 2
weeks. A total score can be computed by summing the scores across all items,
with higher scores indicating a higher intensity of somatic complaints.
The Parent version of the CSI (PCSI) was identical to the CSI, except that parents completed the questions with regard to their children's somatic complaints during the past 2 weeks using the same response format as the child version.
The Amsterdamse Biografische Vragenlijst voor Kinderen [Amsterdam
Biographical Questionnaire for Children] (ABVK;
Van Dijl & Wilde, 1982) is
a 115-item self-report questionnaire of which 86 items are intended to measure
the following personality traits in children: neuroticism (30 items; e.g.,
"I am quite nervous"), neuroticism as manifested through somatic
complaints (20 items; e.g., "I am often bothered by headaches"),
extraversion (20 items; e.g., "I always feel fine and at ease when I am
with other people"), and social desirability (16 items; e.g., "I
always think friendly thoughts about everyone"). Each item has to be
answered with yes or no, and thus yields one point for the
pertinent personality dimension. In the present study, reliability
coefficients (Cronbach's alphas) were .94 and .80 for neuroticism, .94 and .79
for somatic complaints, .92 and .75 for extraversion, and .89 and .74 for
social desirability in the healthy and the pediatric sample, respectively.
The SPPC consists of 36 items that can be allocated to six subscales representing various aspects of children's self-esteem, viz., scholastic competence, social acceptance, athletic competence, physical appearance, behavioral conduct, and global self-worth. Each SPPC item consists of two opposite descriptions, for example, "Some children often forget what they have learned" and "Other children are able to remember things easily." Children have to choose the description that best fits and then indicate whether the description is somewhat true or very true for them. Accordingly, each item is scored on a 4-point scale, with a higher score reflecting a more positive view of oneself. In the present study, a total score (Cronbach's alphas were .90 and .89 in the healthy and the pediatric sample, respectively) was computed by summing all items.
The trait anxiety scale of the STAIC contains 20 items that measure chronic symptoms of anxiety. Children are asked to rate the frequency with which (s)he experiences anxiety symptoms such as "I am scared" and "I feel troubled" using a 3-point scale: 1 = almost never, 2 = sometimes, and 3 = often. Three STAIC items that referred to somatic symptoms of anxiety were removed from the scale. The internal consistency of the resulting 17-item scale was satisfactory, with Cronbach alphas of .92 in the healthy sample and.84 in the pediatric sample.
The short version of the Depression Questionnaire for Children (DQC;
De Wit, 1987) consists of nine
items such as "I feel depressed lately" and "I often think
that other children do not like me." Children have to indicate whether
items are true or not true for them. One somatic-like item
("I do not feel well lately") was deleted from the scale; the
internal consistency of the eight-item DQC was .70 in the healthy sample and
.69 in the pediatric sample.
The CBCL comprises 118 items addressing emotional and behavioral problems of children. Parents are asked to indicate on a 3-point scale the extent to which each item applies to their child: 0 = never, 1 = sometimes, and 2 = often. The CBCL addresses two broad domains in which problems of children and adolescents manifest themselves. One domain is externalizing, which reflects behavioral problems, and the other is internalizing, which refers to emotional problems. In addition, factor analysis has yielded eight narrow-band factors that have been replicated across different gender and age groups: withdrawn (e.g., "Likes to be alone"), somatic complaints (e.g., "Suffers from pain"), anxious-depressed (e.g., "Is nervous"), social problems (e.g., "Doesn't get along with other kids"), thought problems (e.g., "Hears things that aren't there"), attention problems (e.g., "Can't concentrate"), delinquent behavior (e.g., "Steals things at home"), and aggressive behavior (e.g., "Argues a lot"). CBCL scores are computed by summing relevant items. In both samples, Cronbach's alphas were in the .80 -.90 range for internalizing and externalizing, and higher than .60 for the narrow-band factors (except for the factor thought problems).
In order to prevent possible order effects and minimize the impact of fatigue, half of the healthy children and adolescents first completed the CSI and then the ABVK, DQC, STAIC, and SPPC, whereas the other half received the questionnaires in the reverse order. In pediatric patients, questionnaires were administered in random order.
Statistical Analysis
The Statistical Package for Social Sciences was used for carrying out
statistical analyses. For the healthy sample, data were included in the
statistical analyses only when children/adolescents had completely filled in
all measures, that is, the CSI, ABVK, SPPC, STAIC, and DQC, and when parents
had completed both the PCSI and CBCL. Cases that contained missing values were
deleted. This procedure yielded a final sample of 479 children and adolescents
and 438 parents. Principal-components factor analysis was conducted using
Varimax rotation (following the procedure that was employed in previous factor
analytic studies of the CSI; Garber et
al., 1991; Litcher et al.,
2001). Allocation of individual items to a designated factor was
based on the largest item-factor loading that exceeded .30. Reliability
(internal consistency) of the CSI and PCSI was assessed by means of Cronbach's
alpha. Parent-child agreement was investigated by means of Pearson
correlations. To establish discriminant validity, scores of healthy and
pediatric children were compared by means of analyses of variance in which we
covaried confounding variables (i.e., gender, age, and socioeconomic status).
Construct validity of the CSI and PCSI was tested by means of Pearson
correlations with p < .05/17 (i.e., Bonferroni correction). As
previous studies by Garber et al.
(1991) and Litcher et al.
(2001) have demonstrated that
gender and age are associated with CSI scores, partial correlations were
computed, in order to control for the possible influence of these variables on
the association between the CSI and criterion variables. Due to the
nonindependence that exists between (sub)scales of ABVK, CBCL, SPPC, STAIC,
and DQC, the method devised by Meng, Rosenthal, and Rubin
(1992) of comparing correlated
correlation coefficients was employed to compare the magnitudes of partial
correlations.
|
Results |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionReferences |
|---|
Healthy Sample
Descriptive Statistics1
The mean CSI score for the total group of healthy children and adolescents was 10.16 (SD = 10.36). Girls scored significantly higher on the CSI than boys, means being 11.31 (SD = 11.65) and 8.80 (SD = 8.43), respectively [t(477) = 2.66, p < .01]. Thus, girls reported a higher intensity of somatic complaints than did boys. Furthermore, a small but significant negative association was found between age and CSI scores, r = -.16, p < .001, indicating that somatic complaints somewhat decreased as children became older. Finally, the number of somatization symptoms on the CSI that children reported as having been present a lot or a whole lot were examined in order to get a picture of the frequency of severe somatization symptoms (see Garber et al., 1991
). Results
showed that 72.0% of the children showed no severe symptoms; 14.2% reported
one symptom, 6.7% reported two, 2.1% reported three, 1.5% reported four, 1.3%
reported five, and 0.6% reported six symptoms. Only 1.7% of the children
reported a total of seven or more severe somatic complaints, which is the
number required for a DSM-IV-TR diagnosis of somatization disorder
(when excluding sexual symptoms that are not covered by the CSI). The most
prevalent (severe) symptoms were "headaches" (9.8%), "pain
in stomach" (5.6%), "nausea/upset stomach" (4.8%),
"low energy" (3.8%), and "pain in lower back"
(3.8%).
Reliability
The CSI had a Cronbach's alpha of .90, indicating that the internal
consistency of the scale was good. Item-total correlations varied from .24
("blindness") to .64 ("nausea or upset stomach"), with
a mean of .43. The Cronbach's alpha of the PCSI was also reasonable: .80. The
mean item-total correlation was .28, ranging between .00 ("muscle
weakness") and .58 ("weakness in body parts") for individual
items.
Factor Structure of the CSI and PCSI
Factor analysis performed on the CSI data of healthy children (N =
479) yielded nine factors with eigenvalues greater than 1.0, which accounted
for 58.83% of the variance. The four-factor structure was inspected first, as
this solution had proved to be satisfactory in previous factor analytic
research with the CSI (Garber et al.,
1991; Litcher et al.,
2001). Three of the factors were clearly in keeping with those
obtained in earlier studies, viz., pain/weakness, gastrointestinal symptoms,
and pseudoneurological symptoms. However, the fourth factor contained only
four items (difficulty urinating, constipation, pain when urinating, and
memory loss). Thus, no trace of a "cardiovascular symptoms" factor
was found. Therefore, it was decided that the three-factor solution as
presented in Table I provided
the most parsimonious and satisfactory solution for the structure of the CSI
data. The three factors were pain/weakness, gastrointestinal symptoms, and
pseudoneurological symptoms, which together accounted for 37.4% of the
variance.
|
The factor structure of the PCSI data was more in line with that obtained in previous research. As can be seen in Table II, four factors emerged that accounted for 37.8% of the variance: pain/weakness, gastrointestinal symptoms, cardiovascular symptoms, and pseudoneurological symptoms.
|
Construct Validity
Validity of the CSI was examined through its correlations (corrected for
gender and age) with measures of personality and psychopathology. As is shown
in Table III, CSI scores were
significantly connected to ABVK neuroticism (r = .32) and
neuroticism-somatic complaints (r = .43), but not to extraversion and
social desirability. Further, a negative association was found with SPPC
self-perceived competence (r = -0.26), while positive connections
emerged with STAIC trait anxiety and DQC depression (r = .38 and .44,
respectively). Correlations with CBCL scores also showed the predicted
pattern. That is, a significant link was found with internalizing (r
= .24) but not with externalizing (r = .07), and CSI scores were more
strongly associated with somatic complaints (r = .37) than any of the
other CBCL factors (Z 
3.20 for all, p < .01).
|
A comparable pattern of results was found for the PCSI. That is, PCSI
scores were positively associated with ABVK neuroticism and
neuroticism-somatic complaints (r = .20 and .32, respectively), STAIC
trait anxiety (r = .19), and DQC depression (r = .19).
Furthermore, PCSI scores were more strongly related to CBCL internalizing
(r = .55) and CBCL somatic complaints (r = .63) than to CBCL
externalizing (r = .23) and other narrow-band factors (Z
5.11 for all, p < .001).
Parent-Child Agreement
The correlation between CSI and PCSI was 0.44 (N = 438, p
< .001). Furthermore, a paired t-test indicated that children
reported higher levels of somatization symptoms than their parents, means
being 10.21 (SD = 10.24) versus 4.16 (SD = 5.20)
[t(437) = 13.75, p < .001].
Pediatric Sample
Descriptive Statistics
In the pediatric sample, no gender difference was found for the CSI. Mean
scores were 23.27 (SD = 18.60) and 19.85 (SD = 19.12) for
girls and boys, respectively [t(61) = .71, p = .48].
Furthermore, no significant association was found between age and CSI scores
(r = .17, p = .19). Finally, the majority of the sample
(66.7%) reported at least one severe somatization symptom on the CSI: 14.3%
reported one symptom, 12.7% reported two, 9.5% reported three, 9.5% reported
four, 4.8% reported five, and 3.2% reported six symptoms. Eight children
(12.7%) reported seven or more severe somatic complaints, the number of
symptoms mentioned in the DSM-IV-TR for the diagnosis of somatization
disorder. Most common symptoms in the pediatric sample were headaches (30.2%),
pain in the stomach (20.6%), nausea/upset stomach (17.5%), pain in the lower
back (15.9%), and weakness in body parts (15.9%).
Reliability
The internal consistency of the CSI in the pediactric sample was good:
Cronbach's alpha was .92. All item-total correlations were between .21 (pain
in chest) and .76 (blurred vision), with a mean of .48. For the PCSI,
Cronbach's alpha was .88. Mean item-total correlation was .40, with only five
items clearly showing values too low (i.e., < .20): loss of voice,
blindness, fainting, seizures, and muscle weakness.
Construct Validity
Correlations (corrected for gender and age) between CSI and PCSI and
personality and psychopathology measures in the pediatric sample are displayed
in Table IV. Results were
largely in keeping with those obtained in the healthy sample. That is,
significant associations were found with ABVK neuroticism-somatic complaints
(CSI and PCSI), SPPC self-perceived competence (PCSI), STAIC trait anxiety
(CSI), CBCL internalizing (PCSI), and CBCL somatic complaints (CSI and
PCSI).
|
Parent-Child Agreement
The correlation between CSI and PCSI in the pediatric sample was .42
(N = 38, p < .05). A paired t-test indicated
that there was a trend for children to report higher levels of somatization
symptoms than their parents, means being 19.79 (SD = 18.82) versus
14.16 (SD = 13.55) [t(37) = 1.92, p = .06].
Discriminant Validity
Figure 1 displays the mean
CSI and PCSI scores of healthy and pediatric children. Analyses of variance
with age, gender, and socioeconomic status as covariates revealed that
pediatric children displayed significantly higher somatization scores than
healthy children, means being 21.67 (SD = 18.65) versus 10.12
(SD = 10.35) for the CSI [F(1, 553) = 45.24, p <
.001] and 14.05 (SD = 13.39) versus 4.13 (SD = 5.17) for the
PCSI [F(1, 484) = 87.02, p < .001]. When comparing these
scores to the normative data of our healthy sample, it appeared that 30.2% and
63.2% of the pediatric sample scored in the highest decile of CSI and PCSI
scores, respectively. Because of heterogeneity of variance in CSI and PCSI
scores in the healthy and the pediatric sample, the analysis of covariance was
repeated employing square root transformed data. Results were highly similar
to those obtained with the untransformed values [for CSI: F(1, 553) =
38.26, p < .001, and for PCSI: F(1, 484) = 65.93,
p < .001].
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionReferences |
|---|
The present study examined the psychometric properties of the CSI in Dutch children and adolescents. The main results can be summarized as follows. To begin with, factor analysis yielded some evidence for the presence of a number of the somatization factors that have been found in previous research, viz., pain/weakness, gastrointestinal symptoms, and pseudoneurological symptoms. Furthermore, the reliability (internal consistency) of the scale appeared to be good. Evidence was also obtained for the validity of the CSI. More specifically, the scale correlated in a theoretically meaningful way with child and parent reports of personality and psychopathology, and discriminated well between healthy and pediatric children. Finally, highly similar psychometric properties were obtained for the PCSI. Altogether, the current findings are in keeping with those of previous studies in other countries (i.e., United States and Ukraine) and confirm the notion that the Dutch version of the CSI is a reliable and valid self-report measure for assessing somatization symptoms in children and adolescents.
Factor analysis of the CSI data revealed a number of factors that have been
reported in previous studies with this instrument. More specifically, most of
the items convincingly loaded on the factors pain/weakness,
gastrointestinal symptoms, and pseudoneurological symptoms (see
also Garber et al., 1991;
Litcher et al., 2001). The
cardiovascular symptoms factor was not found, at least when analyzing children
and adolescents' self-report data. A number of remarks can be made with regard
to this finding. First of all, whereas the DSM-III-R criteria for
somatization disorder still included a number of cardiovascular symptoms
(i.e., shortness of breath, palpitations, chest pain, and dizziness), the
current edition no longer lists such symptoms in the criteria. Thus, the three
factors that were found in the current study nicely correspond with the
symptom clusters that are mentioned in the DSMIV-TR (sexual symptoms
excluded). Second, the CSI contains a rather heterogeneous set of symptoms
that are relatively rare in healthy children and adolescents. So, it is not
that surprising that factor analysis of the CSI in different samples will
yield somewhat different solutions. In line with this argument, it should be
mentioned that whereas the studies by Garber et al.
(1991) and Litcher et al.
(2001) both obtained a
four-factor solution for the CSI, almost half of the items (45.9%) did not
load on similar factors.
At a theoretical level, the current data replicate a number of findings
that have been previously found in the literature on children's somatization
symptoms. First, evident gender differences were found for the CSI. This
result accords well with earlier studies showing that girls generally report
higher levels of somatization symptoms than boys (for a comprehensive review,
see Fritz et al., 1997).
Second, a small but significant age effect was found for the CSI, with younger
children displaying somewhat higher levels of somatization symptoms than older
children. Again, this finding corresponds with previous studies that reported
a decline in somatization symptoms with increasing age (e.g.,
Garber et al., 1991). Third,
headaches, pain in the stomach, and low energy consistently appear among the
most prevalent somatic symptoms in children and adolescents
(Garber et al., 1991;
Litcher et al., 2001). Fourth,
significant associations were found between somatization as indexed by the CSI
and anxiety-depression. This result fits well with findings from
epidemiological studies showing that there is comorbidity between somatization
disorder and anxiety disorders and depression (see
Fritz et al., 1997). Fifth, in
agreement with previous research (e.g., Garber et al.,
1991,
1998;
Litcher et al., 2001), the
present findings illustrate that the correspondence between children and their
parents is rather modest in the case of somatization symptoms. More precisely,
parents appeared to report lower levels of somatization symptoms than
children. Assuming that children are the most important informant in the case
of internalizing problems such as somatization
(Achenbach, McConaughy, & Howell,
1987), this finding suggests that parents are either only partly
aware of their child's symptoms or underestimate the severity of such
complaints. It should be noted, however, that the latter findings are in clear
contrast with the study by Garber et al.
(1998), who found that mothers
of patients with recurrent abdominal pain reported more symptoms in their
children than did the children themselves, and this was particularly true for
mothers who displayed high levels of maternal stress. Most children and
adolescents who were included in the present study (even those in the
pediatric sample) were relatively well functioning. Perhaps, then, levels of
distress in parents was rather low, resulting in reserved ratings of
children's symptoms.
Altogether, the current results provide further support for the utility of the CSI as an index of somatization symptoms in youths. The scale seems to be a valuable addition to the standardized assessment of somatic complaints, particularly in pediatric settings. However, some notes of caution should be made while interpreting the results of the present study. First of all, the number of pediatric patients was relatively small, posing a clear limitation to the validity study. Furthermore, it should be borne in mind that high CSI scores not unambiguously point in the direction of a somatization disorder. High scores may also reflect the presence of a physical problem, another somatoform disorder, an anxiety disorder (in particular panic disorder or generalized anxiety disorder), or depression. The present study did not include a standardized diagnostic instrument in order to examine which children and adolescents in our healthy and pediatric samples really suffered from somatization disorder. Future studies should address the issue in order to establish the sensitivity and specificity of the CSI. Such data could also be helpful for providing a sound empirical basis for the diagnostic criteria of this mental disorder in children and adolescents.
|
Note |
|---|
1. Gender-specific normative data based on the healthy sample are available upon request from the first author.
Received June 17, 2002; revision received October 14, 2002; revision received December 2, 2002; accepted December 18, 2002
|
References |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionReferences |
|---|
Achenbach, T. M., & Edelbrock, C. (1983). Manual for the Child Behavior Checklist and Revised Child Behavior Profile. Burlington: University of Vermont, Department of Psychiatry.
Achenbach, T. M., McConaughy, S. H., & Howell, C. T. (1987). Child/adolescent behavioral and emotional problems: Implications of cross-informant correlations for situational specificity. Psychological Bulletin, 101, 213 -232.[CrossRef][Web of Science][Medline]
American Psychiatric Association (1987). Diagnostic and statistical manual of mental disorders (3rd ed., revised). Washington, DC: Author.
American Psychiatric Association (2000). Diagnostic and statistical manual of mental disorders (4th ed., text revision). Washington, DC: Author.
Burbach, D. J., & Peterson, L. (1986). Children's concepts of physical illness: A review and critique of the cognitive-developmental literature. Health Psychology, 5, 307-325.[CrossRef][Web of Science][Medline]
Campo, J. V., & Fritsch, S. L. (1994). Somatization in children and adolescents. Journal of the American Academy of Child and Adolescent Psychiatry, 33, 1223 -1235.[Web of Science][Medline]
Charman, T., & Chandiramani, S. (1995). Children's understanding of physical illnesses and psychological states. Psychology and Health, 10, 145 -153.
Derogatis, L. R., Lipman, R. S., Rickels, K., Uhlenhuth, E. H., & Covi, L. (1974). The Hopkins Symptom Checklist (HSCL): A self-report symptom inventory. Behavioral Science, 19, 1-15.[Web of Science][Medline]
De Wit, C. (1987). Depressie Vragenlijst voor Kinderen (Depression Questionnaire for Children), DVK en KDVK. Amersfoort, Netherlands: Acco.
Eiser, C. (1989). Children's concepts of illness: Towards an alternative to the "stage" approach. Psychology and Health, 3, 93 -101.
Fritz, G. K., Fritsch, S., & Hagino, O. (1997). Somatoform disorders in children and adolescents: A review of the past 10 years. Journal of the American Academy of Child and Adolescent Psychiatry, 36, 1329 -1338.[CrossRef][Web of Science][Medline]
Garber, J., Van Slyke, D. A., & Walker, L. S. (1998). Concordance between mothers' and children's reports of somatic and emotional symptoms in patients with recurrent abdominal pain or emotional disorders. Journal of Abnormal Child Psychology, 26, 381-391.[CrossRef][Web of Science][Medline]
Garber, J., Walker, L. S., & Zeman, J. (1991). Somatization symptoms in a community sample of children and adolescents: Further validation of the Children's Somatization Inventory. Psychological Assessment, 3, 588 -595.[CrossRef]
Garber, J., Zeman, J., & Walker, L. S. (1990). Recurrent abdominal pain in children: Psychiatric diagnoses and parental psychopathology. Journal of the American Academy of Child and Adolescent Psychiatry, 29, 648 -656.[Web of Science][Medline]
Ghys, A., & Meesters, C. (1993). Dutch translation of the Children's Somatization Inventory. Maastricht, The Netherlands: Academic Hospital Maastricht, Department of Medical Psychology.
Harter, S. (1985). Manual for the Self-Perception Profile for Children. Denver: University of Denver.
Kovacs, M. (1981). Rating scales to assess depression in school-aged children. Acta Paedopsychiatrica, 46, 305 -315.[Web of Science][Medline]
Litcher, L., Bromet, E., Carlson, G., Gilbert, T., Panina, N., Golovakha, E., et al. (2001). Ukrainian application of the Children's Somatization Inventory: Psychometric properties and associations with internalizing symptoms. Journal of Abnormal Child Psychology, 29, 165 -175.[CrossRef][Web of Science][Medline]
Lipowski, Z. J. (1988). Somatization: The concept and
its clinical application. American Journal of
Psychiatry, 145,
1358 -1368.
Meng, X. L., Rosenthal, R., & Rubin, D. B. (1992). Comparing correlated correlation coefficients. Psychological Bulletin, 111, 172 -175.[CrossRef][Web of Science]
Offord, D. R., Boyle, M. H., Szatmari, P., Rae-Grant, N. I., Links,
P. S., Cadman, D. T., et al. (1987). Ontario Child Health Study
II: Six-month prevalence of disorder and rates of service utilization.
Archives of General Psychiatry,
44, 832
-836.
Pennebaker, J. W. (1982). The psychology of physical symptoms. New York: Springer Verlag.
Perquin, C. W., Hazebroek-Kampschreur, A. A., Hunfeld, J. A., Bohnen, A. M., Suijlekom-Smit, L. W., Passchier, J., et al. (2000). Pain in children and adolescents: a common experience. Pain, 87, 51-58.[CrossRef][Web of Science][Medline]
Sigelman, C., Maddock, A., Epstein, J., & Carpenter, W. (1993). Age differences in understanding of disease causality: AIDS, colds, and cancer. Child Development, 64, 272 -284.[CrossRef][Web of Science][Medline]
Spielberger, C. D. (1973). Manual for the State-Trait Anxiety Inventory for Children. Palo Alto, CA: Consulting Psychologists Press.
Van Dijl, H., & Wilde, G. (1982). Handleiding bij de Amsterdamse Biografische Vragenlijst voor Kinderen, ABVK, en de Korte Amsterdamse Biografische Vragenlijst voor Kinderen, KABVK. Amsterdam: F. van Rossen.
Walker, L. S., & Garber, J. (1992). Children's Somatization Inventory: Preliminary manual. Nashville, TN: Vanderbilt University Medical Center.
Walker, L. S., Garber, J., & Greene, J. W. (1991). Somatization symptoms in pediatric abdominal pain patients: Relation to chronicity of abdominal pain and parent somatization. Journal of Abnormal Child Psychology, 19, 379 -394.[CrossRef][Web of Science][Medline]
Walker, L. S., Garber, J., & Greene, J. W. (1993). Psychosocial correlates of recurrent childhood pain: A comparison of pediatric patients with recurrent abdominal pain, organic illness, and psychiatric disorders. Journal of Abnormal Psychology, 102, 248 -258.[CrossRef][Web of Science][Medline]
Walker, L. S., & Greene, J. W. (1989). Children
with recurrent abdominal pain and their parents: More somatic complaints,
anxiety, and depression than other patient families? Journal of
Pediatric Psychology, 14,
231-243.
Walker, L. S., & Greene, J. W. (1991). Negative
life events and symptom resolution in pediatric abdominal pain patients.
Journal of Pediatric Psychology,
16,
341-360.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  M. Vila, T. Kramer, N. Hickey, M. Dattani, H. Jefferis, M. Singh, and M. E. Garralda Assessment of Somatic Symptoms in British Secondary School Children Using the Children's Somatization Inventory (CSI) J. Pediatr. Psychol., October 1, 2009; 34(9): 989 - 998. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. J. Bakker, E. M. van de Putte, W. Kuis, and G. Sinnema Risk Factors for Persistent Fatigue With Significant School Absence in Children and Adolescents Pediatrics, July 1, 2009; 124(1): e89 - e95. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. S. Walker, J. E. Beck, J. Garber, and W. Lambert Children's Somatization Inventory: Psychometric Properties of the Revised Form (CSI-24) J. Pediatr. Psychol., May 1, 2009; 34(4): 430 - 440. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E M van de Putte, R H H Engelbert, W Kuis, J L L Kimpen, and C S P M Uiterwaal How fatigue is related to other somatic symptoms Arch. Dis. Child., October 1, 2006; 91(10): 824 - 827. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Vervoort, L. Goubert, C. Eccleston, P. Bijttebier, and G. Crombez Catastrophic Thinking About Pain is Independently Associated with Pain Severity, Disability, and Somatic Complaints in School Children and Children with Chronic Pain J. Pediatr. Psychol., August 1, 2006; 31(7): 674 - 683. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. M. van de Putte, L. J. P. van Doornen, R. H. H. Engelbert, W. Kuis, J. L. L. Kimpen, and C. S. P. M. Uiterwaal Mirrored Symptoms in Mother and Child With Chronic Fatigue Syndrome. Pediatrics, June 1, 2006; 117(6): 2074 - 2079. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Y. Konijnenberg, E. R. de Graeff-Meeder, J. van der Hoeven, J. L. L. Kimpen, J. K. Buitelaar, C. S.P.M. Uiterwaal, and and the Pain of Unknown Origin in Children Study G Psychiatric Morbidity in Children With Medically Unexplained Chronic Pain: Diagnosis From the Pediatrician's Perspective Pediatrics, March 1, 2006; 117(3): 889 - 897. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||