The revised version of the Screen for Child Anxiety Related Emotional Disorders (SCARED-R): Factor structure in normal children

https://doi.org/10.1016/S0191-8869(98)00130-5Get rights and content

Abstract

The revised version of the Screen for Child Anxiety Related Emotional Disorders (SCARED-R) is a self-report questionnaire that measures symptoms of DSM-IV linked anxiety disorders in children. The present study examined the factor structure of the SCARED-R in a sample of 674 normal Dutch school children aged 8 to 13 years. Exploratory factor analysis (principal components with oblimin rotation) clearly pointed in the direction of a 1-factor solution, suggesting that when applied to samples of normal children, the SCARED-R is a unidimensional measure. Additional exploratory and confirmatory factor analyses carried out on parts of the SCARED-R provided some support for the presence of the following factors: panic disorder, generalized anxiety disorder, separation anxiety disorder, school phobia, social phobia and three types of specific phobias. Implications of these findings for the use of the SCARED-R are briefly discussed.

Introduction

Anxiety disorders are one of the most common psychiatric problems in children. Estimates for the presence of any anxiety disorder range between 5.7 and 17.7%, with half of them exceeding the 10% rate (Costello and Angold, 1995). The latest edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV; APA, 1994) which is the most current classificatory system in many countries (including The Netherlands), recognizes the following anxiety disorders in children: generalized anxiety disorder, separation anxiety disorder, panic disorder, specific phobia, social phobia, obsessive–compulsive disorder and posttraumatic or acute stress disorder. Epidemiological studies have revealed that generalized anxiety disorder, separation anxiety disorder and specific phobias are the most commonly diagnosed anxiety disorders, occurring in about 5% of the children. Social phobia and panic disorder are less frequent with prevalence rates generally below 2%, while obsessive–compulsive disorder and posttraumatic or acute stress disorders are even more rare. There is evidence to suggest that subclinical manifestations of anxiety disorders are also relatively prevalent among children. For example, Bell-Dolan et al. (1990)found that symptoms of generalized anxiety disorder, separation anxiety disorder, and specific phobias were present in 20–30% of a sample of never-psychiatrically ill children.

Structured and semistructured interviews can be used to reliably assess anxiety disorders symptomatology in children. However, these interviews are time-consuming and often require trained interviewers. Alternatively, self-report questionnaires can be employed to measure anxiety symptoms in children. Currently, the three most widely used scales for this purpose are the Revised Children's Manifest Anxiety Scale (RCMAS; Reynolds and Richmond, 1978), the State-Trait Anxiety Inventory for Children (STAIC; Spielberger, 1973) and the Fear Survey Schedule for Children, Revised (FSSC-R; Ollendick, 1983). It should be noted that these scales are global and unidimensional and not keyed to the separate anxiety disorders as listed in the DSM.

Recently, several attempts have been made to develop multidimensional questionnaires for assessing childhood anxiety symptoms. A first example is the Multidimensional Anxiety Scale for Children (MASC; March et al., 1997), a self-report measure comprising four empirically-derived domains of childhood anxiety: physical anxiety, harm avoidance, social anxiety and separation anxiety. Another example is the Children's Anxiety Scale (CAS; Spence, 1997) that taps symptoms of a number of DSM-defined anxiety disorders, namely panic disorder, separation anxiety disorder, social phobia, obsessive–compulsive disorder and generalized anxiety disorder. In addition, the CAS contains a physical fears scale which represents the DSM-category of specific phobias. Interestingly, Spence (1997)showed in a factor-analytic study that anxiety symptoms as listed in the CAS cluster into subtypes of anxiety problems that are largely consistent with the classification of anxiety disorders as proposed in the DSM.

The Screen for Child Anxiety Related Emotional Disorders (SCARED; Birmaher et al., 1997) is also a multidimensional questionnaire that purports to measure DSM-defined anxiety symptoms. Factor analysis performed on the SCARED scores of clinically referred children revealed that the 38 items of the SCARED can be allocated to five separate anxiety subscales. Four of these subscales represent anxiety disorders that correspond with DSM categories, namely panic disorder, generalized anxiety disorder, social phobia, and separation anxiety disorder. The fifth subscale is school phobia which, according to Birmaher et al. (1997), p. 551) can best be considered as “a common clinical entity that is seen both comorbidly and independently from other anxiety disorders”. Muris (1997)revised the SCARED in three ways. To begin with, school phobia items were joined to the separation anxiety disorder subscale. This was done because the DSM (see especially the DSM-III-R; APA, 1987) views school phobia as a symptom of separation anxiety disorder. Secondly, 15 new items were added in an attempt to index symptoms of specific phobia. The DSM-IV distinguishes three subtypes: animal phobia, situational–environmental phobia and blood-injection-injury phobia [see Frederikson et al. (1996), Muris et al. (submitteda)who provided evidence for these three separate categories of specific phobias]. Because specific fears and phobias are highly prevalent among children (Bernstein et al., 1996), items of all three subtypes were included in the revised SCARED. Thirdly, although obsessive–compulsive disorder and traumatic stress disorder are relatively rare, an extra 13 items were added so that it would also be possible to tap symptoms of these disorders with the revised SCARED. Thus, the final 66-item revised version of the SCARED purports to measure the symptoms of the entire anxiety disorders spectrum that, according to the DSM-IV, may occur in children.

Previous research of our group (Muris et al., in pressMuris et al., 1998) has provided evidence for the concurrent validity of the SCARED-R. For example, SCARED-R scores were found to be positively related to levels of anxiety as indexed by traditional childhood anxiety measures. More specifically, the SCARED-R total score correlated 0.86 with RCMAS, 0.62 with the FSSC-R and 0.73 with the trait version of the STAIC.

The current study examined the factor structure of the SCARED-R in more detail. Normal school children (N=674) aged between 8 and 13 years completed the questionnaire. The main purpose of the present study was to investigate whether factor analyses would reveal clusters of items that are related to DSM categories in a meaningful way.

Section snippets

Sample

674 children (330 boys and 344 girls) of 8 primary schools in the Maastricht area, The Netherlands, completed the SCARED-R in their classrooms. Ages of the children ranged between 8 and 13 years, with a mean age of 10.28 years (S.D.=1.22). To enhance the representativeness of the sample, schools were selected from urban (n=5) and rural (n=3) areas and from low (n=2), middle (n=5) and high (n=1) social class districts. More than 95% of the children were white. Consent to participate was obtained

General results

Table 1 also shows the mean ratings for the 66 SCARED-R items. As can be seen, the 10 most frequently reported symptoms were “When frightened, my heart beats fast”, “I feel scared when I watch an operation”, “I am nervous”, “I have thoughts that frighten me”, “I don't like to be with people I don't know”, “I worry that bad happens to my parents”, “I am shy with people I don't know”, “I try not to think about a very aversive event I once experienced”, “I don't like being in a hospital” and “I

Discussion

The current study investigated the factor structure of the SCARED-R in a sample of normal school children aged 8–13 years. Results of an exploratory factor analysis suggest that the 66-item SCARED-R primarily has a 1-factor structure. By means of confirmatory factor analysis, several models for the SCARED-R (i.e. 1-factor model, 9-uncorrelated-factors model and 9-correlated-factors model) were tested. However, none of these models was found to be satisfactory. Additional exploratory and

References (25)

  • Bentler, P. M. (1989). EQS: Structural equations program manual. Los Angeles: BMDP Statistical Software...
  • Costello, E. J. & Angold, A. (1995). Epidemiology. In J. S. March (Ed.), Anxiety disorders in children and adolescents....
  • Cited by (191)

    View all citing articles on Scopus
    View full text