Psychometric properties of the child and parent versions of Spence Children's Anxiety Scale in a Danish community and clinical sample

Highlights

• Supportive findings for the psychometric properties of SCAS and SCAS-P.

• First test–retest reliability reported for SCAS-P.

• Danish norms provided for SCAS and SCAS-P split on gender, age groups and sample.

Abstract

This study examined the psychometric properties and norms of the Spence Children's Anxiety Scale (SCAS) and the associated parent version (SCAS-P) in a Danish community and a clinical sample. The total sample consisted of 1240 children (972 from community sample), age 7–17 years, and 805 parents (537 from community sample). Results indicated that SCAS and SCAS-P had good internal consistency on the total scale and all subscales, with exception of the subscale for fear of physical injury. Both scales showed satisfactory 2-week and 3-month retest stability. All subscales and total scales of the SCAS and SCAS-P discriminated between the clinical and community sample. A comparison with the Beck Youth Inventories and the Strength and Difficulty Questionnaire in the clinical sample supported the scales’ convergent and divergent validity. Results of confirmatory factor analyses for SCAS and SCAS-P were in favor of the original model with six correlated factors.

Introduction

Anxiety disorders are among the most common psychiatric disorders in children and adolescents, and a recent meta-analysis found the mean estimate of prevalence for any anxiety disorder to be 12.3% in children (age 6–12) and 11.0% in adolescents (age 13–18) (Costello, Egger, Copeland, Erkanli, & Angold, 2011). Although prevalent, anxiety disorders in youth are often overlooked (Chavira, Stein, Bailey, & Stein, 2004) and have been associated with social and academic impairment (Essau et al., 2000, Strauss et al., 1987) as well as increased risk of suicidal behavior in adolescents (Boden, Fergusson, & Horwood, 2007). If left untreated a great number of children and adolescents carry their anxiety disorder on into adulthood (Keller et al., 1992), where it has been found to be a precursor of comorbid depression (Roza, Hofstra, van der Ende, & Verhulst, 2003), substance abuse (Costello, Mustillo, Erkanli, Keeler, & Angold, 2003), and other anxiety disorders (Kim-Cohen et al., 2003, Pine et al., 1998). Thus, early identification of childhood anxiety disorders is important.

A range of questionnaires have been developed to assess anxiety symptoms in children and adolescents (Silverman & Ollendick, 2005). Early developed questionnaires on childhood anxiety, like the Revised Children's Manifest Anxiety Scale (RCMAS; [Reynolds & Richmond, 1985]) and Fear Survey Schedule for Children – Revised (FSSC-R; [Ollendick, 1983]), were developed by downward extensions of questionnaires designed for adults. Such measures provide information on anxiety symptoms in general, rather than specific symptoms of different anxiety disorders. Spence Children's Anxiety Scale (SCAS) is a more recently developed questionnaire specifically designed to measure anxiety symptoms for different DSM-IV anxiety disorders among children and adolescents. The psychometric properties of the child self-report version of SCAS have been examined in a variety of cultures and languages including Australia (Spence, 1998), the United States (Whiteside & Brown, 2008), Germany (Essau, Muris, & Ederer, 2002), Netherlands (Muris, Schmidt, & Merckelbach, 2000), the United Kingdom (Essau, Sasagawa, Anastassiou-Hadjicharalambous, Guzmán, & Ollendick, 2011), Belgium (Muris, Merckelbach, Ollendick, King, & Bogie, 2002), Greece (Mellon & Moutavelis, 2007), Sweden (Essau, Sasagawa, et al., 2011), Spain (Orgilés et al., 2012, Tortella-Feliu et al., 2005), Italy (Di Riso, Chessa, Bobbio, & Lis, 2013), Cyprus (Essau, Anastassiou-Hadjicharalambous, Muñoz, 2011), Iran (Essau, Olaya, Pasha, O’Callaghan, & Bray, 2012), Colombia (Amaya & Campbell, 2010), Mexico (Hernández-Guzmán et al., 2010), Japan (Ishikawa, Sato, & Sasagawa, 2009), Mainland China (Zhao, Xing, & Wang, 2012), Hong Kong, China (Li, Lau, & Au, 2011), and South Africa (Muris, Schmidt, Engelbrecht, & Perold, 2002).

The parent version, SCAS-P has been examined to a lesser extent with studies from Australia (Nauta et al., 2004), North America (Whiteside & Brown, 2008), Turkey (Orbay & Ayvaşik, 2006), Hong Kong, China (Li et al., 2011), and Japan (Ishikawa et al., 2013). The majority of research on the SCAS and SCAS-P has been conducted in community samples with the exception of only three studies that also included clinical samples (Nauta et al., 2004, Spence, 1998, Whiteside and Brown, 2008).

Cross-cultural differences on mean scores and psychometric properties of SCAS have been found in a number of studies. For example, German children reported significantly higher symptoms of social phobia, separation anxiety, OCD, and generalized anxiety disorder compared to Japanese children, who conversely reported significantly higher scores on items related to fear of physical injury (Essau, Sakano, Ishikawa, & Sasagawa, 2004). Another study found that Western children (Dutch) reported significantly lower symptoms across all anxiety disorders compared to South African children (Muris et al., 2002a, Muris et al., 2002b). However, a study by Mellon and Moutavelis (2007) found that Greek children reported overall anxiety symptoms comparable to the scores of the South African children in Muris, Merckelbach, et al. (2002) and Muris, Schmidt, et al. (2002). Differences in mean scores have not only been found across different continents but also between countries within Western Europe, where adolescents in the United Kingdom reported significantly higher anxiety scores compared to adolescents from Sweden, Germany, Italy, and Cyprus (Essau et al., 2011a, Essau et al., 2011b). Similarly, differences between countries have also been found for SCAS-P, where Chinese parents (Li et al., 2011) had higher mean scores compared to norms from Australia (Nauta et al., 2004), North America (Whiteside & Brown, 2008), and Japan (Ishikawa et al., 2013). Such differences in mean scores have been attributed to differences in socio-economic status (Muris et al., 2002a, Muris et al., 2002b) or cultural differences in parental rearing behavior (Essau et al., 2011a, Essau et al., 2011b) educational practices (Essau et al., 2004), or self-disclosure of emotional states (Mellon & Moutavelis, 2007). Although such post hoc explanations may often be plausible, very few studies have tested pre-study hypotheses about cultural differences (Al-Issa & Oudji, 1998).

Nevertheless, cross-cultural differences in mean scores emphasize the need to establish norms for specific countries. Furthermore, while a large number of studies have been conducted on SCAS and SCAS-P, the investigation of some of their psychometric properties has been neglected. For example, most studies do not assess the test–retest reliability of SCAS, and to date no prior study has examined the test–retest reliability of SCAS-P. Also, studies on the discriminant validity of SCAS and SCAS-P are scarce, as only three studies have included a community and an anxiety disorder sample. Thus, the aims of the present study were to make a comprehensive assessment of the psychometric properties and provide norms for the Danish versions of SCAS and SCAS-P using community and anxiety disorder samples.