RUNNING HEAD: FACTORS ASSOCIATED WITH ANXIETY IN CHILDREN WITH ASD
Anxiety Symptoms in Young People with Autism Spectrum Disorder
Attending Special Schools: Associations with Gender, Adaptive Functioning and Autism Symptomatology
Abstract
Anxiety related problems are among the most frequently reported mental health difficulties in Autism Spectrum Disorder (ASD). As most research has focused on clinical samples or high-functioning children with ASD, less is known about the factors associated with anxiety in community samples across the ability range. This cross-sectional study examined the association of gender, age, adaptive functioning and autism symptom severity with different groups of caregiver-reported anxiety symptoms. Participants were caregivers of 241 6-18 year old children with ASD attending special schools in Singapore. Measures included the Spence Children’s Anxiety Scale and assessments of overall emotional, behavioural and adaptive functioning. Caregivers reported more anxiety symptoms in total, but fewer social anxiety symptoms, than SCAS Australian/ Dutch norms. There were no gender differences. Variance in total anxiety scores was best explained by severity of repetitive speech/ stereotyped behaviour symptoms, followed by adaptive functioning. Severity of repetitive speech/ behaviour symptoms was a significant predictor of separation anxiety, generalized anxiety, panic/ agoraphobia and obsessive-compulsive subscale symptoms, but not of social phobia and physical injury fears. Adaptive functioning and chronological age predicted social phobia and generalized anxiety symptoms only. Severity of social/ communication autism symptoms did not explain any anxiety symptoms, when the other variables were controlled for. Findings are discussed in relation to existing literature. Limitations and possible implications for prevention, assessment, and intervention are also discussed.
Keywords
autism spectrum disorder, anxiety, children, adolescents, correlates, predictors
Anxiety Symptoms in Young People with Autism Spectrum Disorder
Attending Special Schools: Associations with Gender, Adaptive Functioning and Autism Symptomatology.
Introduction
Although core social, communication and behavioural impairments in Autism Spectrum Disorder (ASD) have been extensively studied (e.g. Kuenssberg et al., 2011; Mazurek et al., 2012), co-occurring psychopathology and mental health problems have, until recently, been less well researched (for a recent review see Mannion et al., 2014). Anxiety-related concerns are among the most common mental health problems in ASD (Ghaziuddin, 2002; Wood and Gadow, 2010). Depending on the informant, methodology or sample employed, between 11 and 84% of children with ASD experience anxiety difficulties (White et al., 2009) and up to 40% are diagnosed with at least one DSM anxiety disorder at some point in their lives (van Steensel et al., 2011), compared with rates of 3-24% in typically developing children (i.e. Cartwright-Hatton et al., 2006).
It is suggested that ASD impairments may, to some extent, contribute to increased stress due to perspective-taking and communication difficulties, limited flexibility and sensory sensitivities (Wood and Gadow, 2010; Kerns and Kendall, 2012). In turn, increases in stereotyped and challenging behaviours could be a consequence of efforts to reduce high levels of anxiety-related arousal (see also Hallett et al., 2013; Oszivadjian et al., 2012). Although these hypotheses remain to be adequately tested in prospective longitudinal studies, many different factors likely influence anxiety in this population. However, data on rates and types of anxiety disorders in ASD have largely been derived from clinical samples of individuals with ASD without associated intellectual disability (e.g. White et al., 2015; Kuusikko et al., 2008; Renno and Wood, 2013; see MacNeil et al., 2009 and van Steensel et al., 2011 for reviews), thus limiting our understanding of the possible influences of factors such as level of intellectual or adaptive functioning or autism severity. In the section below, we examine existing literature on the association between child characteristics and anxiety in ASD, with a specific focus on the relationships between child characteristics and different anxiety subtypes (i.e. social, generalized, OCD etc.).
Factors Associated with Anxiety in ASD
Gender. In most of the published research on anxiety in ASD, it has not been possible to examine gender differences, as too few females have been involved. Amongst the few studies that have explored this issue, no gender differences have been consistently identified (see Kirkovski et al., 2013 for review). Moreover, very few, if any, studies have explored the possible relationship between gender and different forms of anxiety symptoms.
Chronological age. Some studies have reported more anxiety symptoms in older as compared to younger children with ASD (Lecavalier, 2006; Kuusikko et al., 2008; Mayes et al., 2011; Vasa et al., 2013); others have not identified any age effects (Hallett et al., 2013; Sukhodolsky et al., 2008; Farrugia and Hudson, 2006; Pearson et al, 2006). However, it is possible that the relationship with age may depend on the types of anxiety presentations studied. For example, in their meta-analysis, van Steensel and colleagues (2011) noted that overall anxiety and Generalized Anxiety Disorder (GAD) rates were higher among older children with ASD; Separation Anxiety Disorder and OCD were more common in younger children.
Intellectual/ adaptive functioning level. Although significant positive associations between intellectual ability/ adaptive functioning and increased anxiety symptoms have been reported in a number of studies (Lecavalier, 2006; Hallett et al., 2013) or IQ (Davis III et al., 2008; Mayes et al., 2011; Niditch et al., 2012; Hallett et al., 2013), others have identified no consistent relationships (i.e. Brereton et al., 2006; Simonoff et al., 2008). Kerns and Kendall (2012) proposed that intellectual abilities may be associated with specific types of anxiety symptoms in ASD but, to our knowledge, Sukhodolsky et al. (2008) and Hallett et al. (2013) are the only groups to have examined the relationship between cognitive/adaptive functioning and different types of anxiety symptoms. Sukhodolsky and colleagues found that youth with IQ>70 were more likely to present with clinically elevated generalized, separation and somatization anxiety symptoms than their peers with IQ<70, but equally likely to present with elevated simple phobia, panic and social phobia anxiety symptoms. Hallett et al. (2013) reported a positive association of IQ with parent-rated social anxiety only, but no relationship with generalized, separation, panic or OCD symptoms.
ASD symptom severity. It has been suggested that children with greater ASD symptomatology may be more vulnerable to anxiety related difficulties (Pearson et al., 2006; Wood and Gadow, 2010). However, Renno and Wood (2013) found that severity of anxiety was generally independent of ASD severity in a clinically referred cohort. Similarly, Simonoff et al. (2008) found no association between ASD symptom severity and the presence of an anxiety disorder in their population sample. Pearson and colleagues (2006) also reported non-significant associations, but their findings were limited by the study’s retrospective design and the use of a general emotional adjustment questionnaire. In contrast, Sukhodolsky and colleagues (2008) found a positive relationship between autistic stereotyped behaviours and parent-reported anxiety and Mayes et al. (2011) found maternal ratings of autism severity to be the single best predictor of anxiety. In the only longitudinal study of anxiety in ASD to date, Baird and colleagues (2012) reported that earlier stereotyped behaviours and interests predicted anxiety 12 months later in pre-school children with ASD over and above other variables, such as IQ, gender or earlier social and communication impairments. Moreover, most researchers have used a total autism symptom severity score, despite the considerable heterogeneity in, and likely independence between, social/ communication and repetitive/ stereotyped autism symptomatology (see Hus, Gotham and Lord, 2014; Mundy and Skuse, 2008). To our knowledge, only Hallett and colleagues (2013) have explored the association between different clusters of autism symptoms and different subtypes of parent-reported anxiety symptoms; they found that children with more severe social and communication impairments based on clinician-rated ADI-R scores had more separation, but fewer social, anxiety symptoms, while higher ADI-R restricted and repetitive behavior scores (indicating more impairments) were associated with increased panic and OCD symptoms.
The present study: rationale, aims and research questions
Data on the association between anxiety in ASD and child characteristics, such as age, gender, cognitive level, and autism severity remain inconsistent and inconclusive Assessments of anxiety are often based on a very small number of items (i.e. Mayes et al., 2011; Davis III et al., 2011) or on brief anxiety subscale scores derived from broader measures of emotional and behavioural problems (i.e. Vasa et al., 2014; Rieske et al., 2012; Eussen et al., 2013; Niditch et al., 2012) and few studies have explored the relationship between child characteristics and different subtypes of anxiety symptoms. Assessments of autism symptom severity are usually based on total symptom scores, with no differentiation between domains, and there has been little attempt to explore these relationships in non-referred samples. Finally, almost all research in this area has been undertaken in Europe, the US and Australia. Among typically developing children, cultural differences have been consistently reported in internalizing symptoms, including anxiety (Anderson and Mayes, 2010), but there are currently extremely few studies of anxiety in other ethnic groups of children with ASD (i.e. Ooi et al., 2011; 2014).
The present study aimed to contribute to the existing literature by (i) recruiting non-clinically referred participants from specialist schools across the age and ability range; (ii) investigating the independent and joint contribution of a number of child variables (gender, age, adaptive functioning, social/ communication and repetitive speech/ behaviour autism symptoms) in predicting overall anxiety as well as specific anxiety subtypes; (iii) examining the shared and independent role of social/ communication versus repetitive speech/ behaviour autism symptoms in predicting anxiety; and (iv) extending the generalizability of existing literature by recruiting a large, ethnically diverse sample of Asian youth with ASD.
Based on existing literature, we proposed the following hypotheses:
(i) Age and adaptive functioning will be positively associated with anxiety symptoms; there will be no gender differences in anxiety symptoms.
(ii) The relationships between anxiety symptoms and child characteristics will differ according to the anxiety subtypes examined; we made no specific hypotheses about the direction or strength of these relationships, in view of the lack of relevant literature.
(iii) Repetitive speech/ behaviour autism-related symptoms will likely be a stronger predictor of anxiety symptoms than social/ communication autism symptoms.
Methods
Participants
Inclusion criteria. Child criteria were: (a) chronological age 6 to 18 years; (b) clinical diagnosis of autism, ASD, Asperger Syndrome (AS) or Pervasive Developmental Disorder, Not Otherwise Specified (PDD-NOS) made by a qualified medical or mental health professional using DSM-IV or ICD-10 criteria[1]; and (c) attending special schools in Singapore.
Participant and informant characteristics. Caregivers of 241 children meeting inclusion criteria participated (181 mothers, 54 fathers, four grandparents, two not identified); mean age=42.7 years (SD 5.9 years); 48.5% were University educated, 35.7% had a General Certificate of Education [GCE] or other diploma; 14.9% completed primary/ secondary school; 0.8% did not report; caregivers in this sample were generally educated at a higher level compared with the Singaporean population; see Table 1).
INSERT TABLE 1 ABOUT HERE
Diagnosis and recruitment. Diagnosis (see Table 1) was determined by caregiver information on: diagnosis given; diagnostic setting/ organization involved; professional(s) who made the diagnosis. Most participants (85%) were diagnosed in one of the three leading public diagnostic clinics in the country. These are staffed by qualified multi-disciplinary teams employing evidence-based approaches to diagnosing ASD using DSM-IV or ICD-10 criteria (AMS-MOH Clinical Practice Guidelines, 2010; Moh and Magiati, 2012).
All participants were recruited from Singapore’s special schools; special education in Singapore is provided only to children with a formal, clinical diagnosis of intellectual disability (IQ and/or adaptive skills standard scores ≤ 70) or to those with a professional diagnosis of ASD. Participants were recruited from six special schools[2]. Of these, three (N=39) accept children with ASD and mild intellectual disabilities (IQ 50-70); one (N=132) specifically caters for 7-16 year old children with a diagnosis of ASD and a non-verbal IQ score of >70 who are considered to be cognitively able to access the mainstream curriculum in a supportive educational setting; the other two (N=61) cater for children with multiple disabilities, including ASD (9 participants did not state which school their child attended).
Measures
Spence Children’s Anxiety Scale-Parent version (SCAS-P; Spence, 1999). This is a 38-item caregiver-rated measure of anxiety symptoms (higher scores = > anxiety). It provides a Total (range 0-114) and six DSM-IV based subscale scores for generalized anxiety (range 0-18); social phobia (0-18); separation anxiety (0-18); panic attack and agoraphobia (0-27); Obsessive Compulsive Disorder (OCD; 0-18); and physical injury fears (0-15). SCAS-P Australian/ Dutch norms have been established by Nauta and colleagues (2004) based on a sample of 261 parents (45% Dutch) of 6-18 year old children (mean age=11.5 years, SD=2; 48% boys) from rural and urban schools. A cut-off of one standard deviation above the normative mean is being used to indicate clinically elevated anxiety symptoms (Spence, personal communication, October 2012). The SCAS-P has strong psychometric properties in Australia and Hong Kong Chinese community samples (Nauta et al., 2004; Li et al., 2011); it has been frequently used in studies of children with ASD (see Grondhuis and Aman, 2012 for review), and good construct, convergent and discriminant validity and moderate caregiver-child agreement have been reported with youth with ASD (Russell and Sofronoff, 2005; Zainal et al., 2014; Magiati et al., 2014). However, there are no normative data for the SCAS-P currently in Singapore.
Developmental Behaviour Checklist, Parent version, Second Edition (DBC-P; Einfeld and Tonge, 2002). The DBC-P is a 96-item checklist of emotional and behavioural problems for 4-to-18-year old children with developmental and intellectual disabilities. It provides a Total Behaviour Problem Score (TBPS; range 0-192; Cronbach’s α=.94 in this study) and six subscale scores (disruptive/ antisocial; self-absorbed; communication disturbance; anxiety; social relating; higher scores = > problems). The 9-item DBC-P Anxiety subscale (raw score range: 0-18; α=.66 in DBC manual and .67 in this study) was used as an additional measure of anxiety symptoms. It correlates highly with the State-Trait Anxiety Inventory for Children (r=.61; STAI-C; Spielberger et al., 1983).
A factor analysis derived Autism Screening Algorithm (DBC-ASA) comprising 29 DBC items discriminates well between children with intellectual/ developmental disabilities with and without autism with a cut-off of 14 (Steinhausen and Metzke, 2004) or 17 (Brereton et al., 2002). In the present sample, 68% obtained a score ≥14; when only participants with adaptive functioning standard scores < 70 were considered, this increased to 80%.