Social anxiety in adult males with autism spectrum disorders

Highlights

• We investigated self-reported social anxiety symptoms in adult males with ASD.

• 52% of participants (n = 26) endorsed clinically significant social anxiety symptoms including cognitive, behavioural and affective characteristics.

• In our sample, social anxiety symptoms were not associated with present state or childhood ASD symptom-severity, or facets of socio-emotional processing.

• Causal and maintaining mechanisms for social anxiety in ASD now require further investigation.

Abstract

Background

Psychiatric conditions, notably anxiety, commonly co-occur with autism spectrum disorders (ASD).

Method

This study investigated self-reported behavioural, cognitive and affective symptoms of social anxiety (SA) in 50 adult males with ASD. Associations between SA, core ASD symptoms and facets of neuropsychological functioning were also examined.

Results

Twenty-six participants (52%) endorsed levels of SA that exceeded the suggested caseness threshold for social anxiety disorder. Categorical and dimensional data analyses indicated that there were no relationships between SA symptoms, present-state or childhood ASD symptom-severity, or measures of socio-emotional processing in this sample.

Conclusions

Study findings suggest that severity of SA is not merely a reflection of ASD symptom-severity. Further research is needed to ascertain the prevalence of SA in adult ASD epidemiological samples, and identify causal and maintaining mechanisms for these co-morbid symptoms.

Introduction

Psychiatric disorders are frequently and consistently found to co-occur with autism spectrum disorders (ASD) (e.g. Lever and Geurts, 2016, Russell et al., 2016, Simonoff et al., 2008). High rates and levels of social anxiety, in particular, have been reported in children and adolescents with ASD (e.g. Bellini, 2004, Kuusikko et al., 2008; Melfsen, Walita, & Warnke, 2006; Russell & Sofronoff, 2005). Data obtained from self- and informant-report instruments suggest that up to 50% of young people with ASD may score above normative levels for social anxiety, although ratings from different informants do not always correlate significantly (Bellini, 2004).

Relatively little is known about social anxiety disorder (SAD) in adults with ASD, despite this being the most common anxiety disorder in the typically developing adult population, with high rates of co-morbid depression, other anxiety disorders, substance use, and increased risk of suicide (NICE, 2013). Cross-sectional studies that have examined general rates of psychiatric co-morbidity in adults with ASD, recruited via community (n = 172, Lever & Geurts, 2016) and clinical settings (n = 122, Hofvander et al., 2009; n = 63, Joshi et al., 2013; n = 474, Russell et al., 2016), have estimated that between 12% and 56% of adults meet diagnostic criteria for SAD. Three studies to date, have focused specifically on SAD in adults with ASD. Cath, Ran, Smit, van Balkom, and Comijs (2008) examined similarities and differences in self-reported SAD, obsessive compulsive disorder (OCD), and affective symptoms in 12 adults with ASD, compared to matched clinical and non-clinical controls. Participants completed several questionnaires including the Liebowitz Social Anxiety Scale, one of the most widely used self-report social anxiety measures (LSAS: Liebowitz, 1987). Comparable levels of anxiety were found in the SAD, and ASD and SAD groups. Bejerot, Eriksson, and Mortberg (2014) found that 28% of adults with ASD (n = 14 of 50) met the criteria for SAD using the clinician-administered MINI International Neuropsychiatric Interview (M.I.N.I.: Sheehan et al., 1998), as well as the LSAS. Finally, Maddox and White (2015) investigated SAD in three adult samples; individuals with ASD (n = 28), individuals with SAD but no ASD (n = 26), and non-clinical controls (n = 25). Using self-report questionnaires and an objective assessment of anxiety, their findings indicated that 50% of individuals with ASD presented with clinically significant SAD as measured by the Anxiety Disorders Interview Schedule (ADIS-IV: Brown, DiNardo, & Barlow, 1994), and the Social Interaction Anxiety Scale (SIAS: Mattick & Clarke, 1998). By contrast, there were no differences between the ASD and ASD + SAD groups on the Brief Fear of Negative Evaluation Scale (Brief FNE: Leary, 1983).

The notion of co-morbid social anxiety in ASD is, however, inherently complex in several respects. First, there is a clear overlap between the symptom profiles of these two disorders (White et al., 2012). ASD is characterised, for example, by qualitative impairments in reciprocal social interaction (WHO, 1992), while hallmark features of SAD also include difficulties with initiating and maintaining interactions and conversations, as well as social avoidance. Second, similar impairments in neuropsychological functioning have been observed in individuals with ASD and those with SAD, such as emotion and face processing deficits (Brunsdon and Happé, 2014, Morrison and Heimberg, 2013; Wong, Beidel, Sarver, & Sims, 2012); again rendering it difficult to demarcate one disorder from the other. Third, both conditions can impair and restrict attainment and independence; symptoms typically affect peer and social relationships, schooling, and employment.

Assessment of SAD in individuals with ASD poses challenges (Kreiser & White, 2014). Individuals with ASD and/or their significant others (e.g. family members) may not spontaneously seek assessment for social avoidance or social evaluative worries, as these characteristics may be attributed to the core disorder. Even when individuals do present to services, impairments in introspection due to theory of mind deficits (Williams & Happé, 2010), or alexithymia (difficulties labelling own emotions, Bird, Press, & Richardson, 2011) can render it difficult for them to describe physical and cognitive symptoms of anxiety. Further, while some studies suggest that individuals with ASD are able to self-report psychopathology symptoms (e.g. Berthoz and Hill, 2005, Cadman et al., 2015), commonly used social anxiety measures are yet to be validated for the ASD population. Use of multiple measures that focus on a range of behavioural, cognitive and affective characteristics associated with social anxiety may therefore enhance the screening and assessment process (Kreiser and White, 2014, Maddox and White, 2015, Tyson and Cruess, 2012).

Perhaps as a result of these issues, the relationship between ASD and SAD has seldom been explored. As in typically developing populations, psycho-social factors, including adverse social experiences, cognitive processes such as information and attentional biases, and safety behaviours such as social withdrawal and avoidance, are likely implicated as risk, causal and/or maintaining mechanisms (see Clark, 1999, Morrison and Heimberg, 2013). However, it is also plausible that there are ASD-specific factors that serve to increase vulnerability for, and perpetuate, SAD. For example, it may be that core ASD characteristics, such as deficits in social skills, and/or difficulties with engaging reciprocally in social interaction, contribute to anxiety about social situations (e.g. Bellini, 2004, Tyson and Cruess, 2012; White, Oswald, Ollendick, & Scahil, 2009). Similarly, an intolerance of uncertainty (IoU), or hypo- and hyper-sensory sensitivities, have been found to be associated with anxiety symptoms (Boulter, Freeston, South, and Rodgers, 2014; Maisel et al., 2016; Wigham, Rodgers, South, McConachie, & Freeston, 2015) and these may encourage avoidance of social situations, e.g. because these seem unpredictable or overly stimulating. Additionally, facets of neuropsychological functioning (such as impairments in socio-emotional processing) could be implicated in anxiety development in ASD (White et al., 2009), for example, impairments in the ability to recognise and understand others’ thoughts and intentions (Baron-Cohen, Wheelwright, Skinner, Martin, & Clubley, 2001), may render social interactions difficult. Finally, poor peer relationships, rejection, and bullying, all of which occur often and repeatedly for young people and adults with ASD (Schroeder et al., 2014), may mean this population is susceptible to developing social evaluative concerns around difference, inferiority, and vulnerability, as well as encouraging social withdrawal, isolation and avoidance.

In summary, research findings indicate high rates and levels of social anxiety in children and young people with ASD, as measured by self- or informant-based instruments. Few studies have explored the frequency or nature of social anxiety in the adult ASD population, particularly adults who do not have a concurrent intellectual disability, and who are potentially more likely to need to face anxiety-provoking situations in the context of employment or independent living tasks. Also, prevalence estimates have varied widely, which may be due to differences in study sampling frames and selection criteria, inclusion/exclusion of individuals with heterogeneous ASD presentations, and assessment of SAD using different measures, not all of which rate cognitive, affective and behavioural characteristics associated with social anxiety. Despite the difficulties with assessing and diagnosing SAD in ASD, there is a clear need to better understand if and why these symptoms might co-occur in order to aid early identification of need, and the development of evidence-based treatments.

The aims of the present study were therefore as follows: (1) to explore the frequency and range of self-reported social anxiety symptoms in a sample of adult males with ASD and no intellectual impairment; (2) to examine the relationship between data from multiple self-report social anxiety questionnaires commonly used in clinical/research fields; (3) to investigate the relationship between anxiety symptoms and ASD symptom-severity given that core impairments may be associated with the development of anxiety; and (4) to examine facets of socio-emotional processing in relation to social anxiety. We hypothesised there would be high rates of self-reported social anxiety symptoms, and that there would be associations between social anxiety, ASD symptom-severity, and socio-emotional processing.