The Relationship between Autistic Traits, Eating Disordered Behaviours and Anxiety in the General Population

Background:

The high co-occurrence of autism and anorexia nervosa is well established (e.g. Huke et al., 2013), with higher levels of autistic traits commonly found in anorexic populations (Westwood et al., 2016). Further, general population studies show significant positive correlations between autistic traits and eating disorder traits relating to both anorexia nervosa and bulimia nervosa (e.g. Coombs et al., 2011).

However, studies investigating traits have predominantly relied on the Autism Spectrum Quotient (AQ; Baron-Cohen et al., 2001), which has not enabled specific examination of the impact of restricted and repetitive behaviours (RRBs) on eating disordered traits. In addition, anxiety is highly co-occurring with both autism and eating disorders (Kim et al., 2000; Swinbourne & Touyz, 2007) but little is known about how anxiety may influence the shared presentation of both autistic and eating disorder traits. Further, whether cognitive anxiety (i.e. related to thoughts such as worry and indecisiveness) or somatic anxiety (i.e. related to physical symptoms such as dizziness and shaking) is more relevant also remains unexplored.

Objectives:

1. To explore the specific associations between social communication difficulties and RRBs and eating disorder traits in the general population.
2. To additionally examine the effect of cognitive and somatic anxiety on these associations.

Methods:

Three-hundred and forty five adults (87% female; mean age 29.7 years, SD = 13.3) completed four online questionnaires: the AQ, the Adult Repetitive Behaviour Questionnaire-2 (RBQ-2A; Barrett et al., 2015; 2018), the Eating Attitudes Test-26 (EAT-26; Garner et al., 1982), and the State-Trait Inventory for Cognitive and Somatic Anxiety (STICSA; Grös et al., 2007). Mediation analyses were used to explore the mediating effects of cognitive and somatic anxiety on the relationship between autistic traits (social and communication difficulties, repetitive sensory motor behaviours (RSMB) and insistence on sameness (IS)) and three types of eating disordered behaviour (dieting, bulimia and food preoccupation, and oral control).

Results:

Preliminary analyses show that social and communication difficulties, RSMBs and IS all indirectly influenced dieting behaviour through their effect on cognitive anxiety. Similarly, social and communication difficulties, RSMBs and IS all indirectly influenced bulimia and food preoccupation behaviours through their effect of cognitive anxiety. A direct effect of social communication difficulties on bulimia and food preoccupation was also identified. There were no indirect pathways between autistic behaviours and oral control, but there was a direct pathway between IS and oral control.

Conclusions:

Using a general population sample, we found: 1) Social communication and RRBs independently contribute to the presence of eating disorder behaviours; 2) Anxiety generally has an important role in relationship between autistic and eating disorder behaviours. The data suggest the importance of considering the full spectrum autistic traits in those with high levels of eating disorder behaviour. Further, anxiety may be important clinical consideration for treatment approaches.