Analyzing How Autism Severity Affects Motor and Social Skills: An Exploration Using the SFARI Base Dataset

Background: Social skill deficits and repetitive behaviors are the defining features evaluated during autism spectrum disorder (ASD) diagnostic assessments. Assessment tools, such as the Autism Diagnostic Observation Schedule, 2^ndEdition (ADOS-2; Lord, Rutter, Di Lavore, & Risi, 2002; Lord et al., 1989), evaluate the level of impairment within these two characteristics to provide an understanding of the extent of the disorder. In addition to the typical deficits associated with ASD, a growing body of research suggests that gross motor deficits are also present in most children with ASD (Liu & Breslin, 2013; Lloyd, MacDonald, & Lord, 2013; Staples & Reid, 2010). However, little is known about the effect of severity has on social and gross motor skills in children with ASD.  

Objectives: The purposes of this study are to determine: (1) What effect does the level of severity have on the motor skills and social skills in children with ASD?, and (2) If a difference exists between severity groups, which dependent variable has the greatest effect?

Methods: Utilizing a sub-sample from the SFARI base dataset, a MANOVA was used to understand differences among the multivariate means of social skills and gross motor skills of 483 children with autism and autism spectrum disorder. The present sample was divided into two groups based on ADOS-2 severity output: (a) those diagnosed with autism (i.e., more severe, N = 444), and (b) those diagnosed with autism-spectrum (less severe, N = 39). Average age of the participants was 66.27 months or about 5.6 years of age. Mean for participants diagnosed with autism was 66.48, and mean for participants diagnosed with autism-spectrum was 63.92 (group differences were non-significant, t₍₄₈₁₎=1.152, p = 0.250).  

Results:  Wilk’s criterion indicated the combined dependent variables were non-significantly affected by severity of the autism diagnosis, multivariate F_{(2, 480)} = 1.640, p = 0.195. Analysis of the between-subject effects univariate test further revealed a non-significant difference between the groups on gross motor scores, F_{(1, 201.883)} = 1.303, p = 0.254. Participants in the less severe group (autism-spectrum) scored arithmetically higher (M = 83.31) than those in the more severe (autism) group (M = 80.93). This non-significant difference was also seen in the social skill univariate evaluation, F_{(1, 4535.986)} = 2.242, p = 0.135. While the less severe group (M = 70.62) again scored higher in social skills than the more severe group (M = 59.37), as would be expected, differences were non-significant. Effect sizes further demonstrate a weak association between the gross motor and social skill of the groups, partial η² = 0.003 and partial η² = 0.005 respectively.

Conclusions: Results suggest that little difference exists between severity groups on gross motor and social skills; however, results may be limited to the age range of the participants (about 5.6 years of age). Differences in groups may become greater as children age. This analysis suggests that at a young age (about 5 years of age) there is little difference between severity groups within ASD and the severity of gross motor and social skills.