Cluster Analysis of Daily Living Skills in School-Aged Children with ASD

Background:  While daily living skills (DLS) are not a core symptom of individuals with autism spectrum disorder (ASD), they typically fall far below chronological age (e.g., Duncan & Bishop, 2015) and have been linked with factors such as language and cognitive abilities (e.g., Kanne et al., 2011). Studies have also found that internalizing and externalizing behaviors may predict DLS, which may in turn make it difficult to target the acquisition of these skills (De Bildt et al., 2005). It is critical to gain a better understanding of DLS profiles in children with ASD in order to determine how to develop interventions that effectively target these skills while also taking into account autism symptomatology, cognitive and language skills, behavior, and other factors.

Objectives:  The primary aims of the current study were to (1) examine whether there are subgroups of children with ASD defined by both their Vineland-II DLS subdomain raw scores and language abilities and (2) explore the relationship between internalizing and externalizing symptoms on DLS subgroups.

Methods:  The current sample included 959 children with ASD between the ages of 6 to 11 who participated in the initial baseline visit of the Autism Treatment Network, which is a registry collection of children with medical diagnoses of ASD across 14 sites in the United States and Canada. Inclusion criteria for the current study included the following: a completed Autism Diagnostic Observation Schedule (ADOS), a completed Vineland Adaptive Behavior Scales, 2^nd Edition (Vineland-II) including DLS domain and 3 subdomains; a completed Child Behavior Checklist (CBCL); and demographic information on parental education, race, age, and gender. Due to limited data on cognitive abilities, the ADOS was utilized as a measure of functioning level. Specifically, children who received an ADOS Module 1 or 2 were classified as being in the Low Language group and children who received an ADOS Module 3 were classified as being in the High Language group.

Results: A five-cluster solution was found when utilizing age, ADOS module, and Vineland-II DLS domains. An examination of the means of the five clusters, found two groups with Low Language; Low DLS group (n = 143, 14.9%) and High DLS group (n = 170, 17.7%); and three groups with High Language; Low DLS group (n = 189, 19.7%), Medium DLS group (n = 254, 26.4%), and High DLS group (n = 203, 21.3%). The High Language, Low DLS group had significantly higher scores on the CBCL domains as compared to the other 4 groups.

Conclusions:  Within the Low Language group, the Low DLS group has DLS that fall far below what would be expected based on their functioning level. Interestingly, the Low Language, High DLS group had better or equivalent daily living skills as compared to the High Language, Low DLS and High Language, Medium DLS groups. Age equivalent scores on the Vineland-II demonstrated the gap between age and functioning level, even for those with intact language skills (e.g., High Language, Low DLS group had age equivalents ranging from 3-5 years). Additional results and implications for intervention will be discussed.