Association of Somatosensory Difficulties with Behavioral Problems in Children with Autism Spectrum Disorder

Background:  Somatosensory difficulties manifest as hyper- or hypo-reactivity across the senses and are found in the large majority of children with Autism Spectrum Disorder (ASD). Previous studies have demonstrated a relationship between sensory difficulties and problem behaviors. We sought to expand on these studies by examining multiple problem behavior measures and using a  large cross-sectional sample.

Objectives:  We sought to determine the relationship between somatosensory difficulties and problem behaviors in children with ASD, including investigation of the relation of individual somatosensory domains with specific problem behaviors.

Methods:  This retrospective study used data collected through member centers of the Autism Speaks Autism Treatment Network. Our study included 6,902 children and adolescents under the age of 18 years who had been diagnosed with ASD. We analyzed the Short Sensory Profile (SSP) total score and subscale scores as measures of somatosensory difficulties and their associations with problem behaviors as measured by the total and subscale scores of the Child Behavior Checklist (CBCL) and Aberrant Behavior Checklist (ABC) and the Autism Diagnostic Observation Schedule (ADOS) Tantrums, Aggression, Negative or Disruptive Behavior item., Linear regression and polytomous logistic regression  were  used for continuous and categorical behavioral problems, respectively, adjusting for gender, age, DSM-IV diagnosis, IQ, race, ethnicity, primary caregiver education level, insurance status, any psychological problem status, taking stimulant, and taking SSRI. A Bonferroni correction was applied to account for analyzing multiple behavioral problems variables for each somatosensory domain. 

Results:  The majority of children in the sample were male, Caucasian, and had an IQ of 70 or more. Baseline psychological problems   (ADHD, anxiety, disruptive behavior, other psychiatric problem) were reported in 15.5%. The mean age at informed consent was 6.2 years. After controlling for covariates, there were significant associations between behavioral and somatosensory problems. Children with fewer somatosensory symptoms had fewer behavioral problem symptoms as measured by CBCL and ABC. For each unit increase in SSP Tactile Sensitivity score (=less sensitivity), ABC Irritability score were lower (=less irritable) by 0.76 units. Hyperactivity and total aberrant behavior were higher in children with lower auditory filtering difficulties (ß=-1.11 and -2.94, respectively). Of the sensory difficulties, for each unit reduction in sensation seeking, rule breaking behavior score were lower by 0.05 unit. Lower conduct problems are associated most with higher auditory filtering and sensation seeking (ß=-0.05 for both). Average levels of aggressive behavior were higher among children with higher scores on tactile sensitivity, movement sensitivity, auditory filtering and sensation seeking (ß=-0.06 for all).

Conclusions: Our study demonstrates a strong relationship between somatosensory difficulties and problem behaviors in a cross-sectional sample of children  with  ASD. Our characterization of the relationship between individual sensory and behavioral domains may aid targeting of sensory therapies, with subsequent behavioral problem improvement.