Measuring Repetitive Behaviors In Toddlers with ASD

Background: Repetitive and stereotyped behaviors (RSB) are hallmark symptoms of autism spectrum disorder (ASD), yet have received little research attention in early childhood.  Some studies have suggested that RSB were not pronounced in toddlers and were later developing symptoms (Cox et al., 1999; Charman et al., 2005).  Other studies reported that RSB distinguish toddlers with ASD from toddlers with typical development and developmental delays (Richler, 2007; Wetherby et al., 2004; Watt et al., 2008).  These discrepancies suggest that there is a need to identify an instrument that is sensitive to RSB in toddlers with ASD.  Further, the relationship between RSB and developmental level in the first years of life remains unclear.

Objectives:  

1. To examine whether an instrument developed to assess repetitive behaviors in typical toddlers and preschoolers (i.e., the Childhood Routines Inventory; 1997) is sensitive to repetitive behaviors in toddlers with ASD. 
2. To explore the relationship between RSB and developmental level in toddlers with ASD.

Methods: Thus far, data is available for 13 toddlers between 19 and 34 months of age (mean = 29.15 months) diagnosed with ASD through The University of Alabama ASD Clinic.  ASD was diagnosed using the ADOS and ADI-R.  Developmental level was assessed using the Mullen Scales of Early Learning.  Two caregiver-report RSB measures were administered, The Childhood Routine Inventory (CRI; Evans et al., 1997) and the Repetitive Behavior Scales Revised (RBS-R; Bodfish et al., 1999; Lam & Aman, 2007).  RSB-R includes 6 subscales measuring both repetitive motor movements and compulsive behavior.  The RBS-R was developed to measure RSB in children with ASD but has not been used with toddlers.  The CRI has 2 subscales that measure compulsive behavior.  The CRI was developed specifically for a toddler and preschool population. 

Results: Compared to normative means provided on the CRI, every toddler with ASD showed repetitive behaviors greater than the 95^th percentile.  Specifically, toddlers with ASD scored four times higher on the CRI than toddlers with typical development in both the 12-23 and 24-35 month normative intervals.  Further, children’s total scores from the CRI and RBS-R were significantly correlated (r = .82; p = .001) such that high scores on one measure was related to high scores on the other.  Finally, total scores on the RBS-R and the CRI were negatively correlated with the Mullen Early Learning Composite (RBS-R; r = -.59; p = .04 and CRI; r = -.70; p <. 01).  Children with high levels of RSB received lower standard scores on all Mullen subscales (visual motor, language, and fine motor).

Conclusions: Results suggest that toddlers with ASD do engage in high levels of RSB.  Further, both the RBS-R and the CRI appear to be useful tools for identifying RSB in toddlers with ASD.  Findings regarding the CRI are particularly exciting as this measure is sensitive to RSB in typical children.  The CRI may be useful in research comparing diagnostic groups or research examining sub-threshold symptoms in at-risk toddlers.  Further research is ongoing to identify specific patterns of RSB in toddlers with ASD.