A Parent-Report Measure of Restricted and Repetitive Behaviors for Early Childhood

Background: Restricted and repetitive behaviors (RRBs) are defining features of autism spectrum disorder (ASD) which emerge early in life and may differentiate children who do and do not develop the disorder during toddlerhood (Wolff et al., 2014; Elison et al. 2014). These behaviors also occur as part of typical early development, supporting the acquisition of more complex and goal-directed behavior (Evans et al., 1997; Thelen 1979). Existing paper and pencil measures of RRBs are largely anchored in pathological features seen in older children and adults. As a result, they and are not intended to capture individual differences in restricted and repetitive behaviors among young children.

Objectives: Our goals were twofold: 1) To adapt the Repetitive Behavior Scale, Revised (RBS-R; Bodfish et al., 2000) for use among typically and atypically developing children during early childhood; and 2) To collect pilot data on the measure from a large cohort of typically developing children.

Methods: We developed the Repetitive Behavior Scale for Early Childhood (RBS-EC), a brief (~10 minute) parent report measure consisting of 34 items. The measure is comprised of 4 subscales constituting categories of repetitive behavior congruent with existing measures and theoretical constructs: repetitive motor, self-directed/self-injurious, restricted, and ritual/routine. Unlike most existing measures, the RBS-EC focuses on frequencies (rather than severity) of behavior and yields both subscale and total scores for frequency and behaviors endorsed. Content validation was performed through feedback from outside experts in repetitive behavior and child development. Initial beta testing and feedback was obtained from parents of both typically and atypically developing children (infants through early school age). We performed preliminary psychometric analyses of 330 data sets collected from parents of 18 to 25 month-old children.  The data were ascertained from a quasi-epidemiological sample recruited from the Institute of Child Development’s research registry as part of a larger study investigating quantitative phenotypes in young children.

Results: The distribution of total repetitive behavior endorsed scores is presented in Figure 1. Internal consistency for the total RBS-EC was excellent (α = 0.91). For subscales, internal consistency ranged from adequate to excellent (repetitive motor: α = 0.94; self-injurious: 0.74; restricted: 0.81; ritualistic: 0.78). Test-retest reliability for total and subscale scores were in the excellent range (ICC > 0.90).

Conclusions: Initial results suggest that the RBS-EC is sensitive to individual differences in repetitive behaviors among typically developing toddlers, providing highly dimensional behavioral data. This is in contrast to the floor effects seen in the application of more clinically-oriented and less developmentally appropriate measures of repetitive behavior to typically developing young children. Next steps include examining the RBS-EC factor structure and psychometric properties in our full sample of 800+ children as well as expanding our sample to include children of different ages and children who are developing atypically.