Growth Trajectories of Problem Behaviors Utilizing the Aberrant Behavior Checklist

Background: Recently there has been an increase in interest in behavioral difficulties that co-occur with Autism Spectrum Disorders (ASD). The majority of longitudinal studies report improvements in behaviors as youths with ASD get older, but there are also repeated references to individuals who experience extreme changes in behavior and sometimes marked deteriorations.

Objectives: To examine predictors of growth trajectories in problem behaviors (lethargy/social isolation, irritability, and hyperactivity) from mid-childhood to late-adolescence.

Methods: The sample (n=120) is comprised of individuals with consecutive referrals for possible autism at age 2 to clinics in North Carolina and Chicago and nonspectrum individuals recruited from North Carolina, Chicago and Michigan with some degree of mental retardation, language delay, or with other disabilities. Growth curve analyses were utilized to determine significant changes in lethargy, irritability and hyperactivity, measured by the Aberrant Behavior Checklist (ABC).

Results: Total scores for the three ABC subscales ranged from 0 to 111 out of a possible 141, with higher scores indicating more problematic behaviors.  At both the first and last measurements, the autism group exhibited significantly more problem behaviors overall compared with children with PDD-NOS and nonspectrum delays.  However, the autism group also improved the most on average.  The overall trend from age 9 to 18 was one of decreases in problem behaviors.  About half of the autism group (49%) showed decreases in problem behaviors of at least one standard deviation (S.D.=12.7), while a smaller proportion of the PDD-NOS and nonspectrum groups underwent similar decreases (24% and 38% respectively).  The hyperactivity subscale accounted for the largest decrease in problem behaviors.  Problem behaviors worsened substantially for a small but significant minority of children with ASD (9-10%). Predictors of outcome and longitudinal “growth” trends were examined separately for the three subscales.  For all three subscales, the pattern of change for the autism group alone was one of an accelerated decrease in problem behaviors between ages 9 and 11 (i.e., a significant positive quadratic slope), after which the decline in problem behaviors continued at a slower rate.  Verbal IQ predicted fewer problem behaviors for all three subscales, but did not affect the rate of change over time.  Greater lethargy/social isolation at age 9 was reported for children who had one or more seizures in their lifetime, however, the rate of decrease in lethargy over time was also greater compared with those who had never seized.   Various other factors were not significant predictors of outcome, including race, gender, caregiver’s education, hours of treatment, and pubertal development.

Conclusions: Findings from this study suggest that problem behaviors such as hyperactivity, lethargy/social isolation, and irritability tend to decrease over time for children with ASD and nonspectrum delays but outcome differs according to diagnosis, IQ, and seizure activity.  In the final analyses, we will examine additional factors likely to influence outcome such as classroom placement and medication usage. Closer examination of those whose problem behaviors increased over time is needed as well.  The clinical implications will be discussed.