Overweight and Obesity in Children with Autism Spectrum Disorders

Background: Childhood overweight and obesity are considerable problems both in the United States and worldwide.  Abnormal weight is often accompanied by increased physical and mental health complications including diabetes mellitus, cardiovascular issues, and depression.  Elevated body mass in childhood generally leads to overweight in adulthood and is associated higher rates of morbidity and mortality.  Youth with intellectual or developmental disabilities, such as autism spectrum disorder (ASD), appear to be at heightened risk for overweight and obesity due to high medication use, atypical eating habits, and sedentary behavior.  Previous literature is mixed as to whether these children actually have higher prevalence rates of overweight and obesity, although the methodology of some studies has been subpar due to use of parent-derived height and weight, and small sample sizes. 

Objectives:  The objective of this project was to reevaluate the prevalence of overweight and obesity for children with ASDs and to compare them to those of typically developing youth. To supplement this information, we also investigated whether there were risk factors associated with abnormal weight.

Methods:  Data were acquired through the Autism Treatment Network (ATN) and included children from 17 sites in the United States and Canada. Body Mass Index (BMI) percentiles were calculated according to the Centers for Disease Control and Prevention (CDC) criteria. These were compared to normative samples in other literature. A hierarchical regression was used to determine whether risk factors for abnormal weight could be identified. Variables of interest included demographic information, medications, sleep problems, comorbid psychopathology, cognitive functioning, and ASD subtype.

Results:  For the sample of 2,610 subjects, 118 (4.5%) children were underweight, 859 (32.9%) were overweight (exceeding top 15% relative to norms), and 452 (17.3%) were obese (exceeding 5% of normative sample) based on BMI.  This means that 1,633 (62.6%) children in the sample had normal weight for their heights.  When compared to the existing literature on the typically developing population, neither the rate of overweight nor obesity in children with ASDs was significantly different.  When these figures were contrasted with other children with ASD from a previous study, there also were no differences in prevalence of overweight or obesity. Asian heritage, high levels of paternal education, stimulant medication use, atomoxetine use, high scores of the Anxious/Depressed CBCL subscale, and having DSM-IV pervasive developmental disorder–not otherwise specified were associated with lower BMI percentile.  Hispanic heritage, SSRI use, alpha 2 agonist use, high scores on the Sleep Disordered Breathing subscale of the Children’s Sleep Habits Questionnaire, and elevated scores of the CBCL Aggressive Behavior subscale and Withdrawn/Depressed subscale were associated with higher BMI percentile (greater likelihood of being overweight or obese), but the amounts of variance explained were very small.

Conclusions:  In this sample, children with ASDs were not significantly more likely to be overweight or obese than their typically developing peers, although both groups had nominally higher rates in comparison to the CDC criteria. We also found that the predictors of abnormal weight were similar to typically developing children.