Overweight and Obesity in a Sample of Children With Autism Spectrum Disorder

OBJECTIVE: To determine the prevalence of overweight/obese status in children with autism spectrum disorders (ASD), identify associated characteristics, and develop a model to predict weight status. METHODS: The prevalence of overweight and obesity was determined in 2769 children with ASD enrolled in the Autism Speaks Autism Treatment Network, a collaboration of 17 academic centers, and compared with a national sample matched for age, sex, and race. Associations in the ASD sample between weight status and demographic and clinical variables, such as age, race, head circumference, and adaptive functioning, were tested using ordinal logistic regression. The accuracy of a final model that predicted weight status based on early life variables was evaluated in a validation sample. RESULTS: The prevalence of overweight and obesity were 33.9% and 18.2%, respectively; ASD was associated with a higher risk of obesity (but not overweight) relative to the national sample (odds ratio [OR], 1.16; 95% confidence interval [CI], 1.05-1.28; P = .003). In the adjusted analysis, overweight/obese status was significantly associated with Hispanic ethnicity (OR, 1.99; 95% CI, 1.37-2.89), parental high school education (OR, 1.56; 95% CI, 1.09-2.21), high birth weight (OR, 1.56; 95% CI, 1.11-2.18), macrocephaly (OR, 4.01; 95% CI, 2.96-5.43), and increased somatic symptoms (OR, 1.41; 95% CI, 1.01-1.95). A prediction model designed to have high sensitivity predicted low risk of overweight/obesity accurately, but had low positive predictive value. CONCLUSIONS: The prevalence of obesity in children with ASD was greater than a national sample. Independent associations with increased weight status included known risk factors and macrocephaly and increased level of somatic symptoms. A model based on early life variables accurately predicted low risk of overweight/obesity.