Prevalence of Obesity in Autism Spectrum Disorders and Associated Risk Factors

Background: Obesity (OBY) is increasingly prevalent in childhood and adolescence, and is associated with medical problems, psychopathology, and impairments in functioning in typical populations. Children with autism spectrum disorders (ASD) may be at elevated risk for OBY for several reasons, including medication use and decreased physical social activities. However, few studies have examined behavioral and psychiatric risk factors associated with OBY in ASD.

Objectives: To examine predictors of OBY in a large clinical sample of children with ASD.

Methods: Participants included 5082 children with ASD (M= 6.1 years; range 1.6 – 17.4 years; 84.5% male) enrolled in the Autism Treatment Network (ATN). CDC and WHO gender-/age-specific cutoffs were used to classify children as OBY (BMI ≥ 95%) or not obese (NOB, BMI < 95%). Bivariate analyses were used to assess the association of OBY with socio-demographics, ASD severity, child functioning and behavior, medication use, and comorbidities.

Results: OBY prevalence was 17.9% overall, and ranged from 12.9% to 26.7% across ATN sites. In terms of socio-demographics, OBY prevalence was lower for children < 6 years than for those ≥ 6 years (16% vs 20.9%, p < .001), and was unrelated to gender (p = .9). OBY was positively associated with lower parental education (p < .001), non-white race (p = .01), and Hispanic/Latino ethnicity (p < .001). There were no differences between OBY and NOB in terms of ASD symptom severity (ps > .4) as measured by ADOS Calibrated Severity Scores. In terms of cognitive levels, children with OBY were more likely to have IQ scores < 70 (34.9% vs 28%), and less likely to have scores ≥ 85 (45.9% vs 52%), compared to NOB children (p = .01). OBY children also had significantly lower adaptive skills overall (Cohen’s d = .15) and in the communication (d = .13), socialization (d = .10), and daily living skills domains (d = .08). In terms of medication use, OBY children were significantly more likely to take SSRIs (7.4% vs 4.8%, p = .001) and atypical neuroleptics (8.0% vs 5.1%, p < .001), but not amphetamines or melatonin (ps > .2). OBY children had significantly elevated sleep difficulties (d = .14) and co-morbid emotional and behavioral problems as measured by the Child Behavior Checklist (d = .13), including higher levels of affective (d = .17), anxiety (d = .09), and ADHD problems (d = .07), but not oppositional defiant problems (d = .02).

Conclusions: OBY was significantly associated with several socio-demographic variables and with increased use of certain classes of medication. OBY was also characterized by increased difficulties in adaptive skills, communication, and sleep as well as elevations in depressive symptoms. However, ASD symptoms were not associated with OBY, suggesting that the some of the same risk factors apply for OBY occurring in ASD as in typical and other atypical populations. Interventions that take into account both general OBY risk factors as well as those that are ASD-specific, such as adaptive skills and comorbid depression, may hold promise for improved weight status in ASD.