Heightened Emotional Eating Behaviors in Children with Autism Are Modulated By Gender.

Background: There is substantial risk for obesity among children and adults with autism spectrum disorder (ASD). There is emerging evidence that overeating behaviors are more common in ASD and might contribute to this obesity risk in ASD. To date, study of food approach behaviors in ASD has not included emotionally mediated eating. Furthermore, although phenotypic sex differences in ASD are gaining recognition, they have yet to focus on eating behaviors.

Objectives: Examine the roles of emotional valence and gender on eating in a sample of children with ASD and typically-developing (TD) children.

Methods: Parents of 231 (145 male) 4- to 17-year-old (mean=8.46, SD=3.31) children (153 with ASD and 78 TD) completed questionnaires about their children’s eating habits. Parent ratings of emotional eating behaviors were obtained using the Eating in the Absence of Hunger questionnaire (EAH) and Child Eating Behavior Questionnaire (CEBQ). Body mass index (BMI) was calculated using parent report of age, height, and weight using CDC norms. Specific subscales of the EAH included Negative Affect, and Fatigue/Boredom. Subscales of interest in the CEBQ were Emotional Over-Eating and Emotional Under-Eating.

Results: The ASD group had a significantly greater male:female ratio than the TD group (chi square=13.91, p<.001). Children with ASD were much more likely than TD children to exhibit over-eating behaviors linked to Negative Affect (t=5.33, p<.001) and Fatigue/Boredom (t=2.42, p=.01), and more likely to exhibit both emotional over-eating (t=7.19, p<.001) and emotional under-eating (t=5.65, p<.001). There were significant interactions between diagnosis and sex (Fs>4.23, ps<.05) for the over-eating related variables. While the neurotypical sample presented no gender differences on any of these measures (ns), females with ASD were significantly more likely than males with ASD to exhibit Negative Affect related over-eating (t=2.76, p=.006), Fatigue/Boredom related over-eating (t=2.86, p=.006), and general emotional over-eating (t=3.03, p=.003). Interestingly, there were no significant correlations (ns) between any of these eating-related ratings and BMI in the ASD group, but in the TD group BMI was significantly correlated with emotional over-eating (r=.28, p=.02).

Conclusions:

To our knowledge, this is the first study to examine emotional eating behaviors (overall and their potential gender differences) in children with ASD. Parents rated emotional eating behaviors as much more common in children with ASD than in TD children. Furthermore, emotionally mediated food approach behaviors were more common in girls than boys with ASD, while no gender differences in these behaviors emerged in the TD children. None of these eating-related metrics was significantly correlated with BMI in the ASD group. This suggests that the reason for eating, but not necessarily the quantity of eating, is more strongly linked to emotions in children with ASD than in TD children, and most strongly in girls with ASD. One possible interpretation is that the act of eating serves as a socially-acceptable “stimming” (self-comfort through self-produced sensory input) behavior. Because females with ASD are more likely to “camouflage,” or to mimic neurotypical behavior (Rynkiewicz et al., 2016), albeit at considerable mental “cost,” eating may be an especially desirable stimming outlet for females.