Eating disturbances in adults with autism spectrum disorder without intellectual disabilities.
Verbally fluent autistic adults carry more eating-disorder and autism-specific food quirks than peers—so screen both domains.
01Research in Context
What this study did
Demartini et al. (2021) compared eating habits in verbally fluent autistic adults and neurotypical peers. They used two checklists: one for eating-disorder symptoms and one for autism-specific quirks like food touching or same-brand rules. The team kept age, sex, and body weight the same across groups so autism was the only difference.
What they found
Autistic adults scored higher on both scales. They showed more dieting, shape concerns, and bulimic episodes. They also reported more autism-only habits such as only eating foods of one color or from one bowl. The gap stayed even after the researchers controlled for weight and mood.
How this fits with other research
The adult spike in eating problems extends what Hamama et al. (2021) saw in kids: autistic children, especially girls, show more emotional over-eating than typical peers. Benedetta moves the lens up two decades and swaps "emotional eating" for classic eating-disorder signs.
Islamoğlu et al. (2025) also found high ARFID risk in autistic children. Together the three papers draw a straight line: picky, rigid eating starts young and can bloom into full eating-disorder patterns by adulthood.
Martins et al. (2008) once reported only mild picky eating in autistic kids. That softer signal may reflect the broad survey they used; Benedetta’s newer, finer scales now capture the sharper adult picture, so the field has moved on.
Why it matters
If you serve autistic teens or adults, add two quick questions to your intake: "Any worries about weight or shape?" and "Any food rules you can’t break?" A positive reply earns a referral to an eating-disorder-informed dietitian. Catching the problem early can spare years of medical and mental-health fallout.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Add two eating questions to your intake form and flag any yes answers for a dietitian consult.
02At a glance
03Original abstract
There is a growing interest in the relationship between autism spectrum disorders (ASDs) and eating disorders (EDs), two relatively common conditions lying on a spectrum from mild to severe clinical features. However, only limited data are available about pathological eating behaviors throughout adults on the autistic spectrum. The aim of the present study is to assess dysfunctional eating behaviors, including EDs manifestations and ASDs-related eating disturbances, in a population of adults with ASDs without intellectual disabilities. We recruited 106 adults on the autistic spectrum, without intellectual disability and 103 neurotypical adults (NAs). Participants completed the "Eating Attitude Test" (EAT-26), to measure symptoms and concerns characteristic of EDs, and the "Swedish Eating Assessment for Autism Spectrum Disorders" (SWEAA), to assess eating behaviors frequently observed within the autistic spectrum. Participants with ASDs scored significantly higher than NA at the EAT-26 and at the SWEAA. Moreover, participants with ASDs scored higher than NA at the EAT-26 subscales Dieting and Bulimia. The difference between groups remained significant after controlling for the effect of age, biological sex, and BMI. These results suggest that adults with ASDs without intellectual disability presented not only a higher prevalence of eating disturbances typical of the autistic spectrum, but also other symptoms of EDs in comparison to NA. LAY SUMMARY: For both scales assessing eating disturbances (EAT-26 and SWEAA), participants with ASDs scored higher than NA, presenting a higher prevalence both of eating disturbances typical of ASDs and of ED symptoms (distorted body image, tendency toward bulimic behaviors, and self-control of eating).
Autism research : official journal of the International Society for Autism Research, 2021 · doi:10.1002/aur.2500