Your browser is no longer supported

For the best possible experience using our website we recommend you upgrade to a newer version or another browser.

Your browser appears to have cookies disabled. For the best experience of this website, please enable cookies in your browser

We'll assume we have your consent to use cookies, for example so you won't need to log in each time you visit our site.
Learn more

Autistic traits may be more common in anorexic girls

  • Comment

“Anorexic obsession with food and calorie-counting could be linked to autism,” the Daily Mail reports.

The headline is based on a study involving a small group of adolescent girls diagnosed with anorexia nervosa. The researchers found that the teenagers scored higher for traits of autistic spectrum disorder (ASD) compared with a large group of similarly aged adolescent girls without anorexia.

They also tended to have reduced levels of empathy while having higher levels of what is known as ‘systemising’. This is an impulse to follow a strict system of behaviour in day-to-day living and is similar to the symptoms of obsessive compulsive disorder.

The researchers suggest that clinicians should consider a potential crossover between the obsessive, systematising and self-focused traits of ASD and those of anorexia. They also suggest that this crossover could be important when thinking about treatment and causes of anorexia.

Although the exact relationship between the two conditions is unclear it is possible that certain behavioural characteristics of both ASD and anorexia, such as social isolation, overlap.

It should not be assumed from this research that people with ASD or related developmental disorders are at risk of anorexia, or vice versa.

Importantly, the results were based on just 66 adolescent women diagnosed with anorexia, so may not be representative of the wider majority of women with anorexia.

The suggestion of a link between ASD and anorexia is certainly plausible, but it remains unproven.

Where did the story come from?

The study was carried out by researchers from Cambridge University and Cambridgeshire and Peterborough NHS Foundation Trust and was funded by the UK Medical Research Council.

The study was published in the peer-reviewed science journal Molecular Autism on an open access basis, meaning it is free for everyone to read and download.

The media reporting was generally balanced. The headlines suggest that anorexia “could be linked” to autism, rather than making more concrete assertions. This is refreshing as, often, a potential link found in preliminary research gets written up as if it were an established fact.

Even so, it is important to remember that this is not necessarily a causative link. It is possible that the findings just reflect a similarity of certain behavioural characteristics common to both conditions.

What kind of research was this?

This was a cross-sectional study examining the prevalence of autistic traits among female adolescents with anorexia nervosa (anorexia), compared with that among adolescent females without the eating disorder.

The researchers define anorexia as a refusal to maintain a minimum body weight (>15% below expected body weight), and an unhealthy obsessive preoccupation with food and weight. They say that the eating disorder may be caused by a combination of environmental (social) pressures, familial dynamics, genetic predisposition and having a distorted perception of the body.

The researchers wanted to investigate whether there is a link between anorexia and traits of autism.
The term autism is sometimes used interchangeably with “autistic spectrum disorder”, which covers both autism and Asperger syndrome. These developmental conditions are characterised by problems with:

  • social interaction with others (for example, not being able to respond to others’ emotions)
  • communication (for example, difficulties in having a conversation)
  • having a restricted, repetitive collection of interests and activities, and rigid routines and rituals

The main distinction between autism and Asperger is that children with autism tend to have some degree of intellectual impairment. While those with Asperger syndrome tend to have normal intelligence and sometimes have specific skills in certain academic or creative areas.

What did the research involve?

The study recruited 66 female adolescents (aged 12 to 18 years) with anorexia and 1,609 without. Both groups were mailed questionnaires that gathered information on autistic traits. The questionnaires gave results in three main areas:

  • Autism Spectrum Quotient (AQ) – this has five elements (social, communication, attention to detail, attention-switching, imagination) and scores range from 0 to 50. Eighty per cent of people with autism spectrum disorders score above 32.
  • Empathy Quotient (EQ)
  • Systemising quotient (SQ). Systemising refers to interest in activities with rules and patterns. People with autism spectrum disorders are expected to score lower on EQ and higher on SQ than people without.

The researchers compared the autistic trait scores for the women with anorexia and those without to look for differences. They split the women into younger (under 16) and older groups (over 16) because some of the surveys were designed to be filled in by the parent for under 16s and by the women themselves if they were over 16.

What were the basic results?

Average AQ score in women with anorexia was higher than in those without. In the younger group (under 16s) average AQ scores were 21.8 in those with anorexia compared with 11.8 in those without. In the older group (over 16s) there was an average AQ score of 21.0 in those with anorexia and 15.5 in those without.

Both differences were statistically significant, which means they are unlikely to be the result of chance.

Similar statistically significant differences were seen for EQ and SQ measures. The only exception was that there was no significant difference between the EQ scores among older women with and without anorexia.

Average EQ scores were:

  • younger groups – 44.7 in those with anorexia compared with 51.2 in those without (significant difference)
  • older groups – 49.6 in those with anorexia and 48.0 in those without (not statistically significant)

Average SQ scores were:

  • younger groups – 39.7 in those with anorexia compared with 33.4 in those without (significant difference)
  • older groups – 64.9 in those with anorexia compared with 51.7 in those without (significant difference)

The EQ and SQ measures use different scoring systems in over 16s and under 16s. This explains the large difference in the scores between the two groups above.

How did the researchers interpret the results?

The researchers conclude that “females with anorexia have elevated autistic traits” and that “clinicians should consider if a focus on autistic traits might be helpful in the assessment and treatment of anorexia”.

They highlighted how “future research needs to establish if these results reflect traits or states associated with anorexia”.


This cross-sectional study found that adolescent women with anorexia had higher levels of certain traits of autism (as measured on an Autism Spectrum Quotient score) compared with a large group of similarly aged adolescent women without anorexia.

Specifically, those with anorexia showed reduced levels of empathy and increased levels of systematising behaviour.

The researchers suggest that clinicians should consider a potential crossover between the obsessive, systemising and self-focused traits of autism and those of anorexia, when thinking about treatment and causes of anorexia.

This research was well designed and recruited a large number of women. However, it has a number of limitations which must be considered, most of which were recognised and reported by the study authors themselves.

Lack of detail

No information was collected about the diagnosis of anorexia in the women (for example, how long they had been diagnosed) or on the height and weight of the women (to indicate the severity of their anorexia). There was also no information available on any wider mental health issues that they might have, either currently or in earlier years (including history of developmental problems). Neither was there exploration of possible predisposing factors (for example, disruptive life events or family history). These factors would be useful for determining whether the findings in these women may be generalised to other women with anorexia.

Potential bias

The participants volunteered to participate in the study. This means there is a risk the results may have been different in the group who chose not to participate in the study. This source of bias is known as selection bias.

Survey scores

The analysis of autistic traits was based on scores from a questionnaire. Observed behaviour may be an alternative, more direct way to measure autistic traits. The researchers used different questionnaires for under 16s and over 16s in some cases, meaning it wasn’t possible to combine the results of the different age groups. This means the results are quite specific to the age groups studied, which may prevent them being generalised to other age groups.

Reverse causation

It should not be assumed from this research that people with autism or related developmental disorders may be at risk of anorexia or that individuals with anorexia have these developmental conditions. As the researchers suggest, it is possible that the consequences of anorexia, such as starvation, are causing some of the autistic traits, causing the women to become more self-focused and socially withdrawn, for example.

This research highlights a potential link between autistic traits and anorexia, but it does not prove there is a link. The reasons behind any relationship are unclear, but it is possible that certain behavioural characteristics of both autistic spectrum disorders and anorexia overlap.

Anorexia can often be a challenging condition to treat as it requires both physical intervention (to combat the effects of starvation) and psychological intervention (to combat the mental health symptoms that encourage starvation).

Hopefully this study will lead to new insights into effective methods of treatment.


  • Comment

Have your say

You must sign in to make a comment

Please remember that the submission of any material is governed by our Terms and Conditions and by submitting material you confirm your agreement to these Terms and Conditions. Links may be included in your comments but HTML is not permitted.