Mental health

Should we rethink the causes of anorexia?

"Anorexia is not about a fear of getting fat, but rather a pleasure at losing weight, experts reveal," says the Daily Mail. The headline oversimplifies the results of a study that looked at women's responses to photos of women of varying weights.

In the study, 71 women with anorexia and 20 without were shown photographs of women who were either a normal weight, underweight or overweight, while monitors recorded sweating caused by emotional excitement. This type of test, known as a skin conductivity test, is reported to be a way of assessing levels of emotional excitement.

Researchers found women with anorexia felt more negatively about images of normal and overweight women, and more positively about images of underweight women, compared with women without anorexia.

This suggests a desire to be thin may be more important than a fear of getting fat, say the researchers. This hypothesis, unproven as it currently is, could explain the continuing popularity of "pro-ana" websites. These sites often use pictures of underweight women to promote the so-called "anorexia lifestyle".

The researchers also tried to see if this "getting thin is pleasurable" attitude was linked to a specific gene type known as Val66Met, but the results were inconclusive.

Anorexia carries the highest risk of death among all mental health conditions. If you are worried you may have anorexia or someone you know has it, it's important to get medical help as soon as possible. You can start with your GP, or by visiting the Beat, a UK charity supporting people with eating disorders.

Where did the story come from?

The study was carried out by researchers from Paris-Descartes University and INSERM UMR in France, and the University of Ulm in Germany.

It was funded by Fonds d'Etudes et de Recherche du Corps Médical.

The study was published in the peer-reviewed journal Translational Psychiatry on an open-access basis, so you can read it for free online.

Despite the Mail's over-simplifying headline, its report covers the study well. Unfortunately, the article is illustrated by a photograph of a very thin young woman in her underwear, exactly the sort of image the study suggests may motivate women with anorexia.

What kind of research was this?

In this case control study, researchers compared the reactions of 20 healthy women with the reactions of 71 women with anorexia when they were shown images of underweight, normal weight and overweight women.

Researchers wanted to know which images provoked the strongest responses in women with anorexia, and whether that was different from healthy women.

Studies like these can help us understand more about an illness, but they can't prove causality – so, in this case, we don't know if the women's reactions are a cause of anorexia, or perhaps a symptom of the disorder, or if they're related in some other way.

What did the research involve?

Researchers recruited 71 women being treated for anorexia at a hospital in Paris, and 20 of their friends or acquaintances of a similar age and education level.

The women were shown 120 images over four sessions, while answering questions about them and having their skin response monitored. Afterwards, researchers compared the results between women with and without anorexia.

The women with anorexia had a mixture of conditions (food restriction or binge/purge) and were at a range of weights. Some were no longer clinically underweight, having put on weight since starting treatment. Half of them were treated as in-patients.

All women were asked to categorise the images they were shown by estimated weight, and to say how they would feel on a scale of one to four if that was their body (one being very unhappy, four being very happy).

They wore devices on their hands that tested skin conductivity to measure the rate of sweating caused by emotional excitement as they watched the images.

This measure was included as a truer measure of reaction, as some women may have felt they were supposed to react in certain ways to images of people who were over- or underweight.

All women also had their saliva analysed for a type of gene (Val66Met) that has been linked to anorexia, although the link is unproven.

Researchers analysed the data for a range of groups and sub-groups. They looked at whether women with anorexia reacted differently from women without, and whether women's weight or length of illness affected the results.

They also looked at how many women carried the gene associated with anorexia, and whether that affected the results.

What were the basic results?

Women with anorexia were more likely to overestimate the weight of people in the images who were underweight and normal weight.

Women with anorexia were likely to be less happy with the thought of having a body like the normal-weight images (average score 1.9, compared with 2.6 for healthy women) and slightly less happy than healthy women at the thought of having a body like the overweight image.

They were happier with the thought of having a body like the underweight images (average score 2.7, compared with 1.9 for healthy women).

Results of skin conductivity tests found women with anorexia were more likely to show a response to underweight images than healthy women were. They were also more likely to show a response to underweight images than overweight or normal weight images.

Carriers of the gene type linked to anorexia were more likely to show a skin conductivity response to images of underweight women, compared with those without this gene type.

How did the researchers interpret the results?

The researchers say their results "suggest an increased attention and motivation toward underweight stimuli [i.e. images of underweight women]" among women with anorexia, which "may promote pathological behaviours, maintaining starvation in patients".

In other words, they believe that viewing images of underweight bodies might encourage people with anorexia to persist in eating too little.

The researchers say their results showed the positive feelings towards images of underweight bodies were stronger than negative feelings towards overweight bodies.

They say having a "positive value of starvation, rather than a more negative value of overweight" might be a more accurate definition of anorexia than the current definition, which emphasises an intense fear of weight gain. They add that the gene type investigated "could partly mediate" this response.


Anorexia is a notoriously difficult illness to treat. While people do recover, many live with this devastating condition for years, and some die from it.

Because anorexia is so difficult to cure, researchers are interested in finding out more about how the condition works. A better understanding of the underlying causes might help find better treatments.

This study is an interesting addition to that understanding. A primary feature of anorexia was always thought to be fear of weight gain, and many people with anorexia say they do fear putting on weight.

But this study found that a desire to be very thin may be as important, or possibly more important, than a fear of gaining weight.

There are some important caveats. Showing an electrical reaction in the skin when people look at images of underweight bodies, and hearing that people with anorexia would be happy to have an underweight body, is not the same as proving that a desire to be thin underlies the illness.

The electrical reaction was assumed to be one of excitement, but it could equally have been one of increased anxiety.

The study involved only 71 people, with different types of anorexia and at different stages in their illness. A bigger study with more specific groups might help us understand more.

For example, we don't know whether a preference for underweight body types triggers anorexia, or whether this preference is learned as anorexia develops.

The control group only included 20 people, meaning their results might not be representative of all healthy women. This would change the effect of comparing responses from women with anorexia to responses from healthy women.

The findings relating to the gene type are quite tenuous. For one thing, the gene type was as common among women without anorexia as with it. More work needs to be done to understand whether genetics does have a role to play in the condition.

NHS Attribution