Just another site

Does having thin friends give you anorexia? …and should there be government intervention about that?

This is a long post about something people might have heard me going on about before, but I think there are some useful points near the end – and a personal confession. If you don’t want to go through all my nit-picking about the research behind these headlines, just scroll down to where it says RANT STARTS HERE.

A story cropping up all over this week (but please read on, past these headlines – because I don’t think the headlines are at all justified):

The Guardian: Anorexia research finds government intervention justified: Economic analysis finds that banning very skinny models from catwalk and pictures from magazines may prevent ‘epidemic’

Vox: Research-based policy analysis and commentary from leading economists: When distorted self-image takes its toll: The effects on the health of European females Joan Costa-i-Font &  Mireia Jofre-Bonet (authors of the original article)
Striving for the perfect body can take its toll, both physically and mentally. This column shows how excessive preoccupation with self-appearance can give rise to preventable eating disorders, such as anorexia and bulimia, among European females. It is time for policy action to shift people’s perceptions of their ideal body closer to what is healthiest.

The Age (an Australian newspaper): Skinny model ban ‘could curb anorexia’
Governments are justified in using the law to stop modelling agencies using very skinny women on catwalks and prevent magazines from printing photographs that suggest extreme thinness is attractive, according to research from the London School of Economics.

These are based on an upcoming paper in Economica. It’s not available online yet, though it looks as though Economica makes the current issue available free online, which is good, so you might be able to get to it in a few months. The authors link to CEP Discussion Paper No 1098 November 2011: Anorexia, Body Image and Peer Effects: Evidence from a Sample of European Women Joan Costa-Font and Mireia Jofre-Bonet in their Vox piece above, which looks as though it’s likely to be very similar to the forthcoming article.

The article is long and complicated and based on economic modelling, with lots of equations. I don’t understand the modelling process, and even if I did, I couldn’t follow the maths. So perhaps I shouldn’t comment, but I think I get the drift of the argument and the evidence – and how that relates, or doesn’t, to the headlines. I don’t think the article justifies the conclusions above, and I think it misses an important psychological point about eating disorders.

The research is a piece of economic modelling about the relative utility of health and body image, and how that might be influenced by various social and demographic factors, which comes to the conclusion:

Our results were consistent with the assumption that individuals trade off health against self-image.

Also, there’s a demonstration that ‘severe anorexia’ rates, as defined by the number of women who had a very low BMI (body mass index & so were extremely thin), who saw themselves as being ‘fine’ or ‘too fat’, and who also thought they were eating adequately, are higher in those European countries where women have lower BMIs generally. I don’t think that’s a great definition of extreme anorexia. But OK, then – and then what are their conclusions?

Also, in agreement with the epidemiological literature, we found that weight-related food disorders happen mostly at younger ages and require attention before they extend to older age groups. Note that the findings showed that anorexia primarily affected women aged between 15 and 34, and that it was primarily socially induced. These results have serious policy implications. They call for urgent action on individual identity, probably while it is still being formed, so as to prevent severe damage to women’s health and in order to improve their well-being and that of their families and friends.

Well, we sort-of knew about the younger-ages bit, and could have guessed that there are social influences (in accordance with a number of ‘you catch being fat from your friends & family’ findings),but does that really provide solid backing for the conclusions in the last two sentences?

Both the newspaper headline stories above talk about how this research supports a government ban on thin models. All I could find out about that in the article was the final paragraph:

In the light of this study, government intervention to adjust individual biases in self-image would be justified to curb or at least prevent the spread of a potential epidemic of food disorders. The distorted self-perception of women with food disorders and the importance or the peer effects may prompt governments to take action to influence role models and compensate for social pressure on women driving the trade-off between ideal weight and health. However, given the nature of the data and the absence of natural experiments we can’t prove our results as being causal and should be taken with caution.

Nothing about banning thin catwalk models (actually nothing about models at all) in the paper. The authors did try out a measure of exposure to inappropriate images by using subscription rates to ‘women’s magazines’ – and found it unrelated to anorexia rates. They comment:

The result of non-significance for the women’s magazine circulation per capita was quite puzzling as it was not consistent with some specific studies on the subject (Turner et al., 1997). This may be due to the crudeness of the country measure and the possibility that the categories are not comparable across countries; perhaps better quality data was required to measure the effect of environmental or media-related variables.

In other words, as good scientists, if we don’t find the results we wanted we presume it must be a problem with our measurements (this isn’t meant to be a snide criticism of the authors: that’s the way most people react to disconfirmation, really – and their measure was pretty crude). More importantly, it was NOTHING to do with skinny catwalk models.

Actually I think other measures in the paper seem pretty crude and/or inappropriate: for instance, their measure of health-consciousness was “the declared number of gynaecological screenings taken in the last 6 months.” The study uses a big general-purpose European dataset, so they have to use whatever measures were taken in compiling the dataset, rather than choosing appropriate measures – but it might be better not to force too much meaning into those measures.

The index of ‘severe anorexia’, as defined above, for women 15-34 varies a lot across European countries, from 4+% in Austria to 0.0% in Northern Ireland, what used to be West Germany, Greece, France and the Netherlands. What used to be East Germany (right next to West Germany) has a rate of 1.45%, and Ireland (right next to Northern Ireland) has 2.66%, so those are medium and high rates compared with other European countries. It’s a bit surprising that what you might think are closely related countries have such different rates of severe anorexia – though the mean BMIs of the population of young women in the those countries (the peer comparison measure) do go in the appropriate direction: higher in WGermany and NIreland than EGermany and Ireland.

So, I’m not convinced by the evidence, and it looks as though the ‘government should ban skinny models’ stuff just comes out of reporters’ fevered imaginations (or, more likely, the headline they’ve used several times before without thinking about it properly then, either). But on top of that….


Two things to rant about.

Yes, anorexia can be dreadful, both for those individuals who want to starve themselves, and for those around them, but it’s not the important weight epidemic. Overeating and obesity are what kills many more people, and looks to be getting to be a bigger and bigger problem. So if underweight models really do encourage young women (and men) to eat less – bring them on. Starve them more: their sacrifice will be worth it for the good of the nation. When I walk down the street, I don’t see much evidence of the malign influence of skinny models; more the effect of cheap calories and low-effort transportation, and I’m sure that’s the case in the diabetes clinics, too. [Disclaimer: my BMI is around 30, so I could definitely do with some of that influence, if it worked].


Whenever I read first-hand accounts of anorexia, the thing that strikes me most are issues of control, not body dysmorphia or inappropriate models. A couple of recent, anecdotal, examples: Gok Wan talking about his anorexia on TV last week: ‘I felt I couldn’t control anything in my life except what I put in my mouth, so I started to control that’.

Laurie Penny (identifying herself as a recovered anorexic) in the New Statesman, 5 March, 2012 (this article may appear on her blog:, which has other interesting stuff on it, though it’s not there as I write):

The most important thing to recognise about eating disorders is that starving, bingeing, purging and puking are not causes of distress, they are symptoms of it. The diseases are replete with contradictions, at once about denying hunger for food, for rest, for fun, for sex, for freedom while the sufferer – a curious combination of aggression and compliance. Eating disorders are what happens when youthful rebellion cannibalises itself.

She compares anorexia with work-to-rule strikes:

Women, precarious workers, young people, and others for whom the stakes of social non-conformity are high, lash out by doing only what is required of them, to the point of extremity. Work hard; eat less; consume frantically; push yourself to the point of collapse.
We followed all the rules, sufferers seem to be saying – now look what you made us do.

Seems a more psychologically (and socially) sophisticated account to me – and suggests that even if we locked up all those skinny models, the problem won’t go away.

Final admission That’s my position, intellectually, but actually, deep down, I’m influenced by the skinny models, too, and they’ve led me into dysmorphia. I would love to be able to put on a light-coloured linen suit and look like Bill Nighy

 or Dan Cruickshank,

 but when I look in the mirror, all I see is Sydney Greenstreet (Casablanca, The Maltese Falcon, and below):


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: