Getting off the scale: Non-weight severity markers in anorexia

One of the most common catchphrases in the eating disorder community is, “It’s Not About the Weight!” Although I’d love nothing more than to change this to “It’s About More Than Just Weight!,” the fact is that, in most cases, treatment for an eating disorder depends mostly on weight. If your BMI doesn’t fall below some sort of arbitrary cut-off, no matter how sick, demented, and symptomatic you are, it’s (to borrow a phrase from Seinfeld*) NO TREATMENT FOR YOU!!

Basically all of the treatment providers, advocates, sufferers, and family members say that this is a really, really dumb system. Not that weight is irrelevant to eating disorder risks (low BMI does increase your risk of dying, but that doesn’t mean the risks at higher BMIs are insubstantial, either), but there are plenty of other psychological and physical dangers that can and do kill and seriously harm people at higher BMIs. Insurance companies and many national health systems, however, typically won’t foot the bill (an exception is labwork, but generally that involves a few days in the hospital, followed by discharge. Lifesaving, certainly, but it’s hardly ED treatment. I’ve also heard insurance that won’t pay unless your labwork is seriously messed up, which is problematic because many times in AN, your bloods remain normal even at very low weights and high symptom use, and then crash abruptly). Which is what gave me a WTF moment when I saw that in the new DSM-5, the severity marker for anorexia was…wait for it…BMI.



Bulimia and binge eating disorder have behavioral severity markers: number of binge and/or purge episodes per week. Granted, the diagnostic criteria for these disorders doesn’t involve weight change like AN does, but still. Again, I don’t think weight and weight lost are irrelevant in AN (they’re not), but for them to be the ONLY measure of severity is rather shortsighted.

Thankfully, I’m not the only one who feels this way. A large group of scientists recently published an article in the International Journal of Eating Disorders that showed restrictive eating behaviors were a separate marker of severity in AN (De Young, et al., 2013).

The researchers recruited 115 individuals with full- or sub-threshold AN that were being treated as outpatients in the Midwest. They were divided 60/40 between the restricting and binge/purge subtypes. The researchers defined “restrictive eating behaviors” as “skipping a meal; limiting calories, carbohydrates, or fat grams; and eating as little as possible,” as well as fasting and strictly limiting overall daily intake.

Using ecological momentary assessment (EMA), which using a mobile device to ping a person randomly throughout and day and ask about current or recent use of ED behaviors, including binge eating, purging, and restrictive eating behaviors. On initial surveys, the AN-BP group had higher levels of ED thoughts and behaviors, which is consistent with what researchers found in other studies.

What the researchers found from the EMA data, however, was much more interesting. The AN-BP group, although having the same average BMI as the AN-R group, had significantly different indicators on ED severity. The group had

  • more episodes of binge eating and purging for week (obvious, since this is required for the diagnosis and regular binge/purge behavior is excluded in the diagnosis of AN-R)
  • more fasting
  • more skipped meals
  • more restrictive intake at those meals
  • and more times they reported “eating as little as possible”

What’s more, these behaviors weren’t a compensatory reaction to binge eating (the authors did a statistical analysis that showed when you factored in binge eating episodes, these restrictive behaviors were just as severe). Instead, they were an independent marker of ED severity. In fact, higher BMI was associated with more fasting, though not with other behaviors.

The authors concluded:

For AN, the severity dimension is based on current BMI. Body weight is an indisputably critical marker of medical severity in AN and has been associated with risk of mortality. The results of this study further indicate that the AN subtypes may be a useful proxy for degree of ED psychopathology, providing information about severity as well as the current configuration of behavioral symptoms. Indeed, the frequency of these restrictive eating behaviors (and perhaps binge eating and purging as well) may represent a nonweight-based dimension of severity in AN independent of BMI…professionals working clinically with individuals with AN may wish to assess and monitor the variety of behaviors in which their clients engage to restrict their food intake, noting that the configuration of these behaviors may differ according to subtype and represent a nonweight-based marker of the severity of ED psychopathology.

It’s not that low body weight isn’t a potential marker of severity in AN, just that it shouldn’t be the ONLY marker. We’re judging a mental illness on medical criteria. We don’t do that for depression or anxiety. Behavioral severity is also important in that it has a tremendous effect on people’s lives and overall well-being, independent of how much they weigh. That’s what we need to focus on as well as the physical factors that are associated with EDs.

*True confession: I’ve never seen an episode of Seinfeld. Not ever. Nor most of the TV shows that “normal” people watched when I was younger. Actually, I really didn’t watch TV at all. But my friend was obsessed and shared the Soup Nazi segment with me. I can haz cultural references?


De Young, K. P., Lavender, J. M., Steffen, K., Wonderlich, S. A., Engel, S. G., Mitchell, J. E., Crow, S. J., Peterson, C. B., Le Grange, D., Wonderlich, J. and Crosby, R. D. (2013), Restrictive eating behaviors are a nonweight-based marker of severity in anorexia nervosa. Int. J. Eat. Disord.. doi: 10.1002/eat.22163

Twitter Digg Delicious Stumbleupon Technorati Facebook Email

12 Responses to “Getting off the scale: Non-weight severity markers in anorexia”

  1. Thanks for the post Carrie. Agree. I would argue for an even more comprehensive measure of severity. At our clinic we use The Clinician Administered Staging Instrument for Anorexia Nervosa (CASIAN) developed by Sarah Maguire and Stephen Touyz (amongst others). Severity is staged along BMI (as it is important), weight change over time and acutely, lowest weight, time at lowest weight, desired body weight, length of time engaging in any ED behaviours, any periods of weight normalisation and behavioural abstinence, restrictive eating behaviours, social eating, binge/purge behaviours, mood state, obsessional symptoms, motivation to change, body dissatisfaction, physical complications, interpersonal features and chronicity.
    Long list, but a complicated illness needs a complex assessment of severity.

  2. Hi Carrie
    Could you if possible post what will be the criteria for AN in the DSM-5?
    I have mixed thoughts on this, the people who worked on this are medical doctors. I do wonder though about the whole issue of BMI. For example, a BMI of 13 say on someone who sits normally at a BMI of 18.5, is it really the same on a woman who sits normally at a BMI of 21?

  3. I’m so glad you wrote about this, Carrie. Have you ever seen any positive studies involving BMI?! It’s such an inaccurate measure of everything. I think this study adds fuel to the “I’m not sick enough” fire. It never fails to amaze me that even professionals have such strong misconceptions when it comes to EDs and the recovery process.

  4. I really loved this post, and I thank you sooo much. I identify myself as anorexia, but I would have never diagnosed to be clinically. My BMI didn’t stay low long enough. I pulled it up by myself. I lost weight and became anorexic because I was trying to be right. I researched how much I should eat for my activity level and height, and followed it strictly. I researched for healthy foods to optimize my health, and started to eat them only. Doing right was very important. Having 2 digits for my weight was not right. I pulled it up back to 3 digits, but remained to have a low weight, because I realized that I could not get out from my own rules. That was how I saw myself being anorexia. I can not believe that people think that it is only about food and weight. That’s only a tip of a root problem.

  5. Wait, did I read this correctly? The DSM V is going to use BMI as a criteria for AN, when many physicians and clinicians are questioning the validity of BMI? How many of us have to die before people with AN and other EDs can receive adequate TX that sticks?

    • No, it’s a severity measure which goes alongside the diagnostic criteria. It’s dumb, and it’s not the best measure, outside of the medical complications that more frequently occur with very low BMI.

  6. When it comes to actual mortality in anorexia ( and correct me if I’m wrong) apparently the main cause of death is not starvation, but suicide.

    I suspect most people would agree that suicide might qualify as fair indicator of severity, and yet how exactly it ties in with BMI is rarely addressed.

    So count me with those above who feel that BMI and many of the other quantitative measures used to define anorexia do a poor job of capturing the real essence of the thing.

    While research into EDs is always interesting, as long as psychiatry continues in it’s push to move further away from humanistic measurements, I suspect they will continue to miss the mark when it comes to explaining this disorder in ways that will truly inform treatment.

    • The percentage of deaths by suicide in AN is very high- the stat I remember is 1/5, but it will probably vary somewhat from study to study.

      I do think researchers are getting better at explaining EDs in ways that inform treatment. For one, we’re slowly starting to move away from the idea that bad parenting causes EDs. It would be nice if we moved further from the thin ideal being causative (it contributes, yes, but there would be way more EDs if it were the primary factor), but still. Understanding the need to normalize eating as well as the role the ED plays in emotion regulation will be hugely important in developing new treatments.

  7. Hi Carrie, thanks for this post. I’ve linked to it in a post I just wrote about a recent Guardian article by Emma Woolf on “skinny-shaming” which made me feel a bit uncomfortable. I’d like to know what you think about it too!

    It’s so hard at the moment in my recovery because everyone assumes I’m better, when in actual fact there’s possibly even more anxiety and stress than ever. But I’m pushing forwards because I know in time it will get easier.

  8. Oops sorry I forgot to provide a link for my blogpost…

  9. I agree with this post and feel like it’s a crucial matter. I’ve just been diagnosed with an eating disorder and, having read about the situation in the US, was pleasantly surprised by the evaluation I went through here, in Switzerland. I met up with a doctor and a nurse who were both ED specialists and we talked for an hour. There was no pro forma questionnaire but my situation was thoroughly assessed. I did mention my height and weight in the discussion, as well as how much I’ve lost, but most of the conversation revolved around mental states and disordered thoughts. Even though as a restricter I do not fulfill the criteria for anorexia (yet) I will have immediate access to outpatient treatment. This will be reimbursed by my insurance (which in this country is both compulsory and very expensive I must say, so all isn’t perfect). It is so important to receive treatment before the situation gets completely out of hand! And I’m sure there’s a rationale for it: early treatment will eventually come out cheaper than relapse after relapse.


  1. Signs Your Diet Has Crossed the Line Into Anorexia - February 3, 2014

    […] Yet you will never find any definition of anorexia that is based solely on weight or body fat. This article points out that you could even be overweight and still have anorexia and that anorexia is best defined as a mental illness with a disordered perception of […]