Rethinking ED Prevention
In honor of the last day of eating disorders awareness week, I thought I would do a post on what we know about preventing EDs. A lot of what is discussed during the week (besides the obvious “awareness” bit) is how to prevent EDs.
So for those of you that don’t have much time to read posts about ED prevention, let me summarize what we know about ED prevention right here: not much.
Speaking as an epidemiologist–what I was in my pre-science-writer life–I can say that using the word “prevention” without any qualifications drives me a bit bonkers. There are actually three different types of prevention, as seen below (thanks, Wikipedia!):
- Primary prevention: Methods to avoid occurrence of disease. Most population-based health promotion efforts are of this type.
- The “Love Your Body” campaigns are a good example of this
- Secondary prevention: Methods to diagnose and treat existent disease in early stages before it causes significant morbidity.
- Treating significant disordered eating or subclinical EDs is a good example of secondary prevention
- Tertiary prevention: Methods to reduce negative impact of existent disease by restoring function and reducing disease-related complications
- Efforts to encourage parents and patients to treat EDs early is a good example of tertiary prevention
Most ED prevention is primary prevention. The “Love Your Body” campaigns that admittedly drive me a bit bonkers are a fantastic example of primary prevention. One of my issues with primary prevention is that there is only one remotely evidence-based method of primary prevention (which I will discuss more fully below), and it doesn’t address all of the ED risk factors we know about. The other issue is that we simply don’t know enough about what causes EDs to really target them on a population level.
We do know a lot more about the importance of addressing and treating EDs early, and of certain groups of the population that are at particularly high risk for developing EDs (although our prejudices about EDs generally keeps us from doing a good job of recognizing them). So, as a way to wrap up ED Awareness Week, and to indulge my nerdy epidemiological side, I’m going to cover the subject a bit more in depth here.
When I looked through the ED literature for work on ED prevention, I found that about 95% of it measured outcomes of body image and disordered eating. If you have been reading ED Bites for any length of time, you will know that I repeatedly harp that disordered eating and body image distortion are NOT eating disorders. What’s more 100% of the literature focuses on things like media literacy and loving your body.
Before I get into things further, I want to clarify: I have NO OBJECTION to teaching media literacy. I think we all should be able to deconstruct what multi-trillion dollar corporations are trying to sell us and how we are being manipulated into being “better consumers.” I also think that we should teach kids and even adults that healthy bodies can come in a variety of shapes and sizes, and that you can and should try to take care of your body by feeding it properly and moving it for fun.
I don’t remain convinced that these messages will do much to prevent eating disorders. However, reading the work of Eric Stice, who has done fairly rigorous work in this area, I am willing to concede that it might help more than I might always like to admit, though not as much as a lot of people might hope. About five years ago, Stice and colleagues published the first results of a randomized control trial of an eating disorder prevention program that deconstructs the thin body ideal (Stice et al., 2008). In brief, the trial randomized 481 adolescent girls with “body image concerns” into one of three programs: one that examined the risks, costs, and reality of the thin ideal, one that was a healthy weight promotion group, and an expressive writing group (control). Each of the programs lasted three hours. During a three-year follow-up, the authors of the study identified 3 new cases of subthreshold AN, 1 showed bulimia nervosa onset, 23 showed subthreshold bulimia nervosa onset, 1 showed binge eating disorder onset, and 12 showed subthreshold binge eating disorder onset. (People who met DSM-IV criteria for an ED were excluded from the trial). Compared to the control group, the girls in the thin ideal deconstruction group showed a 60% reduction in the onset of “eating pathology” (6% vs. 15%).
However, when a follow-up study was performed in 2011, the effect sizes were much smaller (Stice et al., 2011). The thin ideal deconstruction group showed lower levels of ED symptoms, but NOT lower levels of new-onset EDs (again, people with current EDs were excluded from the study). The studies themselves, the course content, the control groups, and the facilitators were all different in this later trial, which likely had an effect. However, whether it was these or the actual intervention that was less effective is unclear. The content of the classes seemed interesting and useful, and I think they should be part of learning for all kids (not just adolescent girls).
My problem with these types of intervention is that they essentially assume that EDs are primarily a psychosocial disorder while ignoring any potential biological influences (though you can’t change biology, so I don’t necessarily expect that a prevention program would target biology specifically). An ED is NOT an attempt to match the thin ideal, and to think so hurts the eating disorder field as a whole. I can’t help but think that no wonder people think EDs are vanity issues when we try to prevent them with a “Love Your Body” week.
As well, they don’t address the numerous other routes into an eating disorder, such as athletics, “healthy eating,” illness, anxiety, and depression. These types of interventions could very well help certain subsets of potential ED sufferers, but I think focusing too much on the thin ideal could be very detrimental.
Despite posters from the National Eating Disorders Association about how much 9-11 year-olds are dieting, not all of these kids are at the same risk of developing an ED. In fact, the large majority of them won’t develop an eating disorder. Although identifying everyone at high risk for developing an ED is going to be next to impossible, we do know of some pretty significant risk factors, such as history of depression and/or anxiety, family history of eating disorders, perfectionism, impulsivity and/or compulsivity, etc. These, combined with signs of an emerging ED (extreme body dissatisfaction, weight loss/gain, food rules), puts someone at very high risk of an ED.
A few years ago, researchers at Stanford University randomized 480 college-aged women with “high weight and shape concerns” to a control group or an 8 week, Internet-based CBT course with a monitored discussion group (Taylor et al., 2006). These women were then followed for 2 years, and control women were given the opportunity to participate in the program after the follow-up period. Overall, the researchers didn’t find any significant differences between the control group and the treatment group. When they looked at subgroups of the participants, however, they found that women with a BMI >25 or those who used compensatory behaviors showed significant benefits from the program. Whereas 11.9% of women with BMI>25 in the control group developed a clinical or subclinical ED during the 2 year follow-up, 0% of the women in the CBT program did. For purging behaviors, it was 4% vs. 14.4%.
The numbers were still relatively small, and I’m not sure if the researchers thought of doing the subgroup analysis before or after they tested the groups as a whole. The statistical math is complicated, but doing a subgroup analysis because the data didn’t say what you want is cheap and not a strong, mathematically speaking. The analyses didn’t have “post hoc” in the name, so in theory they decided to do the subgroup analysis at the outset.
As far as I know, the trial hasn’t been repeated, which is a major weakness, as you saw above in the Stice trials. Still, we know much more about what places people in high-risk groups, which makes us much more able to effectively intervene. I would LOVE for researchers (I know you nerdy types read my blog!) to focus on high-risk groups that have nothing to do with body dissatisfaction and instead look at perfectionism or coping skills or whatever.
Tertiary prevention is all about early recognition, early diagnosis, and early treatment. An online program aimed at parents whose teenage children met “risk criteria” for anorexia (Jones et al., 2012). The parents received 6 weeks of information, chat sessions, videos, quizzes, and behavior logs, and were followed for a year. Of the 19 families who participated, 16 showed reduced AN risk status at follow-up. It’s a small study and needs to be replicated, but it seems interesting and promising.
Mostly, the problem with tertiary prevention isn’t that we don’t know what works best, it’s that we don’t do it. Physicians say “you’re not that bad,” “there’s nothing really wrong,” “well at least you’re not [fill in the blank].” Health care systems don’t cover treatment if you’re not at death’s door. Patients are left to linger at sub-optimal health. We know these don’t work, we know they contribute to the low recovery rates and high mortality rates in EDs, and we know they seriously harm patients. Yet we don’t do anything differently.
Ultimately, the importance of early intervention is summarized in a journal article in Pediatrics in 2003:
There is strong evidence that the longer the duration of illness, the harder it is to achieve recovery. Eating disorders need to be diagnosed early in the disease process in order for treatment to be as successful as possible. By the time a formal diagnosis of an eating disorder is made, the patient is already suffering from serious biopsychosocial problems. Intervention should occur at the first signs and symptoms of disordered eating. Awareness needs to increase at many levels. Early recognition of the disease process by parents, friends, educators, and coaches can facilitate evaluation by the health care system. Practitioners need to be sensitized to the possibility of an eating disorder developing even at a young age, and need better training to improve their recognition of the early stages of the disease process. Screening about body image, dietary changes and dieting habits, and assessment of growth patterns should occur yearly.