How did I ever find things to write about before social media? Recently, someone on Tumblr asked whether eating disorder prevention and awareness efforts do more harm than good. In other words, can attempting to prevent eating disorders actually contribute to their development? Good question, I thought. I’ve often wondered about this myself, especially in light of some emerging studies suggesting that “healthy eating” campaigns may actually contribute to unhealthy (restrictive) behaviours around food in school children. So off I went to scour the literature. I came up with a number of hits, but surprisingly few from the past few years. With this recent silence in mind, I will look specifically at a meta-analysis by Stice, Shaw & Marti published in 2007 that highlights some of the characteristics of effective prevention programs and comment on some of the potential pitfalls of prevention.
As a bit of a primer prevention efforts are generally divided into 3 types: primary, secondary and tertiary.
- Primary prevention includes efforts to prevent the disorder/disease/condition from occurring in the first place; essentially, it looks to intervene by removing risk factors and/or promoting protective factors
- Secondary prevention occurs when there has already been exposure to risk factors or in early stages, often before individuals realize there is an issue; efforts might include screening or early intervention
- Tertiary prevention looks a lot like treatment, in some cases, and can be more of a management technique
A handy little primer on prevention is available here if you’re interested in learning more.
When people talk about prevention, they might also refer to universal, selective and indicated prevention. These divisions are pretty straightforward: universal prevention targets everyone, selective targets groups more at risk, and indicated targets individuals specifically identified as at risk. School-based mental health has a document with more prevention definitions that might be of interest here.
The point here is that not all prevention efforts look the same. Eating disorder prevention programs of different types have been tried to varied effect. Universal prevention might take the shape of a national campaign, whereas selective prevention measures might target, for example, college-aged young women or adolescent girls.
Primary prevention for eating disorders is hotly debated, largely due to lacking consensus on the real “cause” of eating disorders. Of course, how you go about “preventing” eating disorders would depend on how you think they come to pass. I wrote about this in a post on girls’ perspectives following an eating disorder prevention program here, so I won’t go on another rant here; suffice to say that how prevention program designers envision the predisposing and perpetuating factors for eating disorders has a strong impact on the type of program they design.
WHAT HAS BEEN TRIED?
There is a fair bit of disagreement in the research and practice community about whether prevention programs are worth the time, effort, and financial input they require. Looking back to 1997, a study by Mann et al. raised the point that trying to prevent eating disorders might have unintended negative outcomes. In their study, a program trying to use both primary and secondary prevention efforts in college females resulted in intervention participants demonstrating higher levels of eating disorder symptoms after the intervention than participants in the control condition.
Likewise, Carter, Stewart, Dunn & Fairburn (1997) conducted a study of primary prevention efforts with adolescent girls; though they did find that target behaviours decreased in the short term, their results indicated an increase in dietary restraint following the intervention. The authors observed that gains in knowledge about eating disorders were retained at follow up, but behaviours returned to their original levels, with dietary restraint increasing. Obviously, this would have been quite concerning for practitioners: does giving individuals at risk more information normalize or instruct on eating disorders, rather than preventing them?
The call to investigate prevention effort efficacy was taken up with gusto in the early 2000s; notably, Stice & Shaw published several meta analyses looking at various prevention efforts to try to find the most effective strategies.
In their 2007 meta-analysis, Stice, Shaw & Marti found that of existing and evaluated programs aimed at preventing eating disorders:
- 51% reduced empirically-supported risk factors including perceived pressure to be thin, thin-ideal internalization, body mass, body dissatisfaction and negative affect (i.e. effective primary prevention)
- 29% were effective in reducing current/future eating disorder symptoms (i.e. effective secondary or tertiary prevention)
These numbers don’t seem extraordinary; 51% is only slightly over half of the 51 eating disorder prevention programs they examined. Notably, however, this seemingly-low percentage of effective programs is not uncommon in the prevention literature; the authors cite a number of studies suggesting that effectiveness is similar in obesity and HIV prevention programs.
WHAT MAKES FOR A GOOD PROGRAM?
As Stice, Shaw & Marti hypothesized, prevention programs with a selected target, rather than a universal approach, were more successful overall. Even within universal programs, participants noted to be “high risk” within the general population targeted were most impacted. This makes intuitive sense, and is an encouraging finding. The authors suggest that selected interventions might be more effective because individuals who are at risk (and know that they are at risk) might be more motivated to seek out prevention programs.
The authors also note that interventions targeted specifically at females were more effective. However, they link this to “the elevated body image and eating disturbances that occur for this sex may motivate them to engage more effectively in the intervention and because there may be floor effects for samples containing males” (p. 221).
I would argue that the main reason that women might respond to prevention programs might be that our societal discourse around eating disorders is still very feminized; men might not self-select into an intervention, particularly one that centers around body image and thin-ideal internalization, because despite increased recognition of eating disorders among men, the stereotype around eating disorders still follows the “skinny white girl.”
Stronger intervention effects were noted among interactive program designs, rather than programs delivering psychoeducation in a teacher-to-student model. Again, par for the course in prevention; when participants are actively engaged in the program they may be more likely to internalize its messages.
Interventions that were more impactful:
- Had participants older than 15
- Were delivered by “trained interventionists” rather than providers with other roles, such as teachers
- Focused on body acceptance
- Induced dissonance (a disconnect between attitudes and behaviours) in participants around thin-ideal internalization
Of note is that many of the impacts of interventions faded over the course of time; the authors attribute this to “the ubiquitous sociocultural pressures for thinness in our culture” (p. 222).
Just like the other prevention study I reviewed for this blog, I noticed a strong focus on thin-ideal internalization in the studies Stice, Shaw & Marti examined, and in their analysis of why some effect sizes were larger than others. While body image and thin-ideal internalization might be the most “socially modifiable” aspects of eating disorders, I wonder if the focus on this (and only this) in many prevention programs might have something to do with the limited effectiveness of these programs.
Though the authors declare a number of programs empirically supported as they have replicated positive effects in a number of iterations of running their programs, I remain somewhat skeptical about the long-term effectiveness in truly preventing eating disorders.
WHAT ABOUT UNINTENDED EFFECTS?
A brief glance at the literature around whether prevention programs might do more harm than good reveals that few studies seem to look specifically at whether giving information about eating disorders to those at risk might actually prove instructional. However, as Stice, Shaw & Marti note, purely psychoeducational programs do not seem to be as effective.
Obviously, I would be remiss if I were to say that “not as effective” means that these programs actually perpetuate the disorders they are trying to prevent, but it is telling to me that just giving people information is unlikely to produce any real protective effects. The skeptic in me would like to see more analysis of potential unintended or iatrogenic (negative and caused by the program itself) impacts of these programs before we keep rolling them out.
Some authors have commented on the potentially harmful effects of those interventions specifically aimed at improving body image. Notably, O’Dea (2002) discusses the importance of designing “positive body image” messages that do not inadvertently reproduce “body beautiful” ideals.
In a study examining both intended and unintended effects of a program designed to provide information about eating disorders, Schwartz, Thomas, Bohan & Vartanian (2007) report that such programs can reinforce stereotypes about eating disorders. The authors’ conclusions are somewhat mixed and could be seen in either positive or negative lights: for one, after seeing a video designed to raise awareness about eating disorders, participants noted that individuals with eating disorders were “pretty, were in control of their lives, and could go on to lead normal lives when they recover” (p. 191).
This could be a good thing, because obviously, individuals with eating disorders are just “normal people”; these responses could demonstrate stigma reduction. However, as the authors noted, there is a potential that this might reflect “glamourizing” eating disorders or portraying them as passing fads, or something it is easy to get over.
This brings me to a question that keeps coming up for me as I survey this literature: is normalizing eating disorders a good thing or a bad thing? Does it help people to see that anyone could develop an eating disorder, or does it make behaviours associated with eating disorders seem “not that bad”?
So, essentially, this is a long-winded way of saying that I can’t really answer the question that Anon posed on Tumblr. There (seems to be) a lack of literature looking at whether prevention programs might serve as “instruction manuals” as it were to eating disorders. The literature points to relatively weak effectiveness for many prevention programs, and some potentially negative “side effects” to prevention programs.
But researchers and practitioners still seem optimistic about potentially finding an effective way to prevent eating disorders. Whether this is possible… I’m on the fence. On the one hand, why not try things and see if it helps? On the other, I share the reticence about the potential for unintended negative and even instructive effects. I certainly wouldn’t want to discount the strong efforts of prevention-oriented practitioners, who have certainly made strides toward improving prevention efforts.
So what do you think, readers? Is prevention worth the effort, or should we focus our efforts elsewhere? Also, has anyone read an article about whether prevention and awareness campaigns might be instructive rather than preventative? I’d love to read such a thing, if it exists.
Stice, E., Shaw, H., & Marti, C.N. (2007). A meta-analytic review of eating disorder prevention programs: encouraging findings. Annual Review of Clinical Psychology, 3, 207-231 PMID: 17716054