Does This Study Show That Restriction Doesn't Lead to Binging? Part 2 - Claims

This is the Weight and Healthcare newsletter. If you like what you are reading, please consider subscribing and/or sharing!

Content note: this piece will discuss research around binge eating, including behaviors and frequency.

In part 1, we looked at the authors and methodology of a 2024 study called “Evaluating dietary restriction as a maintaining factor in binge-eating disorder” by Bartholomay et al., whose authors claim that the findings “challenge the assumption that dietary restriction maintains BE [binge eating] among all individuals with BED [Binge Eating Disorder]. Rather, results suggest that dietary restriction may be largely unrelated to BE maintenance in this population, and that reducing dietary restriction generally does not have the intended effect on BE frequency.”

Too Long Didn’t Read Version:When analyzing research, it’s important to check to see if the research that the authors cite to support their claims actually does support their claims. In this case, these authors are drawing large, broad-based conclusions that would support a weight-centric/weight-loss paradigm based on small, short-term, and decades old studies that offer only weak conclusions. They are also completely ignoring the success of a weight-neutral intervention that is included in on of the studies they cite.

Let’s dig in!

When I teach research analysis, I recommend checking something I call “The Do-Do’s” which encompasses two questions: Do the citations support the claims? Do the study data support the results?

In this part of our three-part analysis, we’re going to examine the first question.

Regular readers may have noticed me pointing out in my analyses how often the studies that are cited don’t support the claims. In the case of this study, the question “do the citations support the claims” is particularly important since they are using them to bolster a very limited experimental design, so I’ve devoted this section to some examples.

Bartholomay et al. claim:

“randomized controlled trials conducted among individuals with overw*ight and ob*sity demonstrate that assignment to a low-calorie weight loss diet (e.g., 1200 calories per day) produced greater decreases in binge-eating symptoms than assignment to a wait list control group (Goodrick et al., 1998; Klem et al., 1997; Reeves et al., 2001).”

What I want to note here is that the study authors have created a false binary between low calorie weight loss diets and wait list control group which would be bad enough. In this case, they’ve done so by blatantly ignoring the inclusion of a weight-neutral intervention in the very first study they cited.

Let’s look at the three studies they cited to, in theory, support this claim:

The first study they cite is

1998 Nondieting versus dieting treatment for overw*ight binge-eating womenG K Goodrick 1, W S Poston 2nd, K T Kimball, R S Reeves, J P ForeytDOI: 10.1037//0022-006x.66.2.363

This study evaluated non dieting vs dieting treatment for what they described as “overw*ight binge-eating women.” There were 219 subjects and they were assigned randomly to diet treatment, non-diet treatment, or wait-list control (which is a form of control group in which participants are told that they were on a wait list for the intervention, while being offered no intervention.)

The diet intervention was a restricted diet “reinforced with behavioral strategies” and the non-diet treatment was “therapy designed to help participants break out of their dieting cycles.” It involved 18 months of contact - 6 months of treatment followed by 26 bi-weekly maintenance meetings. They checked in on the subjects 6 months later and 18 months later and at both follow-up points the Binge Eating Scale of both the diet and non-diet group was significantly better than the weight list group and at 18 months both experienced weight gain. They concluded that “Results indicate that neither intervention was successful in producing short- or long-term weight loss.”

I would suggest that this means that the weight loss intervention is a failure since its goal was weight loss and binge eating reduction but the non-diet intervention was a success since it’s goal was to reduce binge eating disorder. I would also suggest that it may have been more effective if the therapy included more specific techniques to reduce binge eating.

What it means for sure is that a higher-weight cis woman (the only group included in the sample and thus the only group we can draw conclusions about) who wants to reduce binge eating without risking the experience of another failed diet (and the physical and psychological risks that can come with it) would be best served by a non-diet approach.

Second, they cite:1997 A descriptive study of individuals successful at long-term maintenance of substantial weight lossM L Klem 1, R R Wing, M T McGuire, H M Seagle, J O HillDOI: 10.1093/ajcn/66.2.239

The study discusses the ridiculous National Weight Control Registry (NWCR) which I often use as a peak example of the embarrassingly poor methodology that gets passed off as research in the weight-centric paradigm. I’ve written about it in detail previously but the short story is that they collect commonalities among an incredibly small group of dieters (literally 10,000 out of what is estimated to be over a billion attmpts) and then make unsupportable claims about those commonalities.

I am unclear why Bartholomay et al cited this since the only reference to binge eating disorders is two studies that happened to include those who binge that tracked the mean lifetime weight loss of participants in order to compare it to the mean lifetime loss claimed in the NWCR.

Not only could I find no claim here about whether weight loss attempts impacted binge behaviors, what I did find was that:”20% of the sample indicated a worsening in time spent thinking about weight and 14% reported a worsening in time spent thinking about food.”

I would argue that, being as generous as I possibly can, this study has nothing to do with the authors claims and, being a bit less generous, this could be seen as a bit of a refutation of the claims.

And this is why we check the references.

Finally they cite

2001 Nutrient intake of ob*se female binge eatersR S Reeves 1, R S McPherson, M Z Nichaman, R B Harrist, J P Foreyt, G K GoodrickDOI: 10.1016/S0002-8223(01)00055-4

In this very small, short study 46 cis women were given a behavioral self-management intervention (6 months of weekly, 1-hour classes taught by registered dietitians) and the other group was a wait list control group. The researchers wanted to measure any change in calories consumed, percentages of calories from fat, protein, and carbohydrates, grams of fiber per 1,000 calories and change in number of self-reported binge days.

After 6 months they found no significant difference between nutrients in either group. The behavioral self management group reported a greater reduction in binge days than the control group.

Their conclusion was quite weak, finding that “Our results suggest that collecting dietary information from participants identified with binge eating disorder is challenging. Dietitians who conduct behavioral weight management programs may require additional training in identifying and understanding the psychological characteristics of participants with binge-eating disorder.”

For this conclusion to have merit, behavioral weight management programs would have to be effective and there is no research supporting that and plenty suggesting that it is not.

Overall these studies are small, short term, and quite old. The most recent was conducted about 24 years ago. It makes me wonder if there really isn’t more recent data, or if more recent data exists but didn’t support their conclusion? Also, remember that Barholomay et al. claimed “randomized controlled trials conducted among individuals with overw*ight and ob*sity demonstrate that assignment to a low-calorie weight loss diet (e.g., 1200 calories per day) produced greater decreases in binge-eating symptoms than assignment to a wait list control group.”

I think their claim is seriously overstated. First of all, they should have been clear that this was the case in short-term, small sample studies. They also failed to mention that a non-diet treatment also produced greater decreases in binge-eating symptoms than in weight list control group - I wonder why they left that out of everything?

Ok, let’s look at another claim from Bartholomay et al.. They claim that adolescent girls with BN (which they explain is “bulimia nervosa, a disorder characterized by both binge eating and extreme compensatory behaviors; American Psychiatric Association, 2013”) who were assigned to participate in a healthy dieting intervention promoting weight control through moderate reductions in caloric intake, exhibited greater reductions in binge eating than girls who were assigned to a wait list control group.

The first issue here is the claim that there is any such thing as a “healthy dieting intervention.” This is, in fact, a claim and not a fact, and it requires evidentiary support, which they don’t offer. They don’t even offer a definition of “healthy” nor proof that this intervention is, in fact, healthy by any definition. Again as we often see in research propping up the weight loss paradigm the authors have substituted “just saying stuff” for anything resembling scientific precision. In this case they are repeating a mistake originally made in the 2006 study they are citing by Burton and Stice called “Evaluation of a healthy-weight treatment program for bulimia nervosa: a preliminary randomized trial.” This study included 85 cisgender female participants with “full- and sub-threshold bulimia nervosa” who were randomly assigned to a 6-session “healthy dieting intervention” or a waitlist control group and assessed after a 3 month follow up. Their conclusion was that “These preliminary results suggest that this intervention shows potential for the treatment of bulimia nervosa and may be worthy of future refinement and evaluation. Results also provide experimental evidence that dieting behaviors do not maintain bulimia nervosa, suggesting the need to reconsider maintenance models for this eating disorder.”

In using this study to support their claim, Bartholomay et al seem to be heavily glossing over words like “preliminary” and “may be worthy.” Also, even if we believe the results (without a deep dive into them) the fact that people on diets can suppress binging (or report suppressing binging) over 9 months is a far cry from showing what impact that restriction has over the long term.

Back to Bartholomay et al., they make another claim that

“Importantly, although longitudinal studies indicate that self reported dietary restraint predicts the future onset of binge-eating symptoms (Field et al., 1999; Killen et al., 1994; Stice, 2001; Stice et al., 2002), these results stand in stark contrast with findings from experimental treatment studies testing the causal effect of dietary restriction and restraint on the maintenance of binge eating.”

This is an odd claim that, to me, is made with far too much confidence/bias. Longitudinal studies look at a longer time period, often quite a bit longer, than experimental treatments. Bartholomay et al. do not have a method to determine whether any difference is due to experimental design or simply due to the fact that experimental treatment studies capture a much shorter time frame. If what is true is that people who are on diets can suppress binge behaviors short term, but that the restriction drives additional binge behavior long-term, then the longitudinal studies could be more accurate.

Let me offer an example to help illustrate the issue: It takes from 10-40 years for symptoms of asbestos conditions to appear. Let’s stay that a study claimed “Importantly, although longitudinal studies indicate that exposure to asbestos causes mesothelioma, these results stand in stark contrast with findings from [6 month - 1.5 year] experimental treatment studies testing the causal effect of asbestos on mesothelioma, which found no relationship.”

If the effect takes longer to appear than the time over which the experiments are conducted, there is a significant risk of missing the effect. That effect may be identified by longitudinal studies.

Overall, I don’t think the studies they are citing come anywhere close to, as they claim:“challenge the assumption that dietary restriction maintains BE [binge eating] among all individuals with BED [Binge Eating Disorder]. Rather, results suggest that dietary restriction may be largely unrelated to BE maintenance in this population, and that reducing dietary restriction generally does not have the intended effect on BE frequency.”

In part 3 we’ll investigate to see if the rest of the article’s data supports their conclusions.

Did you find this post helpful? You can subscribe for free to get future posts delivered direct to your inbox, or choose a paid subscription to support the newsletter (and the work that goes into it!) and get special benefits! Click the Subscribe button below for details:

Liked the piece? Share the piece!

More researchThe Research Post

More resourcesThe Resource Post

*Note on language: I use “fat” as a neutral descriptor as used by the fat activist community, I use “ob*se” and “overw*ight” to acknowledge that these are terms that were created to medicalize and pathologize fat bodies, with roots in racism and specifically anti-Blackness. Please read Sabrina Strings’ Fearing the Black Body – the Racial Origins of Fat Phobia and Da’Shaun Harrison’