Why Don’t You Like My Studies?

credible hulkOne of the reasons that I’m no longer interested in attempting weight loss is that my review of the literature informs me that it simply has no basis in evidence as being an effective way to either lose weight or become more healthy (which are two separate things).  When I say that, people often object insisting that there are studies where people have lost weight.

The problem is that any old research where a couple of people lost weight won’t do.  The research we would need for weight loss to meet the criteria of an evidence-based medical intervention is twofold.  First, we would need a study where the majority of the participants lost the amount of weight that we are told we need to lose to change our health and maintain that weight loss long term (over 5 years).  If we had those studies – and we don’t –  we would then need some proof that weight loss actually caused health improvements – and a new study already brings that into question.

This rules out the National Weight Control Registry because they’ve chosen to study 10,000 people who experienced weight loss while completely ignoring the up to 800,000,000 failed attempts that happened in the same time frame.  When they say things like “eating breakfast contributes to weight loss” what they actually mean is that they asked the 10,000 people who succeeded what they did, and a majority of them said that they ate breakfast.  Note that they didn’t ask how many of the up to 800,000,000 people who did not lose weight also ate breakfast – that would be important information to have since if a majority of the people who didn’t lose weight also ate breakfast then breakfast may have absolutely nothing to do with it.

Imagine if I got together everyone who had survived a skydiving accident when their parachute didn’t open and started looking for things they have in common.  Even if every single one of them wore a green shirt and had oatmeal for breakfast, I cannot say that wearing a green shirt and eating oatmeal will allow you to survive a skydiving accident, or start Ragen’s school of skydiving without a parachute, free green shirt and oatmeal with every jump.  When your sample is the statistical anomaly your research is useless, and when all you’re looking for is random coincidence among a select group of participants you probably shouldn’t call what you do research at all.

Other times people bring up studies where phase 1 was weight loss and phase 2 was maintenance, the study lost between 40% and 70%  of participants during or after phase one, and then the researchers continued on as if the remaining people were the complete study group.  Not ok. Why did all of those people quit?  How will their experience be accounted for? Often the remaining subjects start gaining back the weight they lost so that at the end of phase 2 the average participant has gained back half of their weight with a net loss of less than 10 pounds.  Or they only follow up for a year or two when we know that most people gain their weight back by year 5.

People list study after study and all of them have one or more of the above problems, which I, or someone else in the discussion, points out.  At that point, the person listing the studies often gets frustrated and says something like “Why don’t you like my studies? ” or “You just don’t want to believe.”   If they examined it I think they’d find that their frustration isn’t with me, it’s with the fact that they’ve been sold a lie and they bought it at full retail price.

I certainly know that frustration, when I did my first literature review of weight loss research I expected to find that all diets worked – I was just looking for the “best” one, the one that had the most solid success.  I was so shocked at what I found that I read through all of the literature again.  I simply couldn’t believe that this thing – weight loss – that had been marketed to me more aggressively than anything else in my life had no basis in evidence.  I couldn’t believe that doctors had been giving me an intervention which had been shown repeatedly to almost always end in failure, and the majority of time had the exact opposite of the intended result.  When I found out that there weren’t even any studies that showed that weight loss caused changes in health I was just stunned.

It took me a lot of time and a lot of work to accept the truth.  It was hard to find out that I’d been lied to (on purpose and inadvertently), it was hard to find out that the thing that I’d been promised would solve all of my problems was never going to happen.  In many ways, at least for me, Health at Every Size was about giving up, but that’s what I do when I find out that I’ve been harboring a mistaken belief.  That’s what scientists (well, good scientists) do when their evidence does not support their hypothesis (however strongly held or widely believed it might be.)  They don’t suspend the rules or research and logic and argue for a belief that they can’t support with evidence.   It’s not that I don’t like your studies, it’s that they are insufficient to change my mind about my personal prioritization and path to health.  Of course, nobody is obligated to do the same things that I do, only to respect my choices.

Like my blog?  Looking for some holiday support or gifts?  Here’s more of my stuff!

The Book:  Fat:  The Owner’s Manual  The E-Book is Name Your Own Price! Click here for details

Become a member: For just ten bucks a month you can keep this blog ad-free, support the activism work I do, and get deals from cool businesses Click here for details

Dance Classes:  Buy the Dance Class DVDs or download individual classes – Every Body Dance Now! Click here for details 

Interviews with Amazing Activists!!  Help Activists tell our movement’s history in their own words.  Support In Our Own Words:  A Fat Activist History Project!

If my selling things on the blog makes you uncomfortable, you might want to check out this post.  Thanks for reading! ~Ragen

20 thoughts on “Why Don’t You Like My Studies?

  1. Atta girl! Another home run! I love that you cut through all the nonsense and just tell it like it is. It’s beautiful thing to see.

  2. Damn, there’s just no arguing with this, is there?!

    The only problem I see is actually attempting to explain all this to a EAT LESS LOSE MORE GET HEALTHY blank-eyed devotee without having their pickled brains turn everything into white noise half-way through. Buying into a chant or a specific dogma whole heartedly usually means turning off the thought processes. Not much headway to be made once that kicks in…

  3. Also, even if you ignore attrition rate and far too short follow-up periods, there is still the issue that weight loss of a few pounds can easily be statistically significant if your participant pool is large enough. I have never ever seen any evidence that an intervention aimed at weight loss was successful at reducing the average participant’s BMI enough to go from “morbidly obese” to “normal” or even just “overweight”, not even in the short term. Hell, even people in the National Weight Control Registry (a quite exclusionary group as Ragen has explained above) can have lost as little as 30 lbs. And yes, for someone with a BMI > 40 or, hell, even > 35, 30 lbs. often IS not a huge loss compared to total body mass. In fact, for me losing 30 lbs. would mean a drop in BMI of less than 5 points. Hell, according to this measure, I would be a “weight loss success” because I have lost the weight “overshoot” that I gained when I rebounded after my last diet – yet no doctor would actually see me as a success because I am still quite fat. Plus, if they did, how would that make sense considering that I only lost weight which I probably would have never gained without repeated dieting in the first place?

    What is more, doctors (and other medical professionals) have repeatedly told me to just “lose the same amount again”. There is no evidence in the literature to support this advice – not even if one makes the highly questionable assumptions that people dropping out of weight loss studies lost/regained weight at the same rate as people staying in the studies in question and that people who underwent weight loss stop regaining after a one- or two-year follow up.

  4. I like to follow the results of my study of how my body reacts to calorie restriction. It sometimes leads to minor temporary weight loss… but it invariably leads to an extremely foul temper with a very short fuse.

    To paraphrase the Incredible Hulk: you wouldn’t like me when I’m hungry.

    1. I had the same experience (the calorie restriction causing foul temper). No one associated it with dieting though, since I was still fat and gaining. They blamed me for being a “harpy”.

  5. I wrote about studies on my blog today, too! The conclusion I came to is that I’m going to live my life in the best, healthiest way for me, despite what any study says.

    Thanks for another great post, Ragen. As always, you hit the nail on the head.

  6. I do not like their studies because they are only half of the studies I do like. Their studies find that X participants lose ten pounds in two months. My studies also find that X participants lost ten pounds after two months, BUT THEN gain every one of them back after two years, even though they are still on the diet plan. Their studies not only lack the “but then,” they deliberately excise it with the express intent to decieve.

    RE: “You just don’t want to believe!”, wow. What reality are they living in? Most fat people not only want to believe, we DID, and we lived like the myths were true for years, often to the detriment of our health. I applaud and admire the person who can look diet culture in the eye, recognize it for the sham it is, and say “screw you!” right off the bat. I was not that person. I had to see myself gain weight on dangerously low-calorie diet before my suspension of disbelief was finally strained to the breaking point and I began seriously researching WHY my experiences differed so much from the ‘eat less-move more-lose weight-live happily ever after’ myth I’d been sold.

    Turns out you can’t believe a myth true. Who knew.

  7. I have a friend who believes every “study” she reads about obesity absolutely causing poor health, etc. yet when I mention studies showing the opposite her reply is always “it was probably written by a fat person making excuses.” That seems to be the general public opinion.

    1. Confirming the general public has never even *glanced* at the material we’re discussing, since Glenn Gaesser, Linda Bacon, Paul Ernsberger, and Abigail Saguay are all thin.

  8. I’m a biologist/biochemist myself, and I’ve collaborated with people in the clinic as well as doing my own *in vitro*, micro, and animal work. While I agree with you that much of the body of weight loss research (and, frankly, almost all nutritional research done until quite recently) is problematic in terms of both methodology and in terms of the sweeping conclusions the popular media makes about the results, I think there are a lot of practical concerns to be kept in mind when talking about cutoffs for considering the findings of a study. Frankly, science is hard, and science using people is more so. Studies are expensive, the time frame you’re considering sufficient is longer than the average research grant in a lot of fields, and the average biomedical study on human subjects has dropout rates ranging anywhere from 25% to 50% or even higher, depending on the demands of the study (how frequently follow-up happens, how invasive the tests are, how demanding the everyday changes in the study are).

    Follow-up is expensive. The more comprehensive the metrics you’re using (say, blood glucose, glucose tolerance, blood lipids, etc. to monitor health, as opposed to simply half-assing it with BMI and assuming from there) the more expensive follow up becomes. The more invasive the metrics (drawing blood, exercise stress tests, MRIs, etc.) the more people will drop out. And if you don’t get a renewal on your two-year grant? Well, now it’s two years of follow-up and not five, because you don’t have any money.(Grant success rates are notoriously poor. I was once at a conference where someone made a joke that you’d have better odds if you took your grant to Vegas and put it on red than you would getting your renewal. We looked at some figures, and it turns out in fact it would be *much* better odds.)

    Besides that, things happen–people move, they die, they acquire medical issues or suffer accidents that make them no longer fit the criteria for the study, etc.. Or they simply drop. Demanding a study have a low dropout rate is like demanding a pink unicorn, regardless of the field. For example, in one study on dropout rates in urology studies with no lifestyle changes demanded and relatively noninvasive methods, about 8% of the sample died during follow up and another 28% dropped out. For musculo-skeletal disorders, dropout rates range from 7 to 57% depending on a variety of factors, and again, any study that has participants adhere to any lifestyle changes will almost necessarily have a higher dropout rate than one which simply looks for correlates. For example, in a study on dropout rates in exercise studies (which did not look for weight loss, just participation in an exercise program), 45% dropped out early on and that figure neared 60% during follow up even though the regimen was only *two* years long. Moreover, factors like being on medications and mental state (depression, job satisfaction, etc.) influence dropout rates across the board.

    Frankly, I would expect absolutely any study that asked for maintaining lifestyle changes, looked for half a decade of follow-up, and focused on a population which is treated very poorly by society at large and has higher rates of medication use to have a very high dropout rate before you even told me what the study was for. This is precisely why studies must enroll more participants than, strictly speaking, would be needed for statistical analysis.

    But this isn’t a reason to specifically discount weight loss studies, but rather to consider *all* human studies in the context of other research in the area, including *in vitro* work, work in animal models, and other human studies which have been done in the field.

    1. Hi Cass,

      Thanks for your comment, I wanted to address a couple of things. .

      First I want to clarify that I didn’t demand that studies have a low dropout rates, I suggested that those who dropped out not be ignored in study conclusions. For example, there is a study where almost 70% of participants dropped out, but the studies conclusions suggest that since the 30% of remaining subjects lost a couple pounds the weight loss intervention was successful, ignoring the fact that almost 3/4 of their participants may have dropped out because it wasn’t successful.

      Second, I am well aware of the difficulties of doing research – especially that involving people – however, weight loss is an intervention being recommended to over 60% of the public so one would think that we would have a decent basis in evidence for doing so, we simply don’t but everyone seems to think that we do. While it’s true that most study grants do not fund a five year follow up, much of the weight loss research is funded by diet companies themselves. When the FDA asked them for longer follow up they refused stating that it would “be too depressing for our clients.”

      If people said that it’s too difficult to get the research we would need to provide evidence of efficacy for weight loss, that the research that we have suggests an almost 100% rate of weight regain and that a majority of people will have the exact opposite of the intended effect, but that they think we should try anyway that would be honest, but that’s not what they are saying. The message that successful weight loss has a basis in evidence is simply wrong, for the reasons I stated in the blog.



      1. Not only that, but

        a) it is well documented that people regain after weight loss. Even the remaining sample in weight loss studies, which on average might still be below the starting weight at 12- or 18-months follow-up, clearly shows an upward trend in weight after the initial intervention has ended. See for example http://www.indiana.edu/~k536/articles/behavior/failure%20Garner%201991.pdf

        b) physiological mechanisms (decreased metabolic rate combined with increased hunger) have been identified that can explain regain after dieting, see http://www.fatnutritionist.com/index.php/why-diets-dont-work/ for a summary as well as references

        c) quite a few people actually regain more than they lost after dieting, see http://motivatedandfit.com/wp-content/uploads/2010/03/Diets_dont_work.pdf

        d) there are at least some researchers who have suggested that dieting often leads to eating disorders. (Sorry, but I do not have a source handy and limited time to look for it in my files/online – but there are papers out there, and they are not that hard to locate).

        The last point is not as clear cut as the others. But the first three points are very clearly documented in the literature. This does not mean that there are not a relatively small number of people out there who actually maintain a weight loss. But it clearly, clearly does show that long-term maintenance of intentional weight loss is a) very, very unlikely, especially if we are talking about more than a few pounds and b) probably not something that should be routinely recommended to people – not by medical professionals and particularly not by lay people.

        (As a side note: Personally, I would be okay if weight loss was recommended as ONE treatment option to fat people who experience health problems that might be exacerbated by weight if these people were also informed at the same time about the actual likelihood of long-term success as well as of the potential negative consequences. But it is highly irresponsible to recommend it as the only option and to deny any dangers associated with it – which frequently happens. It is even more irresponsible to recommend it to people who have actually already gone through the loss-regain cycle repeatedly.)

        1. Totally agree. It’s weird how people have to have such black-and-white thinking on this issue. When people talking about fat acceptance make the claim that 90% (or whatever huge number) of diets fail, that is LITERALLY what it means – that the vast majority of people regain the weight, but that a small fraction may maintain weight loss. But for some reason, when people hear us saying that, they interpret it as “NO ONE EVER LOSES WEIGHT, EVAR.” That is not what we are saying – we are saying that the success rate of weight loss is low enough that it is irresponsible to continue using it as a first-line health promotion strategy for the majority of the population. And especially for the many people who have a demonstrated history of weight loss and regain to higher levels, it is totally inappropriate. And that for people who wish not to chase the golden unicorn of weight loss, for whatever reason, other treatments should be made available. Reading comprehension seems to be a really big problem here. Magical thinking is a hell of a drug.

  9. I tried to be part of the NWCR. The forms you have to fill out are long, very tedious and time-consuming. So you can add “self-selecting group of OCD people who like filling out forms” to the description of the subjects of the NWCR. (perhaps this self-selecting group of detail-oriented people are also those who can keep up crazy things like calorie counting or food journaling for life) I also found that the questions tried to steer me toward particular answers and did not give me a viable way to tell the truth, so I didn’t even complete the first packet of questionnaires.

    1. I am actually a member of the NWCR. I don’t fit into the typical mold of a member, so I used comments wherever I could to tell the truth. Since I joined last year, I’ve only gotten one follow up study, which was about sleep.

      1. It was 2011 when I joined and the first packet they sent was the size of a book! I got about half way through it and gave up.

  10. I saw the exchange on FB that most likely prompted this post, and I just wanted to say one thing to the person who was posting studies: representative sample. As Ragen said, studies that ONLY follow a self-selected group of people who have already proven successful at losing weight are not applicable to the rest of the population. They might be useful for other purposes, such as studying whether there are differences between that small minority and the rest of the population, but they cannot be used to prove that weight loss works for the majority of people. It would be like using an apple to prove that oranges exist. Doesn’t work. Representative sample. The end.

    1. As Ragen said, studies that ONLY follow a self-selected group of people who have already proven successful at losing weight are not applicable to the rest of the population. They might be useful for other purposes, such as studying whether there are differences between that small minority and the rest of the population, but they cannot be used to prove that weight loss works for the majority of people.

      Also, in order to make those studies with self-selected groups useful, we have to start looking beyond mere behavioral differences. It’s all nice and well to know that x group of people eats breakfast and exercises an hour every day. And if you can show that group y actually does not do these things/does not do them as often while having worse outcomes, and you can also show that these outcomes are actually caused by the difference in behavior (and not, e.g., the other way round) you might actually be onto something. But the truly interesting question is WHY these groups behave differently. Sure, it COULD be education, it COULD be self-control, and it COULD be motivation. But it also could be a less supportive social environment, it could be less access to material resources, and it could be *gasp* an actual difference in biology, which might cause one group of people to be hungry in the morning, while the other group isn’t. It also could be a combination or interaction of all those factors. (Also… since I am already on a science and logic rant here: Just because it requires self-control to lose weight and keep it off does not mean that thin people have more self-control than fat people. It means that formerly fat people who kept the weight off might have more self-control on average than people who do not artificially keep their weight down. That’s a very different thing.)

Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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

This site uses Akismet to reduce spam. Learn how your comment data is processed.