I read an article in Slate last week called “Are Diets Just Placebos?“, and though it’s taken me a week to get around to it, I wanted to write a quick post about it with my thoughts. If you haven’t read the article, it’s fairly quick.
The author of the article (Erik Vance) hypothesizes that diets could conceivably work only (or at least partially) through a placebo effect—that is, we get the results we expect to get. It’s an interesting perspective and worth exploring, but I think ultimately there’s significantly more evidence to suggest that success and failure in the realm of dieting can be traced back to simple human error.
Vance starts by arguing that no single diet has proven to be more successful than any other diet, and that because of that there must be some reason why.
If no diet has turned out to be a silver bullet for weight loss, then what could explain why some of them at least seem to work, at least for some time? … Perhaps it’s not the contents of the diet that matters. Perhaps it’s simply the act of dieting. Is it possible that, rather than the specifics of the food regime you undertake, it’s the mere act of starting a diet—any diet—that makes you thinner? Could it be that the inherent placebo effect that comes with any diet is what’s causing you to lose weight?
There is a huge logic gap here. It is completely possible, even probable, that the “mere act of starting a diet” will make you thinner—but it’s unlikely to be because our brain is being “tricked” into losing weight and much more likely due to changed dietary behaviors. Because while the specifics of different diets are often quite different, the root behaviors successful diets encourage (eat fewer calories, eat more unprocessed foods, eat less processed foods, be mindful of your diet, exercise more, etc.) remain the same. Since motivation is generally high in the first days to weeks of a diet, people are usually successful at these goals and lose weight regardless of the specific “rules” of the diet.
Vance does bring in some science to defend his position:
First, he mentions that FDA trials on weight loss drugs typically observe around 5 pounds (2.5 kilograms) of weight loss, even in the placebo arm (the weight loss drugs all outperformed placebos, however). But, as he mentions shortly afterwards, most of these trials also imposed caloric restrictions—and also very likely came with at least some dietary counseling to help the participants accurately measure caloric intake. Both of these interventions by themselves are capable of causing weight loss; indeed, whenever I see a client for weight loss, these are the primary two ways I am able to help them achieve their goals: restricting their calories and helping them count calories accurately. So I’m not convinced here.
Second, he brings up a few studies that demonstrate that our expectations shape our hormonal response to foods. These studies are interesting from the perspective of understanding how our psychology can influence our hormonal response to food, but Vance overinflates the importance of their results.
For example, in the “landmark study” on the different ghrelin responses to identical shakes calorically labeled as either “sensible” or “indulgent”, the authors report a significant difference in ghrelin (a hormone related to hunger) but no subjective difference in ratings to hunger. This tells us that our perceptions may at least partially influence our hormonal response—we know this happens elsewhere, too, such as with the small increase in insulin upon tasting something sweet—but since hunger itself wasn’t modified we should be skeptical that we could actually influence weight loss one way or another by mislabeling foods.
The other study Vance mentions is similarly misleading. In this study, participants were fed either a solid or a liquid and told that it would either remain the same upon entering the stomach (liquid-to-liquid, solid-to-solid) or change upon entering the stomach (liquid-to-solid, solid-to-liquid). The trickery was achieved visually by using gelatin, which will solidify from a powder when immersed in cold water and melt to a liquid from a solid when immersed in warm water.
In his article, Vance makes it sound like the liquid-to-solid groups had equivalent responses to the solid-to-solid group and that the solid-to-liquid group had an equivalent response as the liquid-to-liquid group—but this isn’t what happened. Instead, what the study demonstrates is the our expectations were able to significantly alter our hormonal and physical response to a meal, but only in a clinically marginal way. The groups that were deceived about what would happen in their stomach had slightly different responses, but overall those who ate the solid gelatin had responses that much more closely resembled the other solid group and those who consumed the gelatin-free liquid responded much more similarly to the other liquid group. Again, this demonstrates that our expectations can mildly alter how we process food, but cannot change it in a way that is likely to result in placebic weight loss.
The final point Vance brings up is that most diets work for a short period of time and then cease to work—exactly like placebos. Except that that’s not really how placebos work because placebo effects don’t change objective outcomes. Indeed, many much more knowledgeable people on the topic than myself have argued that the placebo effect is just an “artifact” of the way we run trials and that such an effect doesn’t really exist, it just appears because of the way a trial has been set up, the way the data is interpreted, and other such factors. In this light, placebo effects could never significantly change a person’s weight because they’re a false effect.
That brings us back to my own point, which is that the success and failure of any given diet are ultimately due not to the placebo effect but to human error. I don’t say this in a cruel way, but rather as a reflection on the significant challenge that changing lifelong habits entails, and the many ways that humans can (and frequently do) underestimate their actual caloric intake (and output from exercise).
When diet and exercise are fully controlled—which rarely happens due to the challenge, and is probably not a good way to change habits long-term for humans anyway—it’s easy to get a person to lose weight (easy on the person in control, not necessarily on the dieter). It is easy, for example, to get an obese dog to lose weight simply by reducing its food intake and increasing the amount of time it spends exercising. In such a situation where one party has almost zero control over their energy intake and output, weight loss happens in a predictable manner. It’s only when you insert the human psyche into the equation that things get challenging, especially when motivation to count calories starts to wane and hunger pangs start to bite harder.
Losing weight is hard; we shouldn’t make it more so by suggesting there’s an internal element of “belief” that makes it easier or harder to lose weight. If your goal is weight loss, you need to understand that caloric balance and the motivation to accurately stick to a predetermined goal have much more to do with success than whether you believe this or that diet works better.
Vance finishes his article by writing “Sometimes belief is as good as the real thing.” I disagree. Belief doesn’t make homeopathy effective, it doesn’t make ineffective supplements work, and it doesn’t allow a person to lose weight. If you have a real problem, real solutions always work better than make-believe ones.