American diet culture has been macronutrient-obsessed for decades, and I’ve been exasperated by this obsession since I first started paying attention to it in the late 1990s. During this period, efforts to identify the “best” diet and the “right” way to eat have been narrowly focused on carbohydrate, fat, and protein. Which of these calorie sources are “good”? Which are “bad”? What is the optimal balance of macronutrients for health and fitness?

The reason these never-ending debates exasperate me is that it has been obvious to me from the beginning that, despite the disparate conclusions the various diet schools have came to, they all rest on the same false premise, which is that healthy eating is defined by macronutrients. Well it’s not. A healthy diet can be high- or low-carb, high- or low-fat, and high- or low-protein. To define healthy eating on the basis of the diet’s macronutrient composition is to bark up the wrong tree.

So, if macronutrients don’t matter, then what does? Quality. In the general public, the concept of quality as it relates to food is typically associated with grades of fineness, as in, “This is a high-quality cut of beef.” But nutrition scientists define quality differently. Simply put, a high-quality food type is one that is broadly associated with positive health outcomes and a high-quality diet is a diet that combines high-quality food types in a way that minimizes the risk for all diseases and conditions that have a dietary connection and maximizes longevity and other indices of good health.

What epidemiological diet research has consistently shown is that all of the food types that are commonly consumed by humans all over the world are good for us, but only when they are consumed in natural, unprocessed forms. Vegetables are good, fruit is good, nuts and seeds and plant oils are good, whole grains are good, dairy is good, and unprocessed meats and seafood are good. Processed versions of all of these foods, such as fried potatoes, are bad.

I must confess that I do enjoy it when a major scientific study comes along to validate my position and shatter the ideologies of folks like anti-carb zealot Gary Taubes who think it’s all about macronutrients. And it’s even more enjoyable when the study that does so was initiated by Gary Taubes himself!

The specific study I’m referring to is the latest in a series called the DIETFITS Randomized Clinical Trial. What’s distinctive about the study is that, although it was funded by Taubes’s Nutrition Science Initiative, which is on a mission to prove that high-fat, low-carb (HFLC) diets are healthier than other diets, it was very well designed and executed and in no way put its thumb on the scale (so to speak) to favor HFLC.

A large subject pool consisting of 609 overweight adults was separated into two groups and placed on one of two diets for 12 months: high-fat low-carb or low-fat high-carb. Importantly, though, both diets were high quality, meaning that whether they restricted fat or carbs, all of the subjects ate mostly unprocessed foods. Also, calories were not restricted, and therefore any differences in outcomes between the two diets would be directly attributable to their macronutrient composition.

The purpose of the study was to compare the effects of the two diets on insulin levels and body weight. Taubes’s pet theory is that carbohydrate causes obesity by flooding the body with insulin, which promotes fat storage. This theory would predict that the HFLC group would exhibit reduced insulin secretion and greater weight loss, with a strong correlation between these two outcomes.

That’s not how it worked out. At the end of the 12-month intervention, weight loss was significant and about equal in the two groups, with members of the low-carb group shedding just over 13 pounds, on average, and members of the low-group group dropping a little less than 12 pounds. While individual weight loss differed drastically within both groups, there was no relationship between changes in insulin secretion and weight loss in either group. In short, everything that supporters of the carbohydrate-insulin hypothesis of obesity would have expected to happen (other than weight loss within the low-carb group) didn’t happen.

There are two possible explanations for these results. One possibility is that both diets produced equal amounts of weight loss for completely different reasons—that, in other words, reduced fat intake caused weight loss via one mechanism and reduced carbohydrate intake caused weight loss via another. The other possibility is that factors common to both diets caused weight loss across groups. While the first possibility can’t be ruled out, the second possibility is much more plausible. Energy intake decreased markedly and equally in both groups despite the absence of explicit calorie restrictions, suggesting that by replacing processed (low-quality) foods with unprocessed (high-quality) foods, low-carb and low-fat dieters alike were able to fill up on fewer calories.

Predictably, the unreachable tinfoil hat-wearing anti-carb crusaders out there are now racing around with their hair on fire, crying out to anyone who will listen that this study was not a fair test of the carbohydrate-insulin hypothesis because carb intake was too high among subjects in the HFLC group. It’s true that, while these subjects did start at a very low carbohydrate intake level (just 20 grams a day), their carb consumption increased over the course of the study, ending at just over half of baseline daily carb intake for all subjects. My response to this objection is twofold:

  1. It smacks of desperation and faith-based belief. Statistically, there is enough in the results of this study to demonstrate that carbs per se are not THE cause of weight gain.


  1. The reason the subjects on the HFLC diet started eating more carbs was that they found 20 grams per day unsustainable. Even if we granted that HFLC is singularly effective for weight loss in theory (which it isn’t), how effective is it really if it cannot be practiced by real people in the real world? I’m reminded of the way that, several years ago, barefoot running fanatics defended the theory that running barefoot prevents injuries against a tidal wave of injuries caused by barefoot running. Do you remember what they used to say? “Barefoot running works if you do it right. It’s just that nobody does it right.”