Junkfood Science: All diets work (wait for the punchline)...

March 08, 2007

All diets work (wait for the punchline)...

The story of the week has appeared practically everywhere: “Atkins wins!”

Despite the glut of coverage the latest diet study has received — in mainstream media, medical literature and even the blog world — no one’s gotten the secret yet.

The press release issued by the Journal of the American Medical Association has been the script followed by most reports, giving us headlining stories such as:

Study shows Atkins diet helps women lose more weight (San Jose Mercury News)

Atkins beats other diet plans in study (Associated Press)

Atkins Diet Wins Kudos in a New Study (U.S. News)

Briefly, this $2 million clinical trial, conducted by Stanford University researchers, enrolled 311 healthy women during 2003-2004. The women were about 40 years of age, college-educated, mostly caucasian, with an average weight of 187 pounds and BMI 32 (“obese”). The women were randomly assigned to either the Atkins (very low-carbohydrate), Zone (low-carbohydrate), LEARN (low-fat, high-carbohydrate and based on national Dietary Guidelines) or Ornish diets (high-carbohydrate, low-fat vegetarian). They attended weekly classes on their diet plan, taught by a dietitian, for the first two months. To encourage maximum participation, throughout the study they received emails and telephone calls from staff members and incentive payments. Their diets were monitored by unannounced 3-day dietary recall telephone assessments using food amount books to help them report portion sizes more accurately and Stanford 7-day physical activity recalls. Their weights and other health measures were measured directly in the clinic by healthcare and laboratory professionals blinded so that they didn’t know what diet each participant had been assigned.

At the end of one year, what did they find? The story we’ve heard reported is that the Atkins diet resulted in the greatest weight loss: an average of 10.4 pounds, as compared to 5.7 pounds on LEARN, 4.8 pounds on Ornish, and 3.5 pounds on the Zone.

While a 4 3/4-pounds difference after a year of dieting is being heralded as significant, the undeniable fact is that all of the weight losses were modest at best and clinically insignificant. And all were considerably less than the diets claim and certainly what most consumers believe will be their pay off after a year of effort. The Zone, for example, says the average dieter may lose 8 to 10 pounds a month.

The U.S. government’s Dietary Guidelines tell us that all we need to do to lose a pound is to eat 3,500 calories less or burn 3,500 calories more. Given the calorie reductions made by all of the dieters in this study, they “should” have each lost about 36 pounds. They didn’t, of course. But the results were not surprising.

The actual findings

The impression given by the news reports this week is that all the dieters in this study were steadily losing weight and the Atkins dieters came out ahead of the pack. Reporting just the weight loss differences at the end of a year is misleading, however, and paints a very different picture from what was actually seen.

It would be more accurate to say that the Atkins dieters experienced the biggest weight yo-yoing, and the fastest and most significant weight regain.

During the last 6 months of the trial, all of the dieters regardless of the diet plan, experienced weight regain. At the end of the first year, they were already at or above the weight they lost during the first 2 months of the study. The Atkins dieters lost weight fastest at the beginning, but the trajectory of their weight regain was also the most dramatic. The study period simply ended before their skyrocketing weights jumped above the other diets.

While I can’t pull the weight loss graph from the study, I’ll describe it: It looks like a giant “V.”

This is consistent with other diet studies which have shown that the speed and magnitude of weight loss, regardless of the method, corresponds with the speed and magnitude of the weight regain. The bigger and faster the loss, the bigger and faster the regain.

Had the diet study been continued for 5 years, which all obesity experts recognize is obligatory before making any conclusions about a weight loss intervention, the results would be no different than a century of diet studies.

All diets work (wait for the punchline)...

The fact is, virtually all weight loss diets work — the specific diet plan is irrelevant. They are merely whatever gimmick people will buy. By cutting calories, most everyone will lose weight initially. Just long enough for those before-and-after photos.

But, regardless of the diet or weight loss method, weight regain within five years is guaranteed for virtually everyone.

This was not lost on the researchers, either. Buried in the discussion section they mentioned that “weight-loss trajectories for each group had not stabilized...suggest that longer follow-up would likely have resulted in progressively diminished group differences.”

Only long-term results, after weights have stabilized, are relevant when evaluating any diet and, more importantly, any actual impact on health outcomes. This short-term study was unable to credibly determine health outcomes — beneficial or harmful — so we’re not even going to bother going there.

Doesn’t it ever strike you odd that virtually no popular diet study ever lasts even five years? In a century of diets and weight loss programs, you would think more than one would have. Yet, it is well-known and openly acknowledged among professionals that to demonstrate any degree of effectiveness for a weight loss intervention, people need to be followed for at least five years.

Popular diet studies commonly stop...conveniently when the initial weight losses still look good and before the weight regain and health complications manifest themselves.

[There is a notable recent exception: a careful, well-designed, long-term clinical diet trial. We’ll discuss it in an upcoming post. Hint: Not surprising in the least, it was a dismal failure on all counts, which is probably why you heard little about it.]

A few other points clearly obvious in this JAMA study have not been even mentioned in the news or medical literature. These startling insights are important to understanding this study and putting it into perspective with the body of evidence.

Calories in-Calories out myth of weight maintenance (aka “eat less and exercise more”)

While the participants’ adherence to the diets waned during the year, they were all still eating notably fewer calories than before they started dieting — even while they were regaining the weight lost during the first months of their diets.

The study states: “There was a significant mean decrease in reported energy intake at all postrandomized time points: -497, -387 and -351 at 2, 6, and 12 months, respectively for all groups combined.”

The dieters on every one of the plans who completed the study were eating fewer daily calories and at 12 months their daily calorie intakes were still: -289 (Atkins), -381 (Zone), -271 (LEARN), and -345 (Ornish). Yet they all were steadily regaining their weight over the last 6 months of the first year.

All dieting groups had also maintained higher energy expenditures through increased physical activity, continuing to burn off around an additional 86 kcal/day at the one-year mark over their activity levels before they began dieting.

This concurs with the body of clinical evidence showing that due to biological processes outside of people’s control, it becomes increasingly difficult to continue to lose weight, and especially to maintain any significant weight loss, as it requires increasingly extreme calorie restrictions and exercise. Further weight loss much below one’s natural weight range can become nearly impossible short of total fasting.

There are rare individuals who manage, with extraordinary and obsessive effort, to remake themselves by dieting. There are others who can starve themselves to death by an act of will, and there have been saints who could sit for decades at the top of a pillar. These examples have little to teach the majority of people, who have other things to do with their lives. — Dr. William Bennett, M.D. and Joel Gurin (Dieter’s Dilemma: Eating Less and Weighing More)

So naturally thin people can lose ten pounds or so relatively easily and believe that fat people can just as easily repeat that ten times to be thin, but it doesn’t work that way.


Among the attempts to discredit or excuse this study’s findings are claims that the dieters didn’t follow the diets exactly. This overlooks the fact that the participants received 8 weeks of counseling from a dietitian on their prescribed diet plans; followup calls, emails and office visits; and incentives, which is much more than anyone else buying a diet book gets. If anything, as these researchers reported, this study is about as good as it gets in the real world.

The researchers noted that “we believe that the adherence levels obtained are a fair representation of the diets and variations in micronutrient intake under realistic conditions and, therefore, increase the external validity of the findings.”


Even though the dieters were paid increasingly more money the longer they stayed on in the study, about 20% dropped out. Attrition rates in most diet studies are considerably higher, despite the fact that clinical trial participants are typically highly motivated volunteers.

For instance, in another randomized trial comparing four popular diets conducted at Boston medical centers from the summer of 2000 to January, 2002, the drop-out rate was 47% for Atkins, 50% for Ornish, and 35% for the Zone and Weight Watchers. This was despite the fact that all of the participants in the study had added motivation because of diagnosed health risks. At the end of one year, by the way, the Atkins dieters in that study had lost the least amount of weight (a mere 4 1/2 pounds) versus roughly 7 pounds for the other plans. But this was another study too short to derive any meaningful conclusions from.

It just keeps going and going....

Los Angeles doctor, Lulu Hunt Peters, A.B., M.D. published the first weight loss book based on the calorie theory in 1918. Her book, Diet and Health With Key to the Calories, recommended calorie reduction and increased exercise. Her’s became the first best-selling diet book. The first bathroom scale, created by the Continental Scale Company, was introduced the following year and America was off and dieting.

There are some 7,500 diet books on the shelves today, each with it’s own claims about how to lose weight and each claiming their plan works. Atkins first introduced his approach in his book, Dr. Atkins’ Diet Revolution, in 1972. Dr. Kelly Brownell wrote the first version of the LEARN Program in 1976. Dr. Barry Sears first published The Zone in 1995. And Dr. Dean Ornish’s Eat More, Weigh Less first hit bookstores in 2000.

According to the latest national survey conducted for the Partnership for Essential Nutrition (a coalition of organizations promoting the government’s programs, policies and research on healthy eating and weight loss), 45% of Americans have dieted in the past five years, including 41% of adults still actively trying to lose weight. The weight loss secrets recommended by the government are to eat less in accordance with the Food Pyramid and exercise more; not unlike those of Peters 90 years ago.

In the past century, the thousands of diets and weight loss prescriptions have had more than enough time to prove themselves or, at least, for one to show itself to actually work. Not one has ever been able to. It is not willpower alone that prevents long-term dieting success, although that’s popularly believed. It’s condoned to disregard the biological realities that largely determine our size. Additionally, thinner people don’t fully grasp the intensity and unfathomable range of negative effects brought on by dieting. Sadly, it’s a rare fat woman in our culture who doesn’t want to lose weight and hasn’t exhibited heroic feats of willpower during countless years of dieting.

The multi-billion dollar diet industry has a remarkable business model, unlike any other industry. Despite a century of proven failures and unsound premises behind it, it has managed to convince the public that its diets work and that it is the customer’s fault when they fail. Can you imagine this tactic working in the automobile industry? If consumers bought a car that failed within a year, let alone several more, they would refuse to buy more and not feel guilty.

We recently looked at what motivates healthcare professionals, who (should) know better, to continue to promote something that’s has been shown to be ineffective after hundreds of the most careful, quality studies. That post also looked at what differentiates science from junk science, and why true scientists often react with disinterest or distaste whenever the same junk science resurfaces. They know it is either based on biophysiological implausabilities or has been disproven long ago and they’ve moved on.

That’s exactly how obesity researchers feel about this and other diet studies — those not selling their own diet books, that is; and there are still a few such professionals. It is unproductive to become mired in the mud-slinging resulting from this study, as various diet interests attack their competitors and promote their beliefs that their diet is the best. The undeniable fact is that in nearly a century, despite billions of dollars behind them, no diet or weight loss measure has been demonstrated to work: to result in sustained weight loss or significant health improvements in people. Short-term stunts are irrelevant and the scant 1% exceptions to the rule do not make for an effective intervention.

Instead, the body of evidence is most significant that weight loss prescriptions increase risks for heart disease, strokes, arrhythmias, high blood pressure, osteoporosis, cancers, weight gain and weight cycling, poor self esteem, eating disorders, nutritional shortfalls, weight discrimination, and higher mortality. Fat people continue to be blamed for the failure of diets, leading to negative attitudes towards fat people as lacking in willpower or being irresponsible, even from their own healthcare providers — attitudes which have been shown be significant barriers to seeking and getting quality medical care.

Whether healthcare professionals and consumers will ever take a lesson from alternative modalities and demand evidence-based health care and clinical practice guidelines is yet to be seen. But how long will we continue to find entertaining and reward lucrative profits, let alone spend billions of dollars of our limited healthcare resources, for diet fads that lack credible scientific support to suggest they are helpful, safe or effective — or ever will be?

© 2007 Sandy Szwarc. All rights reserved.

[Illustrations: Lulu Hunt Peters, A.B., M.D.]

Bookmark and Share