Diet Wars

by Paul Rogers

Tuna, beans, potatoes, olives, capers, tomatoes salad - Photo by WordRidden

Tuna, beans, potatoes, olives, capers, tomatoes salad - Photo by WordRidden

The diet wars are over. Yes, after an exhaustive 10 years of argument and counter-argument about low-fat versus low-carb diets and everything in between, most nutritionists are now coming around to the idea that providing fundamental healthy eating patterns are observed, macronutrient percentages of fat, protein and carbohydrate can vary quite considerably to suit your particular preferences, activity levels and perhaps response peculiarities — and provide weight management and health protection.

This has been encouraged by the recently published New England Journal of Medicine (NEJM) study that compared an American Heart Association (AHA) low-fat diet with Atkins and Mediterranean diets, and which was reported by some health journalists to show that the Atkins diet had superior outcomes for weight loss and heart disease risk factors. In fact, this was a very odd study and few meaningful differences emerged from the diet comparisons.

Even so, it prompted a feast of headlines in the diet and health news, blogs and mags because many analysts considered it to be almost a nail in the coffin of low-fat diets, and, perhaps another life for Atkins low-carb, and certainly for Mediterranean type diets, which traditionally are 35 to 40% fat.

The results tended to show very moderately better outcomes for the Atkins and Mediterranean diets in weight loss and in cardiovascular disease measures like blood cholesterol, glucose and other markers, but the study design was controversial.

The NEJM Diet Study – Questionable Design

Frankly, as someone who has been reading the NEJM, on and off, for over 30 years, I was a little amused, and bemused, that this study was actually published. I really don’t have a problem that it was sponsored by the Atkins Foundation, but I do object to the ‘not so subtle’ manipulation of the study parameters.

The Atkins diet group was encouraged to “choose vegetarian sources of fat and protein and to avoid trans fat.” Since when has Atkins dieting included this advice on choosing vegetarian foods? And yet this was compared it to a Mediterranean diet and a standard low-fat diet of the American Heart Association (AHA) in groups that were not given the same advice. Why not ask all groups to “choose vegetarian sources of fat and protein and to avoid trans fat”? Would not that be the obvious thing to do to make it a fair comparison — or ask none. It’s not as if vegetarian eating is at the core of the Atkins diet.

The main differences between the AHA and Mediterranean diets were that the Mediterranean diet group was encouraged to eat a handful of nuts and some olive oil each day — which more or less made up the 5% difference in fat — 30% fat for AHA and 35% fat for Mediterranean. The Mediterranean diet group was also advised to eat mostly chicken and fish instead of red meat.

All three diets discouraged consumption of cholesterol, trans fat and saturated fat one way or another, and had a similar intake of saturated fat of around 10 to 12 percent, which is not particularly low, but better than the 15 to 20% seen in some Western style fast-food type diets.

The maximum difference in weight loss was trivial — only around a 2 kilograms difference from low-fat to low-carb over the two years of the study. There was also an 8% difference in total cholesterol to HDL cholesterol ratio favouring the Atkins diet, which is useful but not dramatic. It’s easy to speculate that a little refinement of dietary advice would have seen hardly any difference in these outcomes. All three diet groups were advised and encouraged to eat lots of fruit and vegetables and mono- and polyunsaturated fats. (Even the Atkins group via the choice of vegetable protein and fats.)

Really, other than some differences in fat and carbohydrate percentages, one could conclude that there was already a consensus being reached.

AHA Diet, Straw Men and Moving Targets

A straw man argument is one in which you put up a premise that you know is easily attacked or rebutted, especially if you move your position ever so slightly that you hope no one will notice. That’s what has occurred in this study. If the Atkins people were so confident, why not compare the old Atkins diet (sausages, bacon and eggs please), with a modified AHA diet of around 25% fat, mostly mono and polyunsaturated, a little higher in protein, and high in fruit and vegetables, nuts, beans and whole grains, low in red meat and refined carbohydrates and added sugars? Something more like a combination of the old AHA step 2 diet and elements of David Jenkins’ Portfolio diet. Many of these diet comparison studies simply don’t proscribe sufficient stringency for the low-fat diets. Clearly, they need to be high in fibre, perhaps low in glycemic index and low in added sugars. A low-fat diet with lots of white bread, Coke and pop tarts is not what should be studied here.

The unique advantage of the NEJM study was the controlled cafeteria environment that went some way toward validating food consumed in each group. A golden opportunity was missed.

John Tierney over at TierneyLabs at the New York Times published a piece by Dean Ornish, high-priest of very low-fat diets for cardiovascular risk reduction. One of the things that Ornish suggests is that the ‘diet wars’ are pretty much over. Consensus is more or less achieved — at least by a wide spectrum of nutritional scientists, which excludes the more rabid fad diet proponents of course. If you’re keen, read Ornish’s views of the Tierney and Gary Taubes articles — and the responses and comments. Talk about Diet Wars!

In any case, I agree strongly with Ornish’s sentiments about consensus. So, what would that dietary consensus be?

The Diet Wars Really Are Over – Consensus

The seminal consensus position is that you don’t have to be too stringent about macronutrient percentages of fat, carbohydrate and protein within a reasonable range, as long as the basic principles of healthy eating are observed. Athletes and heavy exercisers have special requirements in relation to energy intake, carbohydrate and possibly protein, which allows them more choice, especially in relation to refined carbohydrates and sugars. Other groups with special needs may also benefit from moving macronutrient content around within a healthy framework.

Here are the basic healthy eating principles overwhelmingly supported not only by nutritionists, dietitians and heart, cancer and obesity specialists, but for which there is a strong theoretical and evidence base, even if not a perfect one.

  • Know how many calories you should eat to maintain your normal weight and don’t overeat. To lose weight, eat somewhat less than this and exercise more.
  • Eat mostly plant foods including abundant fruits, vegetables, nuts, beans and wholegrains.
  • Eat few refined carbohydrates or added sugars (including fruit juices)
  • Don’t eat too much red meat or processed meat; eat more fish and chicken or vegetarian substitutes
  • Get adequate calcium and vitamin D from food and sunlight
  • Don’t add salt to food and restrict high-salt foods. Athletes have more choice with salt intake.
  • Don’t eat trans fats in margarines and processed and fast foods
  • Don’t eat much saturated fat and cholesterol; eat most of mono- and polyunsaturated fat, especially omega-3
  • Limit alcohol consumption to one drink/day for women and two for men – and take time out
  • Fat – 20 to 35% – mostly mono and polyunsaturated
  • Protein – 15 to 25% – mostly fish, chicken, low-fat dairy, vegetable
  • Carbohydrates – 45 to 65% – mostly high-fibre, whole foods (athletes at the higher end of carb consumption can choose more sugars and refined carbs)
  • Fibre – aim for 0.5 grams/kilogram bodyweight – mix of soluble and insoluble fibre
  • Fluids – stay well hydrated, but not excessively, for your level of activity – tea and coffee count
  • Exercise - at least 5 days a week, various intensities, with some weights.

That last one is not exactly a dietary item. Even so, more and more food pyramids include physical activity as a fundamental. They’re absolutely correct. Think of it as the fourth macronutrient.

That’s a ‘diet,’ or at least an eating framework, that’s relatively easy to stick to, that everyone can enjoy and that will help prevent heart disease, diabetes, cancer and obesity — eaten within energy limits. Even people who cannot, or do not exercise, observing these general principles should provide a sound dietary environment for health promotion. Diabetics and celiacs can experiment with lower-carb if they feel it helps; people with known or suspected food allergies can move foods around to suit; and people trying to lose weight can experiment a little to find a pattern that works for them within healthy eating patterns.

The thing is, unless you’re overly obsessive compulsive, we tend to eat a varied diet from one day or week to the next depending on available choices, work, family, leisure commitments and so on. We need a healthy range of choices rather than some strict regimen in this modern world. By all means, do Ornish or Paleo or South Beach diets if you like — give it a go — then when you wake up, eat within the guidelines above, most of the time, and you won’t go wrong.


– Hu FB, Willett WC. Optimal diets for prevention of coronary heart disease. JAMA. 2002 Nov 27;288(20):2569-78. Review.
– Hu FB, Manson JE, Willett WC.  Types of dietary fat and risk of coronary heart disease: a critical review. J Am Coll Nutr. 2001 Feb;20(1):5-19. Review.
– Schaefer EJ, Gleason JA, Dansinger ML. The effects of low-fat, high-carbohydrate diets on plasma lipoproteins, weight loss, and heart disease risk reduction. Curr Atheroscler Rep. 2005 Nov;7(6):421-7. Review.
– Schoeller DA, Buchholz AC. Energetics of obesity and weight control: does diet composition matter? J Am Diet Assoc. 2005 May;105(5 Suppl 1):S24-8. Review.
– Kennedy ET, Bowman SA, Spence JT, Freedman M, King J. Popular diets: correlation to health, nutrition, and obesity. J Am Diet Assoc. 2001 Apr;101(4):411-20. Review.