Maybe the diet you are on right now is actually a pretty good one

By James McCormack

A version of this commentary appeared in the Montreal Gazette, the Vancouver Province and the Winnipeg Free Press

Maybe the diet you are on right now is actually a pretty good one

The world has seen a plethora of “experts” providing nutritional advice that sounds definitive and evidence-based.  Many of us have lived through all the recommendations: low fat then high fat; salt is a problem, then salt is no problem; eggs are good, then they are bad; butter is very bad, margarine is good, then butter is good again; high carbs, then no carbs — and so on.

This befuddlement has led both health care professionals and members of the general public to make recommendations — or even changes in our own diets — that, from afar, resemble a great cosmic yo-yo. With so much wayward nutritional advice, the medical profession has come to look indecisive and sometimes downright silly.

So here we go again.

A few weeks ago, a large (18 countries, five continents, 135,000 people) and long (7.4 years) cohort study on nutrition was published in the Lancet. The resulting headlines were full of hyperbole: “low fat diets could kill you” and “huge diet study shows carbs not fats are the problem.”

But when it comes to interpreting nutritional evidence, you shouldn’t just read the headlines because the devil is always in the details.

For starters, a cohort study like this cannot determine cause and effect but only give a suggestion as to what might happen when populations consume varying amounts of macronutrients — carbohydrates, fats and protein.  The people studied ingested a broad range of macronutrients (anywhere from 45 to 75 per cent of calories from carbohydrates, 10 to 20 per cent from protein and 10 to 35 per cent from fat). The investigators then looked at the association between the percentage of macronutrient intake and major cardiovascular disease and overall death.

What they found is that despite broad macronutrient ranges, there was no association between the percentage of macronutrient ingested over 7.4 years and the chance of a person developing cardiovascular disease — a major cause of overall disease and death.

An association was seen for overall death. However, even then, an increase in death was only associated with those people who ingested carbs at the highest percentage (around 75 per cent) of the ranges studied, or those who ingested protein or fats at the lowest percentage (around 10 per cent) of the ranges studied.

Importantly, the increase in the risk of death was only around 1-2 per cent higher for people at these “extreme” ranges — so even for the outliers, 98 to 99 per cent weren’t impacted.  In other words, this study seems to suggest that the macronutrient composition of a diet isn’t a big determinant of what makes a diet healthy or not.

When one looks world wide, macronutrient intake on average consists of carbohydrates at 63 per cent of calories; proteins at 11 per cent and fats at 26 per cent.  In developed countries, it is carbohydrates at 53 per cent; proteins at 12 per cent and fats at 34 per cent. So if this Lancet study is correct, the vast majority of us are eating a “healthy” mix of macronutrients.

Now let’s put this cohort study in context alongside randomised controlled trials of different diets — the highest form of evidence. Many might be surprised to learn only three large trials looking at important clinical outcomes have ever been done in nutrition: the 1994 Lyon Diet heart study (primarily men with cardiovascular disease), the 2013 PREDIMED (men and women without cardiovascular disease) and the 2006 Women’s Health Initiative (women without cardiovascular disease).

The first two trials, studied versions of a Mediterranean-type diet and found that fatal plus non-fatal cardiovascular disease was reduced by around 8 per cent over two years and around 1 per cent over four years, respectively. The WHI found a lower fat diet had no impact on cardiovascular disease or any health outcome over eight years.

In other words, the best available evidence — despite the evidence being clearly limited — seems to support a Mediterranean-type diet, which has slightly lower carbs and higher fat than what was evaluated in the Lancet study.

Taking these studies all together suggests, overall, as long as a person doesn’t eat at the extremes of any macronutrient, they should be just fine.

So, all those people who proselytize low carbs, high carbs, low fat, high fat — there’s no strong evidence to favour one over the other.

There are two important caveats. One, there are clearly people who do not eat in a healthy manner, but by far their biggest issue with food is not so much the type but the amount ingested. And two, evidence around diets is also fairly clear when it comes to excessive intake of over-processed food and refined sugars. These consistently seem to be bad actors in the dietary screenplay.

The good news is there’s no yo-yo this time. Looking at the evidence, many of us are eating a reasonable diet when it comes to macronutrients.

 

James McCormack is an expert advisor with EvidenceNetwork.ca, Professor with the Faculty of Pharmaceutical Sciences at the University of British Columbia and co-host of the popular Best Science Medicine Podcast at therapeuticseducation.org.

This entry was posted in Commentaries, Health is More than Healthcare, Our Commentaries and tagged , , , , , , , , , , , , .

Comments are closed.

« Back to Commentaries

License to Republish: Our commentaries, Infographics and videos are provided under the terms of a CreativeCommons Attribution No-Derivatives license. This license allows for free redistribution, commercial and non-commercial, as long as it is passed along unchanged and in whole, with credit to the author and EvidenceNetwork.ca
EvidenceNetwork.ca supports the use of evidence when reporting on health and health policy in the mainstream media. Specific points of view represented here are the author’s and not those of EvidenceNetwork.ca. Let us know how we’re doing: evidencenetwork@gmail.com