What can we really learn 40 years on from the Minnesota Coronary Experiment?
What a massive week in nutrition science! It was revealed that part of the results from the biggest study ever done to test the diet-heart hypothesis were hidden away and never published. Until now, published in the BMJ by a sort of data archaeologist Christopher Ramsden from the National Institute of Health.
The diet-heart hypothesis says that saturated fat raises cholesterol and thus causes heart disease.
This might not seem like an important question to you, but widespread acceptance of this idea has shaped the way we look and how we eat more than anything else in the history of health and medicine.
It’s still the cornerstone of conventional wisdom and dietary guidelines in virtually every country. It’s still the reason the conventional experts can’t take the biologically obvious leap to embracing low carb healthy fat diets for important treatments like weight loss and diabetes.
So, 40+ years on. Here it is in all its glory, published in the BMJ no less. What did they find?
“Available evidence from randomized controlled trials shows that replacement of saturated fat in the diet with linoleic acid effectively lowers serum cholesterol but does not support the hypothesis that this translates to a lower risk of death from coronary heart disease or all causes. Findings from the Minnesota Coronary Experiment add to growing evidence that incomplete publication has contributed to overestimation of the benefits of replacing saturated fat with vegetable oils rich in linoleic acid.”
That’s right, there was no support for their hypothesis in a randomised trial of more than 9500. Replacing saturated fat with Omega 6 fat did reduce cholesterol, but as we’d predict had no effect on heart disease. The one effect they did see was that older subjects who lowered their cholesterol had higher overall mortality.
Yes, cholesterol lowering in older adults caused more death. This is an effect already seen in some population studies.
So what should we take from all this?
It’s hard to say.
It’s hard because we will never know the exact motivations of Ancel Keys the co-principal investigator and infamous diet-heart hypothesis. The guy who shouted down sugar and had the world believe it was butter that was killing us via clogged arteries. If we knew why he decided not to publish these results, and why he waited 16 years to publish the only report on this experiment that did come out during his lifetime, it would be interesting.
If it turned out that the data just didn’t suit his view of the world, then that would be bordering on scientific fraud.
If he just got too busy because of his career success, and ran out of time in the end. Well…is that even plausible for the biggest diet experiment ever conducted? Its hard to imagine Keys being short of resources to help publish a study this important, especially for the time it was published.
How should it change our minds? Well, that depends on who you are really. I’ve never seen the evidence that says that eating saturated fat, in the context of whole food causes harm including more death or even heart disease. The “saturated fat raises LDL cholesterol, and LDL cholesterol is associated with more heart disease in some groups” argument isn’t a cause and effect argument. It’s just a hypothesis.
So I’m perfectly happy to look at this for what it is. Ancel Keys probably behaved unethically, the full extent of which is hard to say. There was never any evidence to demonise saturated fat to the extent that it should become public health enemy number one.
Will convention change?
There is now mounting evidence to advise people to eat less sugar and starch, especially if you are insulin resistant. This is creeping into policy.
But the status quo around fat? Why won’t this change, even with these new (but old) studies being revealed. Even with the evidence of publication bias like this?
I think we all see the word through our own dissonance.
Conventional experts will find problems with Key’s study. I’ve already seen many of these in the rapid response section of the BMJ (we have one there too). Such things as trial length, lack of knowing what they actually ate, low numbers in the autopsy results and so on. These are all issues to be sure, but not such that they’d somehow magically change the outcome to suit the diet-heart hypothesis. All of the diet-heart trials suffer from some of these issues, and usually from others as well (like telling the subjects cutting saturated fat to eat less refined food and more whole food, which at least didn’t happen here, making it a purer test of the hypothesis)
I’d add an issue – they replaced saturated fat with corn oil in high doses (about 19% of calories). Does anyone in modern nutrition science think this is a good idea? Yet they had to do this to lower cholesterol enough to make a difference, in terms of the diet-heart hypothesis.
Predictably, the experts have told us there is no reason to change current dietary advice which is low fat, and saturated fat is to be substituted by polyunsaturated fat.
Why? They cite the results of the meta-analyses of prospective cohorts (following large numbers of people living freely in the population with no actual experiment done). We agree that adding some essential polyunsaturated fats may be a good idea up to the point you get what you need (probably well under 10% of energy). But these studies don’t show that replacing saturated fat is the cause of any benefit, because you see the same small equal benefit of substituting in polyunsaturated fat for carbs.
These cohort studies give us the hypothesis that adding polyunsaturated fats in smallish amounts regardless of what they substitute might have some benefit. The experiments need to be done to find this out for certain.
They cite the Hooper Cochrane review of all the actual diet experiments where saturated fat lowering was part of the diet intervention. In this review there is no effect of these interventions on anything much at all, except one small effect of the interventions on combined coronary events.
OK, so digging deeper what is this effect? This small effect and how it was worked out is critical for the perpetuation of the modern diet-heart hypothesis and mainstream nutrition rejecting low carb eating.
You’ll be surprised about this little effect. When all experiments are added together we see that for the “Reduce saturated fat (treatment)” there were 21 791 people, of which 1774 (or 8.14%) had some sort of cardiovascular event over the study period (52 months). In the “usual diet (control)” there were 31509 people, or which 2603 (or 8.26%) had some sort of cardiovascular event over the same period.
To more directly compare what might happen here we can extrapolate. If we extrapolated and pretend we randomised 20 000 people – 10 000 to a reduced fat diet and 10 000 people their usual diet then we would expect to see 826 CVD events for the usual diet and a reduction to 814 for the reduced saturated fat. That’s a difference in 12 CVD events in 10 000 people over just about 4.5 years, because of eating less saturated fat (compared to doing nothing). And these 12 CVD events that make up the difference do not include either heart attacks or deaths; they include events like angina instead.
That’s a change in absolute risk of 0.12%.
Underwhelmed? Hell yes. So am I, especially because the trials that did show some positive effect like the Oslo diet heart study weren’t just about substituting out saturated fat. They were about more exercise, more vegetables, less processed food, eating fewer calories and often more. In fact, the statistical adjustment the authors make means that these smaller, multifaceted studies end up being given extra weighting because of the chance the bigger studies overwhelm the results.
In the end, it’s hard to see a path, other than a new generation coming through, for the diet-heart hypothesis to go away. It’s hard to see conventional wisdom changing anytime soon.
Remember – in all these studies, there has never been a diet-heart trial of replacing carbohydrate with polyunsaturated and/or monounsaturated fat. There hasn’t even been one that was a test of reducing sugar. All the government funding for long term trials to test the effect of diet on heart health over the years has gone to one team with one idea – to try to lower cholesterol by replacing saturated fat in the diet and see what happens.
Butter or corn oil?
I said before that we all end up with our own dissonance. I’m not naïve enough to think I am excluded from this. I obviously have a bias for low carb healthy fat eating and may see the world that way. That’s why it’s critical for you to take a little time to see the world of nutrition science for what it is and make up your own mind.