Fact-checking the backlash against recent salt studies


As of September 2017, new Sweat Science columns are being published at www.outsideonline.com/sweatscience. Check out my bestselling new book on the science of endurance, ENDURE: Mind, Body, and the Curiously Elastic Limits of Human Performance, published in February 2018 with a foreword by Malcolm Gladwell.

- Alex Hutchinson (@sweatscience)


Look, I agree that the role of salt in food is complicated. It’s not that I think salt has no possible effect on health, and that people should just eat as much as they want. But I do think the reaction to recent studies questioning salt orthodoxy is ridiculous and closed-minded. I agree entirely with a recent statement from Yoni Freedhoff’s excellent Weighty Matters blog, in discussing a recent Scientific American article on salt:

So while I think healthy debate is in fact healthy, I would have thought that magazines like Scientific American, and many of the intelligent commentators on this and other blogs, would in fact do their due diligence to read and critically appraise studies, before getting on any particular bandwagon.

The thing is, I think SciAm did do its due diligence, and many of its critics didn’t. The most widely linked response to the recent salt studies comes from the Harvard School of Public Health, which posted a piece called “Flawed Science on Sodium from JAMA: Why you should take the latest sodium study with huge grain of salt.” It wastes no time in asserting that conclusions of the latest JAMA study (which I blogged about here) are “most certainly wrong.”

Why should we conclude that the JAMA study is wrong? Harvard doesn’t try to explain the results (which found that a measurement of sodium intake wasn’t linked to blood pressure, hypertension, heart disease in 3,681 healthy adults over a 7.9-year period). Instead, they offer some possible ways that random error could have crept into the results, such as:

  • the study was too small to support its conclusions, with just 3,681 subjects;
  • the study used 24-hour urine collection to assess sodium intake, which just provides a snapshot in time;
  • the study didn’t account for the fact that people who are tall and/or active eat more food (and thus salt) but have lower risk of heart disease.

Okay, fair enough. Getting good epidemiological data on salt consumption and health outcomes is very difficult, and this study certainly would have been better if it had a million people in it and kept them in boxes for 20 years to prevent any confounding factors. Presumably that’s what the salt-is-bad studies did, right? It certainly sounds that way, according the Harvard article:

Furthermore, the study’s findings are inconsistent with a multitude of other studies conducted over the past 25 years that show a clear and direct relationship between high salt intakes and high blood pressure, and in turn, cardiovascular disease risk. (4-10)

Conveniently, the (4-10) refers to links to these studies — the strongest evidence Harvard could marshal to prove that salt is dangerous. So what happens if we actually bother to read and critically appraise these excellent studies — perhaps using the same standards they’re applying to the JAMA study?

Uh-oh. This Intersalt study uses 24-hour urine excretion (“unreliable,” according to Harvard). This BMJ study only had 3,126 subjects, smaller than the JAMA study. This AIM study used 24-hour urine and only had 2,974 subjects — and not only that, it found no significant relationship between sodium levels and heart disease. (They tried to salvage the “right” answer by saying there was a “nonsignificant trend” — imagine if the JAMA study had been so brazen!) This NEJM study only had 412 participants, and based its primary conclusion on a comparison of a regular, high-salt diet with a low-salt version of the DASH diet, which “emphasizes fruits, vegetables, and low-fat dairy products, includes whole grains, poultry, fish, and nuts, contains only small amounts of red meat, sweets, and sugar-containing beverages, and contains decreased amounts of total and saturated fat and cholesterol.” Sounds like a fair comparison to me!

Okay, seriously. There’s no doubt that salt has an effect on blood pressure. That’s just basic chemistry. But does it have a clinically significant effect? The DASH study I mentioned above found that cutting salt intake by about 55% (good luck with that in the real world, and feel free to donate your taste buds to science, since you won’t be needing them) reduced systolic and diastolic blood pressure by 6.7 and 3.5 mmHg respectively. For comparison, to go from stage 1 hypertension to normal, you’d have to reduce systolic pressure by a minimum of 20 mmHg. So if eliminating more than half the salt in your diet is able to (barely) move the needle on blood pressure, isn’t it reasonable to question whether dramatic society-wide efforts to reduce salt consumption even in healthy people are rational and useful? And given these small effects, isn’t it plausible that in a real-world epidemiological study of healthy (non-hypertensive) people (like the JAMA study), sodium intake might have no bearing on subsequent health outcomes? Why would such a finding be “most certainly wrong”?

The point is that applying double standards to evaluate studies doesn’t serve science, and it doesn’t serve the public interest. This latest JAMA study appears to me to be no better and no worse than the studies used to justify the “war on salt,” so promptly dismissing it because of its conclusions (rather than its methodology) is lazy at best, and dishonest at worst.

Final note: I still find it interesting that Walter Willett (the key voice in the Harvard School of Public Health article dissected above) himself published findings showing that salt intake in the U.S. essentially hasn’t changed over the last 50 years, while hypertension has risen dramatically. I’m still not sure how he explains this, if salt is such a key driver of blood pressure.

6 Replies to “Fact-checking the backlash against recent salt studies”

  1. I think the salt debate is less about scientific truth than about how inconclusive scientific evidence should influence policy.

    As I gather from the Scientific American article, at the time of the adoption of the ‘reduce salt’ dietary guideline, there was some evidence of a link between salt intake and heart disease, but no conclusive proof. As these things go, presumed risks are often incorporated in policies, even if the science is inconclusive, on the basis of the ‘precautionary principle’. The very same principle that keeps the climate debate alive.

    The precautionary principle roughly states that when there is a risk of damage, lack of full scientific certainty should not be used as a reason for postponing cost-effective measures. Judging from the excerpts of the 1977 report that SciAm mentions, the precautionary principle played an important role in the salt guideline. One of the experts actually quotes the Canadian Minister of National Health of 1924:

    “Even such a simple question as whether one should severely limit his consumption of butter and eggs can be a subject of endless scientific debate. Faced with conflicting scientific opinions of this kind, it would be easy for health educators to sit on their hands; it certainly makes it easy for those who abuse their health to find a real “scientific” excuse. But many of our health problems are sufficiently pressing that action has to be taken even if all scientific evidence is not in.”

    According to another expert “The question to be asked therefore is not why should we change our diet, but why not? What are the risks associated with eating less meat, less fat, less saturated fat, less cholesterol, less sugar, less salt, and more fruits, vegetables, unsaturated fat, and cereal products-especially whole grain cereals. There are none that can be identified and important benefits can be expected.”


    What hasn’t changed since then is te inconclusiveness of the science. What has changed is the appraisal that bringing about a dietary change is easy and costless. That calls into question the ‘better safe than sorry’ assessment of 1977. As in the climate debate, some are apt to defend science based policies with policy based science, claiming that the debate on costs and benefits is closed, because the science is settled. The Harvard response has that written all over it.

    What it also shows is, that a choice based on the precautionary principle can, but need not be wrong. Though the choice on salt can be called into question, the other recommendations seem mostly sound, even though they were probably based on equally unfinished science.

Comments are closed.