The American Heart Association’s Strong Stance Against Science

Once again the American Heart Association has taken a strong stance against science. Of course, that’s not how the AHA phrases it. In its own words the AHA says it “strongly refutes the findings” of a “flawed study” which “you shouldn’t use… to inform yourself about how you’re going to eat.”

But in fact the AHA position is really a strong blow against science and the scientific process.

The current case revolves around a study published last Friday in the Lancet which suggested that salt restriction in the diet won’t benefit most people and may actually cause harm, though the study did also suggest that salt restriction might help the 11% of the population who have high blood pressure and consume a lot of salt.

The average American consumes about 3,400 mg per day of sodium. The AHA recommends that sodium levels be cut by more than half to 1,500 mg/day. Several other health organizations also recommend reduced salt intake, though their recommendations are less severe than the AHA’s (which is itself a good indication of the lack of scientific consensus).

First it’s important to admit that the AHA has a legitimate point: the Lancet study is by no means perfect and it certainly doesn’t “prove” the case against salt restriction. The study, I think everyone would agree, has all the usual limitations of an observational study. It can detect an association or link but it is unable to demonstrate cause and effect. The study is further weakened by its reliance on a morning urine collection to calculate sodium excretion and then estimate dietary sodium. The authors claim that this technique has been validated in previous studies, but it is important to acknowledge that this is an imperfect measure at best.

The key point here is that the authors of the Lancet study make no claim that their study is definitive. Instead they point out that the study was performed in the first place in response to earlier, less definitive studies hinting at possible harms associated with severe salt restriction.

The AHA is quick to find limitations in the Lancet study but it doesn’t acknowledge that its own evidence supporting salt restriction is itself profoundly imperfect. In fact, the entire case for salt restriction is based on the observation that salt raises blood pressure and that increases in blood pressure can lead to cardiovascular disease. Now it is absolutely true that high blood pressure can cause cardiovascular disease. This is about as well established a fact as can be found in the messy world of medicine. But it does not necessarily follow that any method to lower blood pressure will have a beneficial effect.

We now know that blood pressure lowering drugs, for instance, should not be chosen solely on the basis of their ability to lower blood pressure. Before gaining approval a new blood pressure drug needs to demonstrate not only that it can lower blood pressure but that it is also safe. The reason for this is that there have been several instances in which blood pressure drugs have failed because they were shown to be unsafe, despite their ability to lower blood pressure. The only way to prove these drugs are safe is through randomized controlled trials.

The main difference between drugs and dietary interventions is that we know much less about nutrition. It’s (relatively) easy to test a drug in a blinded and randomized controlled trial. But when it comes to nutrition it’s almost impossible to perform large, long-term, randomized, well-controlled, and well-designed trials of free-living human beings. The fact is no one really knows for sure the long term effects of large scale interventions involving an essential nutrient like sodium.

A former president of the AHA, Elliott Antman, described the AHA as a “a science-based organization dedicated to saving and improving lives.” “Confusion about something as dangerous as excess sodium is unacceptable. We owe it to the public to provide the most scientifically sound dietary advice.”

But although “confusion” about sodium may be “unacceptable” it may also be inevitable, at least for now. Despite what Antman and the AHA say, there is no widespread scientific consensus about salt. In its statements the AHA never acknowledges the lively ongoing debate about salt.

Just how divided is the field? When the New England Journal of Medicine published an earlier version of the Lancet study it was accompanied by an editorial that also expressed disagreement with the AHA’s low salt recommendations. Unless and until there is better evidence, “the results argue against reduction of dietary sodium as an isolated public health recommendation,” wrote the editorialist, Suzanne Oparil. Now here is a truly amazing fact: as I wrote at the time, Oparil is, herself, a former president of the American Heart Association, as well as the American Society of Hypertension. This doesn’t mean the AHA position is wrong, or that its opponents are right, but it does mean that it is patently absurd to pretend that the scientific question is settled.

It’s also worth noting that the senior author of the Lancet study, for instance, is Salim Yusuf, a top cardiologist who has been one of the most frequently cited scientists in the world. Again, number of citations or eminence in the field should not determine who is right and wrong. But it at least suggests that there is another side to the story that the AHA and Antman want us to ignore.

It’s hard to overstate the potential dangers of getting something like this wrong. As I’ve noted several times in the past, the American Heart Association should be especially careful in this regard, since it’s already been down this road before, with disastrous consequences. Back in the 1980s the AHA developed enormously influential guidelines on cholesterol and diet. These guidelines helped spark the campaign against dietary fat and had the catastrophic consequence of pushing people to consume more carbohydrates, including sugar, instead of fat and protein.  We will probably never know the full extent of the damage, but many have speculated that this may have contributed to the obesity and diabetes epidemics. Let’s make sure this doesn’t happen again with salt.

Let me be clear: I see no reason why the AHA shouldn’t make its case for low salt and shouldn’t issue recommendations. But they should first of all admit that this is the opinion of their own experts and that there are many other reputable scientists who disagree. The AHA can’t be judge and jury and simply declare themselves the winner in the court of science.

Science is about the accumulation of knowledge, and there are many things, like the fact that the earth revolves around the sun and not the other way around, that we now know that we didn’t know before. But the health effect of low salt isn’t the same by any means, and science is often about the process of accumulating that knowledge as much as it is about the knowledge itself. The Lancet paper is an excellent example of one small step in that accumulation. It may turn out to be mistaken, but it also may prove correct.

Previous Coverage of the Salt Debate:



  1. Very interesting and disappointing. AHA’s stubbornness on the low-fat thing is a black eye and this oddly defensive position about sodium reduction adds a second.

    Another major flaw in the AHA sodium recommendation, in my mind, is that is assumes hypertension is one thing and the that cause is almost always related to lifestyle (read: ‘the patient’s fault’), so the ‘cure’ is a lifestyle change. This is, frankly, pretty absurd. I was diagnosed with ‘essential’ hypertension at age 12 (well before I had a discernible ‘lifestyle’) and have been treated for it since I was 19 years old (so for 34 years). A diuretic has always been part of my drug regimen, along with watching sodium intake. Not surprisingly, my electrolytes, especially sodium, are often just a little off. The treatment is–you guessed it–supplementing with iodized table salt.

    So which is ultimately worse for cardiac health, three decades of chronic electrolyte imbalance or added dietary salt? Who knows? What I do know is that, at least in my case, unless the sodium intake is really excessive, it does not seem to impact my blood pressure, while the therapeutic diuretic (which actually does work to help control my BP) definitely impacts my serum sodium level. The current AHA guidelines with drastic sodium reduction would just exacerbate this situation, so they simply do not work well for me. Maybe this is due to the, as yet, undetermined underlying cause for my particular form of hypertension, which appears to be familial (shows up in childhood in otherwise healthy, active, non-obese, non-diabetic family members) and not lifestyle-related. Again, who knows? But I think this also illustrates the problem with ‘one size fits all’ recommendations for drastic sodium reduction like what the AHA is suggesting.

  2. I agree with pretty much everything you say…except when it comes to blood pressure medication. There are very few studies addressing the actual death rate in RCTs for blood pressure medication, and those for which there are data are pretty much failures. Reducing blood pressure does not mean a reduction in overall death rate.

    Of course, at least there are some RCTs for blood pressure medication. Meanwhile, there are basically no RCTs for salt reduction and mortality over the long term. My guess is that reducing salt to the levels the AHA wants us to reduce it to would cause more heart disease. But we can’t tell, because the studies haven’t been done. We (and the AHA) are flying blind.

  3. What’s sad is that Mente et al. seemingly insist on using these poor methods when they have been so explicitly disputed for many years, if not decades. These results get widespread publicity, while other studies using better methodology (repeated 24hr Na excretions) don’t get mentioned. Like this new one in JAMA: Sodium Excretion and the Risk of Cardiovascular Disease in Patients With Chronic Kidney Disease.


    An Original Investigation article from Tulane University School of Public Health and Tropical Medicine, New Orleans, LA, titled, Sodium Excretion and the Risk of Cardiovascular Disease (CVD) in Patients With Chronic Kidney Disease (CKD), just came out in JAMA, on May 24/31, 2016, exactly, 4 days after the publication of The Lancet article that is being heavily criticized now. The authors of this prospective cohort study claimed that among patients with CKD, higher urinary sodium excretion was associated with increased risk of CVD.

    These are the conclusions from 8 studies from 1999 to 2014 published in BMJ, JAMA, Circulation, Arch Intern Med and N Engl J Med and referenced in the May 24, 2016 JAMA article: “A positive association between sodium intake and blood pressure is well established. However, the association between sodium intake and clinical CVD remains less clear. While some studies reported a J- or U-shaped association between dietary sodium and CVD, others found a positive monotonic association between sodium intake and risk of CVD, coronary heart disease, congestive heart failure (CHF), and stroke. Methodologic limitations, including inconsistencies in dietary sodium measurement methods, could contribute to these conflicting findings. Blood pressure of patients with CKD is more sensitive to high sodium intake than persons with normal kidney function due to a diminished capacity to excrete sodium.”

    Are these studies confusing or contradictory to the average physician and the public? Does eminence trump evidence in medical research? What kind of public health recommendations come out these research? Is it about time we need to have a new Investigative Medical Journal, funded by the NIH, as a referee in clinical research and their publications to serve the interest of the society at large, other than just trusting the IRB, career researchers, peer review, journal editors and editorials, before having more Retraction Journals. A thought for the 21st century medical research!

  5. Harold Silverman says

    Well done, Larry. The vast majority of practitioners would agree with salt restriction as being beneficial for many reasons. In our society, where salt intake vastly exceeds the body’s needs for sodium, restriction would be helpful to most.

  6. Great discussion. Not sure we will ever find hard conclusions however focusing on one nutrient at a time when in real life there are so many factors at play at one time…

    Average US intake: 5000 mg/day (Range 3,000-20,000 mg)
    Minimal requirement: 500 mg/day
    Increase Na+ intake associated Increase BP and CVD, CVA, LVH, CHD, MI, Death, CRI, AC, Platelet Function, SNS, especially if salt sensitive
    Magnitude BP reduction directly proportional to decrease Na+ 150 mmol  100 mmol  50 mmol : 4-6 mmHg / 2-3 mmHg

    Complementary Health Practice Review 2000; 6 : 11-19;
    Circ 1998; 98: 613-17;
    Current Atherosclerosis Reports 2000; 2: 521-8;
    Hypertension 1989; 14: 570-7;
    JAMA 1996; 275: 1590-7;
    Am J Clin Nutri 1997; 65: 648-51,
    Arch IM 1997; 157: 657-67;
    JAMA 1998; 299: 839-46

    Improve HBP control in pharmacologic treated patients
    Reduce CVD, CHD, CHF, CVA in all subjects, especially the obese
    Reduce LVH, diastolic dysfunction, and vascular hypertrophy
    Reduce renal disease and proteinuria
    Improved effect with high K+, Mg++ and Ca++ intake (especially Na+ sensitive) (DASH I and II) (Montreal)
    Improved effect with restriction of refined carbohydrates

    Current Opin Nephrol 2011;20:37Exper Rev Cardiovasc Ther 2010;8:821
    Complementary Health Practice Review 2000; 6 : 11-19;
    Circ 1998; 98: 613-17;
    Current Atherosclerosis Reports 2000; 2: 521-8;
    Hypertension 1989; 14: 570-7;
    JAMA 1996; 275: 1590-7;
    Am J Clin Nutri 1997; 65: 648-51,
    Arch IM 1997; 157: 657-67;
    JAMA 1998; 299: 839-46

    Excess NaCl intake impairs vasodilation and increases vasoconstriction which reduces blood flow and increases BP in both normal and hypertensive patients with or without salt sensitivity.
    Decrease eNOS and NO
    Increased ADMA
    Increased oxidative stress
    Imbalance of NO and Ang-II
    Endothelial cells act as vascular salt sensors

    Ref: J of Hypertens 2011;29:415

    High sodium intake specifically abolishes the AT2-mediated vasodilation immediately via decreased level of AT2 receptor protein and after 30 days is associated with the abolition of endothelial vasodilation.

    Loss of the AT2-mediated vasodilation increases BP and risk of stroke and CVD.

    Dietary potassium lowers BP in normotensives and hypertensives in a dose related response.
    600 mg K reduces SBP 1.0 mm and DBP .52 mm Hg
    Response depends on race (B>W), sodium, magnesium and calcium intake. Higher sodium intake results in more BP reduction with potassium.
    4.7 gms (120mmol) of potassium lowers BP 8.0/4.1 mm Hg with reduction in CVA 15% and MI 11%
    Lowers risk of CVA, CHD, MI, CHF, LVH , CVD, CRI, DM, arrhythmias.
    Reduction in CVA is both BP dependent and non BP dependent.

    Ref: J Clin Hypertension 2008;10:3-11 Archives Int Med 2010;170:1501 J Am Society of Hypertension 2010;4:79NEJM 2007;356:1966. Current Hyperten Reports 2011;13:309Arch Int Med 2010;170:1745

    Reduce sodium intake to 1500-2000 mg per day.
    Increase potassium intake to 5 grams per day.
    This gives a K/Na ratio of 2.5-3.3/1.
    Each 1000 mg increase in Na intake per day increases all cause mortality 20%.
    Each 1000 mg increase in K intake per day reduces all cause mortality 20%.
    Highest quartile of Na/K ratio increased CVD and total mortality by 46% compared to the lowest quartile.

    Ref: Archives Int Med 2011;171:1183

    In summary, looking at a single mineral in a bubble will never make sense or show much as that is not the way they are in the body… a soup of magnesium,K+ and neuroendocrine peptides and steroid hormones all interacting at the same time. We must look to nutrition and lifestyle as a whole package to lower inflammation in the vascular space with it’s consequent endothelial dysfunction and subsequent loss of Nitric Oxide mediated vasodilation.

  7. The salt recommendation does not seem like good evidence-based medicine. Too much salt seriously tastes very bad & I hate it. Any time I am stuck eating prepared food these people seem to think that I have a severe salt deprivation that they must medicate with at least a gram of salt, sometimes 2-3 grams of salt. This is gross.

    Am I surprised? No. Because the god of low-fat has been worshipped for years. God forbid you could be stuck in a hospital and be killed by these low fat foods. Let me fast it out until real food is available.

    My primary pet-peeve against anti-science AMA is that stupid article they ran 30 years ago stating that you can screen people for diabetes based on if they smell nice. Three decades of harrassment later… You skunky stench is NOT NORMAL. I have always had normal blood sugar. What do JAMA people eat? Maybe they should upgrade their diets to cat food. Maybe practice medicine instead of divination. Yes, it is true that diabetics develop a sweeter odor, but others just naturally smell good.

    It’s not just me: a young father was being repeatedly tested for diabetes and was freaking out about how to provide for his family given this death sentence. (No one outright calls it a death sentence but seriously most people would rather have an agressive cancer which is cheaper to treat). The guy was too upset to listen to me explain that EVERYONE in his family smells nice. It is a combination if genes and diet.

    I realize some if this is off-topic but I held my tongue for 30 years about the trouble AMA’s journal has personally caused. It would be nice if all forms of divination and idiocy would be removed from the practice of medicine. But for some, what would be left?

  8. M. Babcock says

    This is a quote from “Good Calories, Bad Calories” by Gary Taubes from page 146.

    “Systematic reviews of the evidence, whether published by those who believe that salt is responsible for hypertension or by those who don’t, have inevitably concluded that significant reductions in salt consumption – cutting our overage salt intake in half, for instance, which is difficult to accomplish in the real world – will drop blood pressure by perhaps 4 to 5 mm Hg in hypertensives and 2mm Hg in the rest of us. If we have hypertension, however, even if just stage 1, which is the less severe form of the condition, it means our systolic blood pressure is already elevated at least 20 mm Hg over what’s considered healthy. If we have stage 2 hypertension, our blood pressure is elevated by at least 40 mm Hg over healthy levels. So cutting our salt intake in half and decreasing our systolic blood pressure by 4 to 5 mm Hg makes little difference.”

    Taubes cites these references: “See, for instance, Graudel, N. A., A. M. Galloe, and P. Garred. 1998. “Effects of Sodium Restriction on Blood Pressure, Renin, Aldosterone, Catecholamines, Cholesterols, and Triglyceride: A Meta-Analysis. JAMA. May 6; 279(17):1383-91. He, F. J.5 and G. A. MacGregor. 2004. “Effect of Longer-Term Modest Salt Reduction on Blood Pressure.” Cochrane Database of Systematic Reviews. No. 3: CD004937.”

Speak Your Mind