Strange Salt

A new JAMA study finds a strong correlation: the third of folks who eat the least salt die over three times as often as the third of folks who eat the most salt. Yet other studies almost as big find contrary effects. I find it quite disturbing that such big studies can show such different results; something is very wrong in big diet correlation study land. Details:

Among 3681 participants followed up for a median 7.9 years, [heart attack] deaths decreased across increasing tertiles of 24-hour sodium excretion, from 50 deaths in the low, 24 in the medium, and 10 in the high excretion group (P < .001). … In multivariable-adjusted analyses, this inverse association retained significance (P=.02): the [hazard ratio] in the low tertile was 1.56 (95% CI, 1.02-2.36; P=.04). Baseline sodium excretion predicted neither total mortality (P = .10) [though 118, 64, 37 total deaths for low, medium, high tertiles sure looks significant to RH]. … All hazard ratios were adjusted for study population, sex, and baseline variables: age, body mass index, systolic blood pressure, 24-hour urinary potassium excretion, antihypertensive drug treatment, smoking and drinking alcohol, diabetes, total cholesterol, and educational attainment. …

Our current observations on cardiovascular mortality are consistent with several other reports. The National Health and Nutrition Examination Surveys (NHANES) I and II demonstrated an inverse association of cardiovascular and total mortality with salt intake as assessed from dietary recall with a similar trend in NHANES III. Alderman and colleagues followed up for 3.5 years 2937 patients with mild to moderate hypertension. There was an inverse association between the incidence of myocardial infarction and 24-hour urinary sodium excretion at baseline for the total population and for men, but not women. …

At variance with our current findings, other prospective studies suggested that a high-salt intake may lead to a worse outcome. … Cook and colleagues analyzed the long-term results of dietary sodium restriction on cardiovascular outcomes by combining 10 to 15 years of follow-up of 744 and 2382 participants randomized in the Trials of Hypertension Prevention, phases 1 and 2. Net sodium reductions during the intervention period (from 18 to 48 months) were 44 mmol and 33 mmol per day, respectively. … With adjustments applied for trial, clinical site, race, sex, and age, the [hazard ratios] for intervention vs control were 0.80 (95% CI, 0.51-1.26; P = .34) for total mortality. … In a 19-year follow-up study of 3126 Fins, the multivariable- adjusted [hazard ratios] associated with a 100- mmol increase in 24-hour urinary sodium were 1.26 (95% CI, 1.06-1.50) for total mortality, 1.45 (95% CI, 1.14- 1.84) for CVD, and 1.51 (95% CI, 1.14- 2.00) for coronary heart disease. (more)

GD Star Rating
Tagged as:
Trackback URL:
  • Patrick L
    • nazgulnarsil

      great dialogue. I particularly liked the “Doctors are like plumbers, not scientists” bit.

  • My theory is that one day Woody Allen’s forecast in Sleeper will come true and we’ll find out a few hundred years from now that the healthiest thing in the world is tobacco.

    • Aron

      Dietary science is a travesty. It’s a travesty of a mockery of a sham of a mockery of a travesty of two mockeries of a sham.

      • DK

        In other words, it’s like economics.

      • Chris Gregory

        Dietary science is dominated by skinny, obsessive-compulsive middle-class women with eating disorders who want to share their pathological disgust of food with others. As someone who has ended up in the field for very different reasons, I can’t imagine an area of science more strongly dominated by a cargo cult-level scientism (except for maybe psychology…). It’s like being taught portraiture by the colour-blind; they seemingly have no idea how strongly their understanding is coloured by their class values and pathological tendencies.

  • Matthew W. Fuller

    I guess I’ll just make sure my salt level is as close to avg as possible.

  • Alrenous

    Explain to me how anti-salt science could ever be taken seriously?

    Salt is critical to basically all life everywhere ever – therefore, salt regulation is critical. But supposedly the human body can’t tell when its had too much.

    How is that supposed to work? Why is there such a noticeable ‘too salty’ reaction, and how does it fail to cut salt consumption? Why does sea water kill you if your body can’t tell it has had too much salt? How isn’t salt a psychoactive drug? Shouldn’t eating a chunk of salt saturate your brain’s sodium ion channels?

    Are these low-salt eaters getting salt cravings, and are they satisfying them? Why are potato chips so very tasty if you have a salt craving, compared to their usual, if your body can’t properly regulate salt?

    Etc? Etc?

    • Robert Koslover

      Seawater contains large concentrations of NaCl-type salt and also other salts/minerals that, when consumed in sufficient quantity, can kill you through causing dehydration.

      • Peter St. Onge

        Salt is dangerous IFF you are not drinking water. Drink plenty of water and osmolality is maintained.

        Unless you’re shipwrecked on a raft, there’s no need to watch salt intake.

    • jsalvatier

      The human body is extremely complicated and our diet is significantly different than it has been in the past so our intuitions about such issues are poorly calibrated. I agree it’s enough to make you say “huh, that’s weird” and look a little closer into the matter, but not much more than that.

    • tylerh

      You are conflating two different time scales, so your argument does not hold

      1. osmotic transport happens on the time scale of minutes, which is shorter than the reproductive cycle of almost any living thing. So evolution has finely tuned moment-to-moment salt transport.

      2. chronic disease happens on the time scale of decades, which is longer than almost any living thing. So there is no reason to believe that evolution will have adapted to the long term effects of too much salt. Quite the contrary:: evolution should favor traits that make an individual more like to procreate at the expense of post-procreational longevity.

      • rmv

        Re: post-procreational longevity

        Not necessarily. Kin/group selection would be a factor.

      • Peter St. Onge

        If we accept parental survival positively influences offspring fitness, which seems fairly obvious, then yes, evolution does tend to control for long-term effects. The question becomes how far along this evolutionary process is. In other words, how long have humans been consuming salt. Considering hunter-gatherer societies at moment of contact, they indeed consume large amounts of salt, even at great expense (c.f. inland salt prices).

      • Alrenous

        So, the only reason is that it may, possibly, not be well-balanced for the very long term?

        That feeling fit and healthy in the short term (re: diet) tends to lead to illness in the long term?

    • ezra abrams

      vitamin A is essential, yet to much is highly toxic (thats why you don’t eat polar bear liver, should it ever be offered to you)
      in other words, the human body has a limited capacity to adjust; to much oxygen is bad, to little is bad, etc etc

  • alice

    the third of folks who eat the least salt die over three times as often as the third of folks who eat the most salt.

    So eat less salt and die more often? There’s got to be a better way to put that.

    • Captain Oblivious

      Yes, there’s definitely something wrong with people who die three times as often as other people. I think death should be a once-in-a-lifetime sort of thing, not something you do regularly just because you don’t like salt.

  • rmv

    Nutrition “science” is about as scientific as macroeconomics.

    Gary Taubes’ “Good Calories, Bad Calories” is illuminating. And depressing.

  • The evidence against salt was never very convincing. I tried to track it down, and all I found were controlled studies that show that salt raises blood pressure by a few mmHg (low single digits), and observational studies that show that people who eat more salt have more problems (just like this study, but with the opposite result). That was never convincing, because you would have to assume that the only difference between french fries and salad was salt if you were going to blame the problems on salt. Now it’s even less convincing.

  • I have posted this many times before:

    We do not know anything about nutrition except that you need to get minimum amounts of certain things, like amino acids and vitamins. The rest is just guessing based on tradition.

    • Jeffrey Soreff

      It isn’t quite that bad… We know a little about excessive doses of some nutrients. Eating enough selenium to experience acute toxicity is generally a bad idea 🙂

  • Peter St. Onge

    I’ve recently come to the conclusion that dietary advice is about as sound as mainstream macro. And shares an uncanny Kuhnian process (Ancel Keys playing the part of JM Keynes). I’d strongly recommend Kurt Harris’ Panu/Archevore site. Harris recently bested (!) Megan McArdle in a comment skirmish, which led me to him. Interesting stuff.

  • Peter St. Onge

    … and I second “rmv” above on Taubes. Highly recommend, right up your alley.

  • otakucode

    It’s almost as if human health was a complex system dependent upon more variables than it is currently possible to reasonably handle. And it seems like if you just ignore the variables you aren’t clever enough to think of, they still have an effect! Pick up a book on complexity and chaos theory. And stop extrapolating findings from studies which only address a very specific population out to humanity in general. If the people you studied had all watched TV at some point, then you can only apply your findings to others who have watched TV, for example. You cannot claim to have found anything whatsoever with regards to non-TV watchers, regardless of how unlikely you believe any interaction to be. This is science. If you don’t like the rigor, get out of the lab.

  • Sister Y

    Big studies with strong, contrary effects – I doubt it’s peculiar to nutrition, but it would be interesting to see where this happened the most and least frequently.

  • Lord

    From this data, I would guess the difference is between short term and long term. In the short term those with the lowest sodium intake are the unhealthiest and trying to reduce their intake to stave off death. In the long term those with the highest intake become the unhealthiest first. The treatment works until it doesn’t.

    • tylerh

      ahh, just like the studies that show “tea-totalers” — those who have no alcohol — have higher death rates the “a drink-a-day” folks.

      That, too is selection bias of the same form: the no drink crowd includes the very sick.

  • schweig

    1) NaCl – type salt.
    “salt” IS NaCl. There are other salts, some of which would kill you a lot sooner, but there is no “NaCl” type salt. Our science teachers failed once again.
    2) What we know about nutrition: If you don’t get any, you die.
    3) People are enormously stupid about food. The misery it causes seems to make them very happy.

    • David C

      Sodium chloride is just one type of salt.

      By the way, it seems like a lot of people on this thread have dubbed themselves experts on dietetics based on reading a smattering of books or a blog or two by somebody they like.

  • I wonder about the AGW in light of this.

  • Interestingly, Gary Taubes wrote a skeptical article on the anti-salt jihad (“The Political Science of Salt”, unfortunately not available for free online) before he became famous for “What If It’s All A Big Fat Lie?” and his subsequent books.

    I spend a lot of time writing about nutrition…and the more time I spend, the more I’ve realized that the “mainstream” nutritional recommendations are complete nonsense. (One example.) I also second the recommendation for Dr. Harris’ Archevore.


    • rmv

      Third the archevore recommendation

      • Peter St. Onge

        third archevore

    • The Taubes article, political science of salt available here.

      • Jeffrey Soreff

        Many Thanks for the link!
        Gaah! It’s amazing how inconclusive observational epidemiology
        can be. I’m giving up on what to conclude about the effect of salt
        on hypertension till I see a 10,000 patient double-blind, randomized
        placebo-controlled prospective study 🙁

  • blastomere

    There are many things that can go wrong with these studies, and some important ones are exhibited in spades in this study.

    Recall “All hazard ratios were adjusted for study population, sex, and baseline variables: age, body mass index, systolic blood pressure, 24-hour urinary potassium excretion, antihypertensive drug treatment, smoking and drinking alcohol, diabetes, total cholesterol, and educational attainment.” First, there is no reliably objective way to do adjust for such factors, because to make these adjustments they’re relying on the reliability of other studies in these various areas, can introduce as well as reduce selection bias, etc. Second, if salt causes the problems it allegedly causes by any of these mechanisms, as is likely the case, these adjustments will mask it. So at most they can say “we’ve shown that salt doesn’t cause substantial mortality via the diseases we haven’t adjusted for.”

    And as another commenter mentioned, there’s massive selection bias here. People who already have health problems are more likely to be the ones on a low-salt diet. We really need studies that force people to eat the diets they want to study regardless of their health status.

    I can imagine these kinds of problems and others also often exist in the studies that found salt to cause health problems, but regardless they certainly exist in spades in this paper. *If* there are reliable studies showing salt causes health problems — a big if, to be sure — this study certainly does not debunk them.

  • Ilya Shpitser
    • That review reports that they used a single measurement of salt excretion to cover an eight year period. That is complete nonsense.

  • Pingback: Humpday Paleo Push Through 2011May11 « Paleo Parents()

  • Pingback: Salt Wars | John Goodman's Health Policy Blog |

  • Dave

    Health parameters take on the absolutist aspects of religious doctrine. This study shows that everyone (practically) eat enough to maintain their sodium level at a very steady level. Population based efforts to control salt intake show no success in terms of increases in longevity. At best you can reduce measurable parameters such as blood pressure. This then becomes an assumed surrogate for increased longevity.
    Observational studies of people who live on salt poor diets don’t get age related hypertension. This is the main evidence that salt restriction in healthy people is beneficial. Salt intake has been declared a national emergency. It is an incitement of American life, capitalism the food industry. This is based, as far as I can tell, upon pretty firm observation that salt /age related systolic hypertension does not occur in people who eat a salt poor diet. There is always a lot of selectivity to these studies. This study could be criticized too and will be. There are only relatively young white people involved and as usual the study is for a limited time frame with limited sampling intervals.
    However, internationally, who are the low salt consuming people? They can’t be the Japanese, the longest living people on average on earth. I have traveled around some and have found nowhere that people don’t eat salt in there food. Perhaps they are the same mythological people who eat all whole grain foods while the Asians eat white rice and the Mediterraneans eat white pasta white, rice and white potatoes.
    Of course there are problems with the Western life style, such as too much animal fat and sugar and lack of exercise. Still average longevity is increasing. Also, salt reduction is a valid part of the treatment of hypertension and some other conditions.

  • Pingback: Book review: The Status Syndrome, by Michael Marmot | Chris Beardsley's Garage Gym()

  • Pingback: Overcoming Bias : Forget Salt()

  • Pingback: Salt (NaCl): Is salt bad for you? - Quora()

  • Pingback: Does Salt Cause Heart Problems…Or Doesn’t It? — Colloids For Life Blog()

  • Pingback: Sick quitters? |