#375: Salt, Sodium & Health

In Podcasts by Danny Lennon2 Comments

Today's Topic in Focus: Sodium & Health [03:42]

In this episode Danny and Alan discuss the current evidence base related to sodium intake and health, most notably cardiovascular disease. The episode walks through understanding diet-disease relationships, the epidemiology of sodium and health outcomes, the reasons for conflicting conlusions, sodium measurements in research, intervention trials, and more. Importantly, the claim that the sodium-CVD risk relationship exhibits a "J-shaped curve" (i.e. risk is low at moderate intakes and higher at both low and high intakes) is dissected, with recommendations given on how to reconcile all the available evidence.

Studies Referenced:

  1. Engberink  et al., 2017 - Use of a Single Baseline Versus Multiyear 24-Hour Urine Collection for Estimation of Long-Term Sodium Intake and Associated Cardiovascular and Renal Risk
  2. Intersalt, 1988 - an international study of electrolyte excretion and blood pressure. Results for 24 hour urinary sodium and potassium excretion
  3. Cook et al., 2014 - Lower levels of sodium intake and reduced cardiovascular risk
  4. Graudal et al., 2014 - Compared with usual sodium intake, low- and excessive-sodium diets are associated with increased mortality: a meta-analysis
  5. Mente et al., 2016 - Associations of urinary sodium excretion with cardiovascular events in individuals with and without hypertension: a pooled analysis of data from four studies
"I Have a Question!"  [83:33]

This episode's questions from listeners:

  1. David asks: "How much sodium does it take to incease blood pressure? Does "natural salt", like Redmond brand salt, make a difference?
  2. Nick asks: "In the low carb/ketogenic diet community there is this notion (I believe it originates from Steven Phinney) that the kidneys excrete more sodium while in ketosis, thus increasing sodium requirements while on a ketogenic diet. This has led to some prominent members of that community suggesting that supplementing up to 10g of salt per day can be beneficial. Any comments?"
Quack Asylum [90:51]

Claims about the health-promoting effects of sodium from author of "The Salt Fix", James DiNicolantonio.

    Random Recommendations [99:09]

    Alan's Recommendation: The Cicero Trilogy - Robert Harris

    Danny's Recommendation:
    Zen and the Art of Motorcycle Maintenance - Robert M Pirsig

    Fan of the Podcast?

    If you regularly enjoy listening to Sigma Nutrition Radio and you'd like to take your support even further, then you can officially support the podcast by:

    1. Making a one-time donation
    2. Subscribing to Sigma Nutrition Premium

    Thank you for considering!


    1. Hi Alan,

      Thanks for the interesting podcast.

      Why would a single urine measurement (and the resulting reduced accuracy of sodium intake at the individual level) lead to a J-shaped curve? Or, how Andrew Mente asked: “Also, there’s a positive association with blood pressure. So why would measurement error result in a positive association with BP but turn into a U-shaped association with clinical events?”

      Thank you.

      1. Hi Ty,

        Thanks for your comment. There are a couple of ways to think about this. The first is that the effect of single measures appears to be to underestimate sodium intake, therefore overestimating the risk at “low” levels of intake. The second is that, even if there were questions over the exact reason, the reality is that no study using multiple measures has demonstrated the J-shaped curve, but consistently demonstrate a more linear relationship. Repeated measures are always preferable in nutritional epidemiology, so I’m not sure why some are happy to reconcile the discrepancy in the literature in favour of the less robust measure. Finally, the limited data that we do have from multiple measurement studies with mortality as an outcome also suggests a linear relationship with events/mortality. For all these reason, we find it difficult to reconcile the current literature other than in favour of the data from multiple measurements.

        Hope that helps clarify things, feel free to fire away with any further comments by return.



    Leave a Comment