Practice Blog

To share practice related stories, create connections and engage readers in the amazing diversity of dietitian experiences.

Rethinking sodium: Reflections on research and implications for practice

A DC member from Cape Breton shares why she thinks we may need to change our thinking around sodium. 

STHS.jpgSusan Taylor is the manager of clinical nutrition at the Cape Breton Regional Health Care Complex. Part of her role is direct patient care on the hospital telemetry unit. She is the main preceptor at her hospital for dietetic interns from three Atlantic Canada universities and enjoys the ongoing learning that being a preceptor entails. Susan decided to delve into the literature on sodium following an intern’s therapeutic update focused on this topic. Susan can be contacted at


Over consumption of salt is a global health issue. Excessive sodium consumption is linked to a higher risk of cardiovascular disease (CVD) and stroke.1,2
Considering that Canadians consume an average of 3400mg of sodium/day,1 should we be concerned? United Kingdom and France also have high intakes, which vastly exceed the World Health Organization (WHO) recommended level of less than 2000mg daily.2


Median intakes of salt in the United States are similar to Canadian levels. It is postulated that these numbers may be even higher than the research shows as studies on salt consumption are generally based on self-reported data. People are likely to underestimate the true amount of food consumed.4 In Canada, the majority of sodium consumed (77%) comes from processed food.1,2 This may explain why people may not be aware of how much sodium they are actually consuming.
The recommendation to reduce dietary sodium has many merits.  There is little question that diets excessive in sodium (i.e. well above the upper limit of 2300mg/day), are associated with health risks. Most of us could benefit from eating less of it. This seems straightforward enough.
The controversy
There is, however, some controversy when it comes to dietary sodium intake:
  • How much sodium should people consume?
  • Should the amount of sodium we recommend be titrated in terms of energy requirements?
  • Is it possible we could actually be harming people by recommending too little sodium?
  • Widespread sodium reduction messages are aimed at the healthy public in an effort to prevent some chronic diseases. Should we apply the same recommendations to hospitalized patients?

Finally, the World Health Organization (WHO) focuses not only on sodium, but on sodium to potassium ratios.4 Similarly, the Dietary Approaches to Stop Hypertension (DASH) diet focuses not only on sodium reduction but also on adequate intake of potassium, magnesium, and dietary fibre.4, 5 In light of this, one wonders if we should be looking at dietary patterns, rather than focusing on specific nutrients.
How much sodium should we recommend?


Let’s start with the amount of sodium we should recommend. Well recognized, leading health organizations are not in agreement on this point (see table below). Each organization uses a rigorous expert consensus building process, yet there is still not enough conclusive data to support unified global recommendations about how much sodium the population should consume.
Examples of current recommendations:
World Health Organization < 2000mg/day
Adequate Intake in Canada 9-50 years old: 1500mg/day
Over 50 years old: 1300 mg/day
Upper Limit in Canada 2300mg
Recommended Daily Allowance (RDA) in France 2000-3200mg/day
American Heart Association < 1500mg/day
US Dietary Guidelines < 2300mg/day (1500mg/day for subgroups at risk)
Institute of Medicine (IOM) 2300mg/day (cautions that there is not enough evidence to support recommending 1500mg/day)
For example, the Canadian Diabetes Association 2013 Clinical Practice Guidelines note that there is concern that sodium intakes near the suggested adequate intake (AI) may be associated with increased mortality in people with both type 1 and type 2 diabetes.6,7,8  While association is not necessarily causation and may be explained by the fact that these patients are sick and may be eating little of all nutrients, it does make one consider possible risks to limiting sodium in this population. Given the high number of patients in hospitals with type 2 diabetes, I am concerned about not providing enough sodium to vulnerable patients.
In hospitalized patients with acute heart failure, there is a lack of evidence reporting benefits of sodium and fluid restrictions.9, 10 Aliti et al. assert that sodium and fluid restrictions in patients admitted with heart failure are unnecessary and widespread recommendations to restrict sodium in these patients have little supportive evidence.9 They conclude that providing liberal sodium and fluid intake in patients with acute heart failure is in the patient’s best interest. Weiss asserts that given the lack of evidence to support the use of low sodium diets in heart failure patients, patients should be discouraged from limiting sodium intake to less than 2300mg/day.11

This investigation into recommended intakes is especially relevant in terms of recent findings by the Canadian Malnutrition Task Force (CMTF). The CMTF have found that the prevalence of malnutrition in adult hospitalized patients in Canada is 45 percent.12 Malnutrition in hospitals contributes to increased length of stay and poorer patient outcomes. It is vital that patients enjoy the taste of foods provided in hospitals so that they will eat enough to meet their needs.
Personally, I have encountered patients who are not eating enough because the low sodium hospital food is not palatable to them. This is especially evident in patients who are used to consuming a high salt diet at home. Such patients need time to adjust to the taste of a diet that is lower in sodium. While low sodium foods may be one factor affecting malnutrition in hospitals, there are many other contributors as well, such as loss of appetite, tray accessibility, opening packaged items, and fasting for tests.
I posit that hospital menus that strive to meet the AI for sodium are unintentionally contributing to malnutrition in hospitals. I question the movement to provide only 1500mg of sodium per day in hospital menus. I also question the merit of ordering sodium restricted diets for every patient that comes in with heart failure or even acute coronary syndrome.
Many hospitalized patients experience poor appetite and it is a struggle to meet their energy and protein requirements. Categorically disallowing a salt package, especially knowing that a standard salt package (at my hospital) provides only 285mg of sodium, does not make sense to me. It could make the difference between patients eating or not. And, in many cases, allowing a salt packet is in the best interest of the patient. Instead of categorically disallowing salt packages, a dietitian should assess whether allowing them is acceptable on an individual basis (as able).


There is little question that diets excessive in sodium are unhealthy.  In general, people consume too much sodium and not enough potassium. However, given that in Canada the majority of the sodium is coming from processed foods, it seems logical that our message to our patients pertaining to sodium should focus on this issue. Less emphasis should be placed on the issue of avoiding salt in cooking and at the table because added salt accounts for a mere 11% of sodium consumed.1,2
The foods served in hospitals should be minimally processed and menus should not be so restrictive in salt that the food is bland and tasteless. Moreover, a menu that provides between 2000 and 2300mg of sodium may be preferable in terms of health benefits to one that contains 1500mg of sodium or less. Hospitalized patients should be assessed by a dietitian to see if providing more salt with their meals is warranted. This has the potential to help increase nutrient intakes, and ultimately, reduce length of stay.
Since doing this research, I have become much more comfortable allowing salt packages on patient trays, even if a patient is on a cardiac diet. Prior to this, I felt torn about providing it. I wanted patients to eat more and I wanted to please them but I worried I may be causing harm if I gave them a salt package.
Knowing that a salt package contains 285mg of sodium, and that the sodium content of the home-cooked meals we serve at our facility is very low, makes me feel confident that providing a salt package is in a patient’s best interest.  I have changed my patient education to focus on limiting processed foods, and feel better informed when speaking with staff, interns, and patients and their families about sodium.  
Editor’s note: The wide variety of sodium recommendations in the table above is bewildering. What do you think about sodium recommendations in your hospital or in general? Should patients be provided more salt with their meals? Please share your thoughts and comments below.
  1. Health Canada. Guidance for the food industry on reducing sodium in processed foods. Retrieved from
  2. Health Canada. Sodium reduction strategy for Canada. Retrieved from
  3. O'Donnell, M. J., Yusuf, S., Mente, A., Gao, P., Mann, J. F., Teo, K., ... & Probstfield, J. (2011). Urinary sodium and potassium excretion and risk of cardiovascular events. Jama, 306(20), 2229-2238.
  4. Drewnowski, A., Rehm, C. D., Maillot, M., Mendoza, A., & Monsivais, P. (2015). The feasibility of meeting the WHO guidelines for sodium and potassium: A cross-national comparison study. BMJ open, 5(3), e006625.
  5. World Health Organization. (2012). Guideline: Sodium intake for adults and children. Retrieved from
  6. Canadian Diabetes Association. (2013). Clinical practice guidelines. Retrieved from
  7. Thomas, M. C., Moran, J., Forsblom, C., Harjutsalo, V., Thorn, L., Ahola, A., ... & Groop, P. H. (2011). The association between dietary sodium intake, ESRD, and all-cause mortality in patients with type 1 diabetes. Diabetes Care, 34(4), 861-866.
  8. Ekinci, E. I., Clarke, S., Thomas, M. C., Moran, J. L., Cheong, K., MacIsaac, R. J., & Jerums, G. (2011). Dietary salt intake and mortality in patients with type 2 diabetes. Diabetes care, 34(3), 703-709. 
  9. Aliti, G. B., Rabelo, E. R., Clausell, N., Rohde, L. E., Biolo, A., & Beck-da-Silva, L. (2013). Aggressive fluid and sodium restriction in acute decompensated heart failure: A randomized clinical trial. JAMA Internal Medicine, 173(12), 1058-1064.
  10. Dietitians of Canada. Cardiovascular Disease-Heart Failure: Background. In Practice-based Evidence in Nutrition [PEN]. 2010. Access by subscription.
  11. Weiss, B. D. (2014). Sodium restriction in heart failure: How low should you go? American Family Physician, 89(7), 508.
  12. Allard, J. P., Keller, H., Jeejeebhoy, K. N., Laporte, M., Duerksen, D. R., Gramlich, L., ... & Teterina, A. (2015). Malnutrition at hospital admission—Contributors and effect on length of stay a prospective cohort study from the Canadian Malnutrition Task Force. Journal of Parenteral and Enteral Nutrition, 0148607114567902.
  1. The randomized trial Susan quoted did not quite address her question of providing some salt to these patients as it tested one hospital’s usual diet’s sodium compared to a more restrictive sodium level (800 mg/day). In contrast, Susan is suggesting liberalizing sodium slightly, by one salt package per meal or per day.

    The randomized trial by Paterna et al, identified in the editorial Susan located is of more importance for this topic since it is closer to what Susan is recommending. In a randomized trial Paterna et al compared 80 mg (1840 mg) with 120 mmol (2760 mg) sodium diets in compensated out-patients with congestive heart failure over 6 months (Ref below). The higher sodium group had a lower incidence of rehospitalization, which suggests that it might be safe to give a little more sodium, but this study cannot give you assurance by itself that giving the patients a salt package is safe.

    Also of importance for consideration to the Tolerable Upper Limit (TUL) is the Adequate Intake (AI) for sodium, sodium is an essential nutrient. Although the DRIs, AIs, and TUL are designed for healthy people, they provide some limited but useful guidance for people with health conditions. I suggest that a useful first step would be to compare the AI for sodium with the actual sodium content of various possible choices among the hospital diets. So far, Susan has assumed that the diet supplies approximately the AI of 1500 mg/day, but the actual amount delivered to patients likely ranges widely depending on whether the patient receives soup, bread, or cereals containing sodium.

    Perhaps those who have no salt containing soup or other high sodium foods could be given the salt package and still have an intake below the AI, or their intake might be greater than the AI of 1500 mg. Some homework is needed here.
    I strongly encourage Susan to discuss her ideas with her local health care team. Not only is their knowledge of this patient group important, the opinions of the Doctors and others are important, but if salt packages are noticed on the trays, incorrect assumptions would be likely if those who observe them are not part of the decision making process.

    A change in practice should be evaluated, ideally in a randomized trial, and the outcomes to consider should include perceived diet palatability, as well as cardiovascular and medical outcomes, as well as mortality. It is possible that people with acute heart failure have poor appetites due to their condition, and giving them salt might not change their food intake.

    I admire what Susan did, searching the literature, and writing her process for peer review. Ideally she will test (at least observationally and ideally to randomly allocated patients) to find out whether patients find the food more palatable with some salt added, and to evaluate the actual sodium content of the hospital diet, either simply using food composition tables, and/or by sending some samples to the lab.

    Paterna S, Gaspare P, Fasullo S, Sarullo FM, Di Pasquale P. Normal-sodium diet compared with low-sodium diet in compensated congestive heart failure: is sodium an old enemy or a new friend? Clin Sci (Lond). 2008 Feb;114(3):221-30. Abstract available from:
  2. Thanks for sharing this review and perspective. I totally agree that removing the salt package especially when meals are already quite modest in sodium and patients may be eating quite little is going to compromise their intake. Dietitians don't have time to screen and assess all hospitalized patients. In BC we have minimal diet technician support to help with menu
    marking in patients with poor intakes. Allowing a salt package makes sense.
  3. Susan, I really enjoyed your summary of the research and I find this information helpful! I am the one who gets the complaints since we removed salt packets from trays and shakers from the cafeteria!
  4. I loved the article. Some fantastic points.
    It's always great to investigate the studies that initiated the original guidelines. Perhaps it is the whole culture around fast/processed food that is the ultimate problem; sodium just being one of the components of this food. I agree, to be more lenient and liberal with some patients, but as RDs it's about assessing overall diet before making the recommendations.
  5. Hi Susan,

    Thank you for putting together a great summary on the differing sodium recommendations and potential implications.

    I think a holistic look at dietary patterns may be the key as you suggest rather than zeroing on specific nutrients.

    - Zannat
  6. Dear Susan,
    I am happy you took the initiative to speak some truths about salt/sodium. I totally agree. Several years ago I wrote an op/ed for the Medical Post with similar views on sodium. I can send it to you if you like.

    As dietitians I think we must be careful not to jump on the bandwagon without questioning what is practical and best for the patient, especially older patients, who often end up not eating at all due to "not always appropriate" restrictions.
  7. Fantastic article. I absolutely agree. The focus should be on the PATTERN not just specific nutrients. Food was created to enjoy not just be functional and it's all about balance. Great job!
  8. Thank-you for that article. I can't tell you how much time I spend liberalizing diet orders for people that are not eating well due to acute medical issues. There is a difference between recommendations in a community vs. needs in a hospital setting. While I am thankful for all the therapeutic/special diets available (e.g. renal, gluten free, vegetarian, halal etc.) I believe we spend too much money on food wastage as a result of trying to over-control a menu, when a liberal approach is needed as a first line intervention. Good intentions.

Leave a Reply

 Security code

Want to subscribe?

To subscribe to the blog with your DC account, please log in here


If you do not have an DC account, click Subscribe


Recent Posts


Tweet Tweet