Home › Topics › Salt and Sodium

Salt and Sodium

GMJ News knowledge hub · last reviewed August 2026 · Georgian Medical Journal

SummaryStatisticsGlossaryGMJ newsFAQDocumentsOrganizationsResearch

Population sodium reduction is supported by one of the more coherent evidence chains in nutrition — a dose-dependent effect on blood pressure demonstrated in feeding trials such as DASH-Sodium, blood pressure as a causal driver of stroke and cardiovascular disease, and the SSaSS cluster-randomised trial in which replacing salt with a potassium-enriched substitute reduced stroke, cardiovascular events and death in nearly 21,000 participants (WHO). Against this sits a persistent counter-literature reporting a J-shaped curve in which very low sodium intake appears associated with harm, and this deserves an honest answer rather than dismissal: the studies reporting it rely predominantly on spot urine estimation of sodium intake, a method with known systematic error, and are vulnerable to reverse causation, since people who are already seriously ill eat less of everything. The practical point for most populations is that intakes are far above any disputed threshold, so the controversy has little bearing on what should actually be done.

Key messages

SETTLED: sodium reduction lowers blood pressure in a dose-dependent manner
Established in controlled feeding trials, most definitively DASH-Sodium, which demonstrated stepwise blood pressure reduction at successively lower sodium levels, with effects larger in people who were hypertensive, older or of African ancestry. Blood pressure is in turn causally related to stroke, cardiovascular disease and kidney disease through overwhelming trial evidence. The chain from sodium to blood pressure to events is therefore supported at each link, which is more than can be said for most nutritional exposures.
THE STRONGEST EVIDENCE: SSaSS demonstrated hard outcome benefit
The Salt Substitute and Stroke Study randomised nearly 21,000 people across 600 Chinese villages to a potassium-enriched salt substitute or regular salt, and found significant reductions in stroke, major cardiovascular events and death from any cause over approximately five years. This moves the field beyond surrogate endpoints: it is a hard-outcome cluster-randomised trial in a real population using a cheap, scalable intervention requiring no behaviour change beyond substituting one product for another. It is arguably the single most important nutrition trial of the last decade.
THE J-CURVE OBJECTION: real literature, identifiable methodological problems
A persistent body of observational work reports a J-shaped relationship in which both very high and very low sodium intake associate with increased mortality. This deserves a direct answer rather than dismissal. The critical problem is measurement: these studies overwhelmingly estimate 24-hour sodium intake from a single spot urine sample using formulae — principally the Kawasaki equation — which systematically overestimate at low intakes and underestimate at high, mathematically manufacturing a J-curve from a linear relationship. Studies using multiple 24-hour urine collections, the reference method, do not reproduce it.
REVERSE CAUSATION: the second explanation for the J-curve
People who are seriously unwell eat less of everything, including salt. Heart failure, cancer, frailty and advanced chronic disease all reduce food intake, and all predict death. A cohort observing that low sodium intake associates with mortality is therefore consistent with illness causing low intake rather than low intake causing illness. This is compounded because sodium excretion falls in heart failure and with certain medications. Together with spot-urine measurement error, reverse causation accounts for the J-curve without requiring any harmful effect of sodium reduction.
PRACTICALLY IRRELEVANT FOR MOST POPULATIONS: intakes are far above any disputed threshold
The disputed low-intake region of the J-curve lies below approximately 3g sodium per day. Average intakes in most countries are 3.5 to 5g or higher, roughly double the WHO recommendation of under 2g sodium — equivalent to 5g of salt. Whatever the truth about very low intakes, essentially no population is near that range, so the controversy has minimal bearing on what should be done. This is a case where a genuine scientific dispute has almost no practical consequence, and where its prominence serves mainly to delay uncontroversial action.
THE POLICY LEVER: reformulation, not individual restraint
Around 70-80% of dietary sodium in industrialised food systems comes from processed and prepared foods rather than the salt cellar, so telling individuals to add less salt targets a small fraction of exposure. The effective interventions are structural: mandatory or negotiated reformulation targets, which the UK programme demonstrated can reduce population intake measurably; front-of-pack labelling; procurement standards for schools, hospitals and public institutions; and potassium-enriched salt substitutes, which SSaSS showed deliver hard outcome benefit. Caution with substitutes is required in advanced kidney disease and with potassium-sparing drugs.

Key statistics

SSaSS
potassium salt substitute reduced stroke, cardiovascular events and death in ~21,000 participants
NEJM 2021
<2g sodium/day
WHO recommendation — equivalent to under 5g of salt; most populations consume roughly double
WHO
Spot urine error
Kawasaki-type formulae systematically manufacture a J-curve from a linear relationship
Hypertension/AJE
70-80%
of dietary sodium comes from processed and prepared foods, not the salt cellar
WHO/PAHO
Dose-dependent
stepwise blood pressure reduction with successive sodium reduction in DASH-Sodium
NEJM 2001
Below 3g/day
the disputed region of the J-curve — far below actual intake in essentially all populations
Epidemiology

Sodium reduction — where the disagreement actually lies

Source: Bars show strength of supporting evidence. The J-curve dispute concerns intake levels almost no population reaches.

Glossary of key terms

Sodium versus salt
UnitsA persistent source of confusion in labelling and guidance. Sodium chloride is approximately 40% sodium by mass, so 1g of sodium corresponds to about 2.5g of salt. WHO recommends less than 2g of sodium daily, which is less than 5g of salt — roughly one level teaspoon. Food labels variously report sodium or salt equivalent depending on jurisdiction, and the fivefold-plus difference between the guidance figures for each unit means that mixing them up produces serious misinterpretation in either direction. When comparing any two claims about sodium intake, the first step is always to establish which unit is being used, and a surprising proportion of apparent disagreement between studies and guidelines dissolves at this point.
The SSaSS trial
Clinical researchConducted in 600 villages across rural China, SSaSS randomised nearly 21,000 participants with a history of stroke or uncontrolled hypertension to receive either a salt substitute containing 75% sodium chloride and 25% potassium chloride, or regular salt, for a mean of just under five years. Results showed significant reductions in stroke, major adverse cardiovascular events and all-cause mortality, with no excess of serious hyperkalaemia in the studied population, which excluded people with severe kidney disease. The design is important: it required no dietary education, no behaviour change beyond using a different product, and the intervention costs very little. It provides the hard-outcome evidence that sodium reduction previously lacked, and it does so through a mechanism that combines lower sodium with higher potassium, both of which lower blood pressure.
Spot urine estimation and the manufactured J-curve
MethodologyThe technical heart of the controversy. Twenty-four-hour urine collection is the reference method for estimating sodium intake but is burdensome, so large cohorts commonly use a single spot urine sample with a prediction equation, most often the Kawasaki formula. These equations perform poorly at the extremes: they systematically overestimate intake in people with genuinely low intake and underestimate it in people with high intake, because they regress towards the population mean. The mathematical consequence is that a truly linear relationship between sodium and mortality is transformed into a J or U shape by the measurement method alone. Studies using multiple 24-hour collections, or careful within-person repeated measures, do not reproduce the J-curve, which is the strongest available argument that it is an artefact.
Potassium and the DASH pattern
NutritionThe sodium story is incomplete without potassium, since the two act in opposition on blood pressure and vascular function, and the sodium-to-potassium ratio predicts cardiovascular outcomes better than either alone in several analyses. Most populations consume too much sodium and too little potassium simultaneously, and the DASH dietary pattern — rich in fruit, vegetables, legumes, nuts and low-fat dairy — addresses both, producing blood pressure reductions comparable to a single antihypertensive drug. WHO issued a guideline recommending increased potassium intake alongside sodium reduction. The clinically important caution is that potassium-enriched substitutes and high-potassium diets require care in advanced chronic kidney disease, and in patients on ACE inhibitors, ARBs, potassium-sparing diuretics or mineralocorticoid receptor antagonists.
National reformulation programmes
PolicyThe UK salt reduction programme, beginning in the early 2000s, set progressively tightening voluntary sodium targets across food categories and combined them with public awareness and monitoring, achieving measurable reductions in population intake and in average blood pressure. Its subsequent weakening after responsibility shifted to a voluntary industry pledge system is itself instructive about the fragility of voluntary approaches. Other models include South Africa's mandatory maximum sodium limits by food category, one of the first legislated approaches worldwide, and Argentina and Portugal's legislated limits for bread. WHO has published global sodium benchmarks for food categories to support such programmes. The consistent lesson is that gradual, uniform reductions across a category are barely detectable to consumers because taste adaptation occurs, whereas unilateral reduction by one manufacturer loses market share.
Individual variation in salt sensitivity
PhysiologyBlood pressure response to sodium varies considerably between individuals. Salt sensitivity — a clinically meaningful blood pressure fall with sodium restriction — is more common with increasing age, in people of African ancestry, in those with existing hypertension, obesity, chronic kidney disease, diabetes or metabolic syndrome, and in primary aldosteronism. There is no practical clinical test in routine use, since formal assessment requires controlled sodium loading and depletion protocols. This variation is sometimes offered as an argument against population-wide reduction, but the argument does not follow: population approaches are justified by the aggregate benefit across a distribution of responsiveness, and the groups most likely to be salt-sensitive are also those at highest cardiovascular risk.

Latest GMJ coverage

Five Key Mechanisms Drive Hypertension Beyond Simple Blood Pressure Numbers
22/05/2026
Evidence Review: Five Health Behaviours Reshaping 2025 Clinical Practice
29/07/2026
Light Roast Retains Most Chlorogenic Acids, But Dark Roast Offers Different Antioxidants
22/07/2026
Eight Common Food Additives Linked to Hypertension and Cardiovascular Disease in Large Cohort Study
13/07/2026
Diabetes and Dementia: New Evidence Reveals Bidirectional Risk and Potential Neuroprotection
09/07/2026
Legume consumption reduces hypertension risk by 16%, major analysis reveals
07/07/2026

Frequently asked questions 12 Q&A — structured for Google featured snippets and AI discovery

Knowledge hub: guidelines, conventions and reports

Organizations working in migration and health

Related health topics

HypertensionResistant hypertension and dietary sodiumStroke preventionKidney disease and potassiumNutritionProcessed food as the sodium source

About this hub. Produced by the GMJ News Editorial Team as a public-good service. Every statistic is linked to its primary source. Documents are preserved in the GMJ Repository with full attribution. Georgian Medical Journal · Contact the editorial team
GMJ BriefsView all →
●From Emergency Response to Durable Policy: Three Actionable Lessons from the 2026 Migrant Health Review4 hours ago●400 Million Cases of Atopic Dermatitis: Nature Reviews Confirms Massive Global Burden4 hours ago●Announcement: The 20 Amino Acids: The Chemical Alphabet of Life and What Every Patient Should Know4 hours ago●What Healthcare Providers Need to Know About Sandwich Generation Caregivers2 hours ago●New Data: Unstable Connectivity Amplifies Loneliness More Than No Internet at All2 hours ago●Announcement: Magnesium deficiency linked to multiple chronic diseases: what clinicians need to know2 hours ago●Ultra-Processed Foods Account for Majority of Calories in Latin American Retail18 minutes ago●What Clinicians Must Know: Screening and Managing Immigration-Related Family Separation Trauma18 minutes ago●Announcement: Vitamin C May Reverse Skin Aging at the Genetic Level, New Research Suggests18 minutes ago●Three Clinical Advances in the UK's Updated Influenza Vaccine Framework2 hours ago