A report from Red Star News published on November 5 and on Baidu's hot search list the next day carries a sweeping promise. Swap the family salt for a potassium-enriched substitute, it says, and the world could avoid close to 3 million deaths every year, based on a modelling study the report places in the journal Nature Health, covering 193 countries and projecting roughly 10 million fewer cases of cardiovascular and kidney disease. [1] The swap itself is backed by a randomized trial of more than 20,000 people that measured strokes and deaths, and by a WHO guideline on record suggesting it. The 3 million figure is a model projection reported secondhand, and it sits above the measured numbers the World Health Organization publishes.
WHO's fact sheet gives the measured side of the ledger. Adults worldwide averaged 4,278 mg of sodium a day in 2021, the equivalent of 11 g of salt, more than double the recommended ceiling of less than 2,000 mg. [2] Excess sodium was associated with an estimated 1.7 million deaths in 2023. [2] Those are counts of what has already happened. A projection of 3 million deaths prevented assumes the whole world changes how it salts its food, so the two numbers can sit side by side without either one being wrong. They answer different questions.
What the trials measured
The largest test comes from the Salt Substitute and Stroke Study, a randomized trial published in the New England Journal of Medicine in 2021. Nearly 21,000 adults across 600 rural Chinese villages used either regular salt or a blend of 75% sodium chloride and 25% potassium chloride for about 4.7 years. [3] In the market the report describes, regular salt is at least 97% sodium chloride and potassium-enriched substitutes contain 10% to 35% potassium chloride, so the trial blend sits inside that range. [1] Strokes occurred at 29.14 vs 33.65 per 1,000 person-years with the substitute and regular salts, a rate ratio of 0.86 (95% CI 0.77 to 0.96), about 14% fewer with the substitute. [3] Major cardiovascular events showed a rate ratio of 0.87, about 13% fewer, and death from any cause 0.88, about 12% fewer, both statistically significant. [3] Serious hyperkalemia, the main safety concern researchers checked for, was no more common in the substitute group (rate ratio 1.04, 95% CI 0.80 to 1.37). [3]
The people enrolled were not a random slice of humanity. Their average age was 65.4, 72.6% had already survived a stroke, and 88.4% had hypertension. [3] A 2025 subgroup analysis in JAMA Cardiology followed 15,249 of the stroke survivors for a median of 61.2 months. Recurrent stroke showed a rate ratio of 0.86, hemorrhagic stroke was 30% lower, and death 0.88, while systolic blood pressure fell by 2.05 mmHg (95% CI 1.08 to 3.03). [4] Hyperkalemia was again no more common (rate ratio 1.01). [4] Every strong event number in hand describes older adults at high cardiovascular risk in China. For everyone else the benefit has to be extrapolated from these populations.
Where the 3 million comes from
In January 2025 WHO published a guideline on lower-sodium salt substitutes, suggesting that regular salt be replaced with substitutes containing potassium. [5] [2] Its evidence review pooled 11 randomized trials with 3,885 participants. Urinary sodium fell by about 459 mg a day, roughly a tenth of the 4,278 mg global average, and potassium rose by about 450 mg. Blood potassium rose 0.12 mmol/L on average, hyperkalemia showed little difference, and cardiovascular death showed a rate ratio of 0.77 (95% CI 0.66 to 1.00) at moderate certainty. [6]
The 3 million figure travels a longer road. The modelling study the report cites, published in November 2026, arrives at close to 3 million deaths and about 10 million avoided cases of cardiovascular and kidney disease under a worldwide switch. [1] These projections estimate what could happen if everyone changed, where the trials counted what did happen when specific people did. WHO's own baseline for sodium-related deaths is 1.7 million a year, [2] and the report's summary does not lay out the model's assumptions in a form a reader can check. The direction of the viral claim fits the evidence. Its size remains a projection.
Who should be careful
Potassium is the reason a boundary list exists. Healthy kidneys clear excess potassium from the blood, and poor kidney function lets it build up. Certain blood pressure medicines also raise potassium, including ACE inhibitors, ARBs (angiotensin II receptor blockers) and potassium-sparing diuretics such as spironolactone, triamterene and amiloride. [7] Severe hyperkalemia can disturb the heart's rhythm. [7] MedlinePlus flags people with kidney problems, particularly those on dialysis, as a group that may need a potassium-restricted diet set by their clinician. [7]
The viral report's own caution line points at the same groups. It names severe kidney impairment, existing hyperkalemia and the people taking potassium-raising medicines or supplements, and it puts a doctor's assessment before any switch. [1] The same WHO review found only one study in children, with 92 participants aged 3 to 17, and drew no conclusion for that group. [6] The clean hyperkalemia results describe monitored trial populations, and people with failing kidneys sit outside them.
What this means for the salt in your kitchen
For most adults, the underlying advice survives the fact check. WHO suggests the swap. [2] [5] The main trial and its stroke-subgroup analysis show fewer strokes and fewer deaths among high-risk adults, with no significant rise in serious hyperkalemia. [3] [4] The 3 million lives a year stays a projection that outruns WHO's measured 1.7 million baseline, so it should be read as the model's output.
Switching salt does not change the limit on total sodium. The measured global average remains more than double WHO's recommended ceiling of less than 2,000 mg. [2] The boundary groups should still raise the change with their clinician first, since kidney disease, a history of high potassium or potassium-raising prescriptions all change the picture. [7] [1] For everyone else the evidence behind the swap is strong even while the headline number stays unproven.