Salt Substitutes and Blood Pressure: The SSaSS Trial in Context

In 2021 a trial in rural China tested a different idea about salt: leave it on the table but change what is in it. The Salt Substitute and Stroke Study (SSaSS) gave 20,995 high-risk adults a salt that was one quarter potassium chloride and followed them for almost five years. Strokes, major cardiovascular events and deaths all fell. Here is what it measured, what other trials add, why its population limits what a healthy reader can take from it, and who should not switch.

What SSaSS tested and found

The product was 75 percent sodium chloride and 25 percent potassium chloride by mass (Neal 2021). Used like ordinary salt, it delivers a quarter less sodium and adds potassium, pulling two blood-pressure levers at once without anyone changing a recipe.

SSaSS was an open-label, cluster-randomised trial across 600 villages. Participants had already had a stroke, or were 60 or older with high blood pressure: a high-risk group. Villages got the substitute or regular salt, and follow-up averaged 4.74 years (Neal 2021). The results, per 1,000 person-years:

• Stroke: 29.14 events with the substitute vs 33.65 with regular salt, a rate ratio of 0.86 (95% CI 0.77 to 0.96).
• Major cardiovascular events: 49.09 vs 56.29, rate ratio 0.87 (0.80 to 0.94).
• Death from any cause: 39.28 vs 44.61, rate ratio 0.88 (0.82 to 0.95).
• Serious adverse events attributed to hyperkalaemia: 3.35 vs 3.30, rate ratio 1.04 (0.80 to 1.37), not significant.

Relative reductions of 12 to 14 percent sound modest, but in absolute terms that is about 4.5 fewer strokes and 5 fewer deaths per 1,000 person-years, from a change that asks nothing of the cook. Open-label design matters less for stroke and death than for symptom scores, but it is a limitation.

A post-hoc analysis (Yu 2024) found lower rates with the substitute for acute coronary syndrome (rate ratio 0.80, 0.65 to 0.99), heart failure (0.88), arrhythmia (0.59, 0.35 to 0.98) and sudden death (0.94), none significant after adjustment for multiple comparisons. Arrhythmia is the feared consequence of excess potassium; here the direction was fewer events, though the authors say post-hoc analyses cannot settle it.

The smaller trials point the same way

Taiwan, 2006. Five kitchens in a veterans’ retirement home were randomised to potassium-enriched or regular salt, covering 1,981 elderly men for about 31 months. Cardiovascular deaths ran at 13.1 vs 20.5 per 1,000 person-years, an age-adjusted hazard ratio of 0.59 (0.37 to 0.95), and the substitute group lived an estimated 0.3 to 0.9 years longer (Chang 2006). Five clusters is few, so the uncertainty is wide.

Peru, 2020. A stepped-wedge trial in six villages replaced household salt with the same 75/25 blend for 2,376 people, a general village population rather than a high-risk group. The average blood pressure change was small, systolic −1.29 mmHg (−2.17 to −0.41) and diastolic −0.76 (−1.39 to −0.13), but among people without hypertension at the start, the substitute was associated with a 51 percent (29 to 66 percent) lower risk of developing it (Bernabe-Ortiz 2020). It is the closest thing to evidence in people not yet hypertensive, but it measured blood pressure, not events.

Chinese care homes, 2023. The DECIDE-Salt trial randomised 48 elderly care facilities (1,612 residents aged 55 and over, mostly men) to salt substitute or regular salt, and separately to restricted or usual salt supply, for two years. The substitute lowered systolic pressure by 7.1 mmHg (−10.5 to −3.8) and diastolic by 1.9 mmHg (−3.6 to −0.2), with fewer cardiovascular events (hazard ratio 0.60, 0.38 to 0.96) but no significant effect on total mortality (0.84, 0.63 to 1.13). Serum potassium rose and biochemical hyperkalaemia was more frequent, without adverse clinical outcomes. Restricting the salt supply changed nothing (Yuan 2023): substitution worked where “use less” failed.

What the meta-analyses add

A 2022 meta-analysis of 21 trials and 31,949 participants found the substitute lowered systolic pressure by 4.61 mmHg (−6.07 to −3.14) and diastolic by 1.61 mmHg (−2.42 to −0.79), with risk ratios of 0.89 (0.85 to 0.94) for total mortality, 0.87 (0.81 to 0.94) for cardiovascular mortality and 0.89 (0.85 to 0.94) for cardiovascular events. The blood pressure effect was consistent across regions and subgroups, which the authors take as a sign it generalises (Yin 2022).

A 2024 review restricted to trials of at least six months was more cautious. Across 16 trials, all-cause mortality (rate ratio 0.88, 0.82 to 0.93) and cardiovascular mortality (0.83, 0.73 to 0.95) were rated low certainty, major cardiovascular events (0.85, 0.71 to 1.00) very low certainty, and serious adverse events showed no clear difference (1.04, 0.87 to 1.25), again very low certainty. The reasoning: the evidence base “is dominated by a single, large RCT”, most trials come from China or Taiwan, and participants had higher-than-average cardiovascular risk, so generalisability to other populations is “very limited” (Greenwood 2024).

Why both halves of the swap matter

Sodium. A meta-analysis of 85 trials spanning sodium intakes from 0.4 to 7.6 g per day found that blood pressure tracked sodium intake in a roughly straight line, for systolic and diastolic readings alike, across the whole range, with no flattening at the low or high end. The slope was steeper in people with higher starting blood pressure and held regardless of trial length, medication use or sex (Filippini 2021).

Potassium. A meta-analysis from the same group, covering 32 randomised trials of potassium supplementation (30 to 140 mmol per day), found a U-shaped curve: blood pressure fell as potassium rose, the benefit weakened beyond a difference of about 30 mmol per day, and above roughly 80 mmol per day it rose again. The rise appeared in people on blood-pressure medication, not in untreated people, and the benefit was strongest in those with hypertension and high sodium intake; the authors caution that few trials tested the highest intakes (Filippini 2020). The salt-substitute trials swapped the product for salt already in use rather than adding potassium on top, and the U-shape is worth remembering if you also use electrolyte products.

Who the evidence applies to

Of the eight trials in the 2024 review reporting deaths or cardiovascular events, seven were run in China or Taiwan and seven enrolled older or above-average-risk people (Greenwood 2024); every trial above replaced the salt used in a household or institutional kitchen. Three things follow.

First, the effect sizes belong to that population. The sodium dose-response predicts a smaller change at normal blood pressure, and the Peru trial’s average of about 1 mmHg is probably closer to what a healthy adult would see.

Second, where much of a diet’s sodium arrives in packaged and restaurant food, a substitute at home reaches a smaller share of intake. No outcome trial has tested that setting, and the 2024 review names Western populations as the group with the least certain evidence.

Third, blood pressure is one risk lever among several, alongside aerobic fitness and ApoB, and a salt swap does not stand in for the others.

Who should not switch without a clinician

Potassium is cleared by the kidneys, and the reassurance from SSaSS comes with a caveat. As a 2024 review of clinical guidelines notes, most trials “took steps to exclude participants at elevated risk of hyperkalemia, primarily those with advanced kidney disease or using medications that elevate serum potassium”, and SSaSS’s pragmatic design meant “there was no information about the occurrence of biochemical hyperkalemia” (Xu 2024). The safety data cover people who resemble the participants and are largely absent for those in whom extra potassium carries the most risk.

Modelling of a nationwide switch in China estimated about 461,000 cardiovascular deaths averted each year against about 11,000 additional hyperkalaemia deaths in people with chronic kidney disease: a large net benefit at population level, even within the kidney-disease group, but a model rather than a measurement (Marklund 2020). Of the four kidney-disease guidelines that mention potassium-enriched salt, three advise against it in chronic kidney disease, and the review’s own draft proposal is that it be recommended in hypertension “unless they have advanced kidney disease, are using a potassium supplement, are using a potassium-sparing diuretic, or have another contraindication” (Xu 2024). UK guidance also names ACE inhibitors and angiotensin receptor blockers, a broader caution whose rationale the review calls unclear.

So the clinician line is not boilerplate here. Anyone with chronic kidney disease, anyone taking a medication that raises serum potassium (potassium-sparing diuretics, and per UK guidance ACE inhibitors or ARBs), anyone on a potassium supplement or told to limit potassium, and anyone pregnant or managing another condition should ask before switching. For everyone else, the trials used the substitute as a straight replacement for household salt, in cooking and at the table, never on top of it.

Bottom line

SSaSS is the strongest single piece of evidence in the salt debate: 20,995 high-risk adults, nearly five years, 14 percent fewer strokes, 13 percent fewer major cardiovascular events and 12 percent fewer deaths, with no significant rise in serious hyperkalaemia events. Smaller trials and two meta-analyses point the same way, disagreeing only on how far one trial generalises. For an older person with high blood pressure who cooks at home, the trial evidence applies most directly. For a healthy adult on a Western diet, the expected effect is smaller and untested on hard outcomes. For anyone with kidney disease or on potassium-raising medication, it is a question for a clinician, not a supermarket aisle.