Salt

What it is

Salt is sodium chloride, and nutrition studies usually measure sodium rather than salt; 1 g of sodium equals about 2.54 g of salt [1]. The World Health Organization recommends adults take in less than 2 g of sodium (about 5 g of salt) a day [2]. Low-sodium salt substitutes replace part of the sodium chloride with potassium chloride; in trials the proportion replaced has ranged from about 3% to 77% [3].

How it might work

High sodium intake and low potassium intake both contribute to higher blood pressure, and salt substitutes reduce sodium while adding potassium at the same time [3]. In the umbrella review of meta-analyses, lower sodium intake lowered blood pressure and reduced urinary sodium and the sodium-to-potassium ratio, while urinary potassium rose [4].

What the research says

Cutting salt: a 2020 meta-analysis of 133 randomised trials (12,197 participants) found that reducing sodium by a mean 130 mmol a day lowered systolic blood pressure by 4.26 mm Hg (95% CI 3.62 to 4.89) and diastolic by 2.07 mm Hg (1.67 to 2.48), with a dose-response pattern: each 50 mmol a day less urinary sodium was linked to 1.10/0.33 mm Hg lower pressure [2]. A 2013 Cochrane review of 34 trials (3,230 people) found a modest salt reduction (about 4.4 g a day for at least four weeks) lowered systolic pressure by 4.18 mm Hg and diastolic by 2.06 mm Hg; falls were larger in people with hypertension (-5.39/-2.82) than in those without (-2.42/-1.00) [5]. An 85-trial dose-response analysis found an approximately linear relationship between sodium intake (0.4 to 7.6 g a day) and blood pressure, in people with and without hypertension [6]. A 2025 umbrella review of 21 meta-analyses reported low sodium intake lowered systolic pressure by 3.39 mm Hg and diastolic by 1.54 mm Hg, and that each 1 g a day increase in sodium was linked to 0.60 mm Hg higher systolic pressure, 4% higher cardiovascular risk and 6% higher stroke risk [4]. A network meta-analysis of 120 trials ranked salt restriction and low-sodium, high-potassium salt among the non-drug approaches that lowered blood pressure more than usual care [7].

Salt substitutes: a Cochrane review of 26 randomised trials (34,961 adults) found low-sodium salt substitutes lowered systolic pressure by 4.76 mm Hg and diastolic by 2.43 mm Hg versus regular salt, and probably slightly reduced non-fatal strokes, non-fatal acute coronary events and cardiovascular deaths [3]. A meta-analysis of 21 trials (31,949 people) found similar blood-pressure falls and lower total mortality (RR 0.89), cardiovascular mortality (RR 0.87) and cardiovascular events (RR 0.89) [8]. In DECIDE-Salt, a two-year trial in 48 Chinese care homes (1,612 residents), substitutes lowered systolic pressure by 7.1 mm Hg and reduced cardiovascular events (HR 0.60) with no effect on total mortality [9]; among residents with normal blood pressure, substitutes halved the incidence of hypertension without increasing low-blood-pressure episodes [10].

Hard outcomes: in the 20,995-person Salt Substitute and Stroke Study (SSaSS), strokes fell 14% (rate ratio 0.86), major cardiovascular events 13% and deaths 12% over a mean 4.74 years, with no significant increase in serious hyperkalaemia events [11]; in the prespecified subgroup with a previous stroke, recurrent strokes and deaths also fell 14% and 12% [12]. A 2024 meta-analysis of 16 trials of at least six months rated the mortality reductions low certainty (all-cause mortality RR 0.88; cardiovascular mortality RR 0.83) and the effect on major cardiovascular events very low certainty [13]. For salt-reduction advice without substitutes, a 2014 Cochrane review of eight trials found insufficient power to confirm clinically important effects on deaths or cardiovascular events, and its pooled cardiovascular event result was driven by one retirement-home trial [14].

Doses used in studies

The 133-trial meta-analysis pooled sodium reductions averaging 130 mmol a day measured in urine, with trials running from under a week to more than six months [2]. The Cochrane review's modest reduction was about 4.4 g of salt a day (75 mmol sodium) over at least four weeks [5]. The main salt-substitute trials used 75% sodium chloride with 25% potassium chloride (SSaSS, four to five years) [11] and 62.5% sodium chloride with 25% potassium chloride plus 12.5% flavourings (DECIDE-Salt, two years) [10][9].

Safety and interactions

Side effects: salt substitutes raise blood potassium slightly - by about 0.12 mmol/L across six trials - and high blood potassium is the main concern [3]. In the largest trials this did not translate into more clinical events: serious hyperkalaemia events in SSaSS (rate ratio 1.04, 95% CI 0.80 to 1.37) [11] and hyperkalaemia in the recurrent-stroke analysis (rate ratio 1.01, 0.74 to 1.38) [12]. In people with chronic kidney disease, lower salt intake may lead to symptomatic low blood pressure (risk ratio 6.70, 95% CI 2.40 to 18.69) [15]; in older care-home residents, substitutes did not increase low-blood-pressure episodes [10].

Upper limits: WHO advises adults stay below 2 g of sodium (about 5 g of salt) a day [2]. The trials did not identify a toxic dose of salt substitutes [3].

Interactions: the main concern is with medicines that impair potassium excretion, which the salt-substitute trials excluded [3].

Pregnancy and breastfeeding: no studies have tested salt substitutes in pregnant women and there are no breastfeeding data; evidence in children is very uncertain [3].

Kidney disease: salt substitutes are not established as safe in chronic kidney disease - the trials excluded people at risk of high blood potassium [3] - and a modelling study estimated additional hyperkalaemia deaths if substitution were used widely in this group [16]. People with chronic kidney disease, or who take medicines that affect potassium levels, should speak to a doctor or pharmacist before switching to a salt substitute; this page is not medical advice [3][15].

Caveats and open questions

Most salt-substitute evidence comes from China or Taiwan in people at higher cardiovascular risk, and the 2024 meta-analysis rates generalisability to Western populations as very limited, with one large trial dominating the results [13]. The hard-outcome evidence for salt reduction by advice is much weaker than the blood-pressure evidence [14]. There are no salt-substitute trials in pregnant women or breastfeeding women, and evidence in children is very uncertain [3]. Ongoing trials listed below are measuring blood potassium in people using substitutes, including kidney transplant recipients.