How much sodium a day is too much?

The Daily Value for sodium on a food label is 2,300 mg a day.

These are the reference amounts used on US food labels and in dietary guidance, not personal targets. What you need shifts with age, pregnancy, medication and health conditions. The % Daily Value shown on every food page here is calculated against the FDA Daily Value for sodium, which is 2300 mg a day.

What this means for you

Two thousand three hundred milligrams is the label figure, which is about a teaspoon of salt, and most people in most countries eat more than that. The part worth knowing is where it comes from: bread, processed meat, cheese, sauces, soups and ready meals carry the bulk of it, so the salt cellar is rarely where the difference gets made.

Established. Eating less sodium lowers blood pressure, and the size depends on where you start. Across 56 systematic reviews, lower sodium intake reduced systolic blood pressure by between 2 and 8.7 mmHg, with the largest effects in people who already have high blood pressure.

How much good it does depends on where your pressure is now

That range hides two different people. Cut sodium by two thirds and systolic pressure falls by about 5.7 mmHg in someone who already has hypertension. In someone whose pressure is normal, the same cut moves it by about 1.1 mmHg. Both figures come out of the same trials.

For the first person the effect keeps going the further they go. Across 85 trials covering intakes from 0.4 to 7.6 grams of sodium a day, blood pressure falls in an almost straight line as sodium falls, with no floor below which the benefit stops. For the second person, a hard cut buys roughly one millimetre of mercury, which is worth knowing before rebuilding how you cook.

What a smaller cut is worth

The biggest pool of trials on this puts a rate on it. Across 133 randomised trials in 12,197 people, with sodium counted from a 24-hour urine collection rather than from what people said they ate, an average cut of 130 mmol a day lowered systolic pressure by 4.26 mmHg and diastolic by 2.07. That is everyone in those trials together, with and without high blood pressure.

The figure worth carrying around is the one per step. Each 50 mmol less sodium in the urine took a further 1.10 mmHg off systolic pressure and 0.33 mmHg off diastolic. A partial cut buys a partial return, so there is no threshold you have to reach before anything happens at all. The return was larger in older people, in non-white populations and in anyone starting from a higher pressure.

The same review found something that changes how you judge your own attempt. In trials shorter than fifteen days, that 50 mmol step showed only 1.05 mmHg against 2.13 mmHg in the longer ones, less than half. Past fifteen days the length of the trial stopped mattering. Short trials understate this, so give a change more than two weeks before you decide it did nothing.

One more thing sits between a trial and a life. When sodium reduction was run through whole communities, workplaces, schools and clinics, pooled across 36 studies in 66,803 people, systolic pressure fell by 2.64 mmHg against usual care. That held in adults and did not hold in the five studies done in children. It was also bigger in people who already had high blood pressure, so the average hides who is getting the benefit.

Not for everyone. Cutting hard has a price, and the same trials measured it. On a low-sodium diet renin, aldosterone and noradrenaline all rise, and so do cholesterol and triglycerides. All of that appears in blood tests long before anyone feels anything, and what it adds up to over years is unknown. For someone with high blood pressure the trade still looks worth making. For someone whose pressure is already normal it belongs on the other side of the ledger.

Is there such a thing as too little? The honest answer is that nobody knows

Unsettled. Three careful analyses disagree about the bottom of the curve, and the disagreement is real rather than a matter of emphasis. One pooling of sixteen cohorts found the risk curve bends at both ends, with low sodium going with more cardiac deaths and high sodium with more stroke deaths. A separate pooling of a quarter of a million people went looking for that same bend in stroke risk and found none, with risk rising in a straight line. And people followed for a median of 24 years after their trials ended showed 12 per cent higher death rates for every extra gram a day, with no bend at the low end at all. Part of why they disagree is measurement. The sharp upward turn above about 2,400 mg a day shows up when sodium is measured in urine and does not show up when it is estimated from a food questionnaire. When the shape of a curve depends on how the intake was counted, the ends of that curve are not settled, and anyone telling you that eating too little salt will kill you is reading one of these three papers and ignoring the other two.

The intervention with the best evidence is a swap

Potassium-based salt substitutes did better in trials than advice to eat less salt. Across six randomised trials in 27,710 people, switching cut deaths from any cause by 12 per cent and cardiovascular deaths by 17 per cent. In absolute terms that is about 12 fewer deaths per 1,000 people.

Not for everyone. Two cautions go with that. The evidence comes almost entirely from one very large trial in China, in older people at higher cardiovascular risk, and the reviewers say plainly that it does not transfer cleanly to a Western diet. And if you have kidney disease, or take an ACE inhibitor, an angiotensin receptor blocker or a potassium-sparing diuretic, a potassium-based substitute is a question for your doctor rather than a quiet swap.

One practical note that has nothing to do with trials. Taste adapts over a few weeks, and most people who cut salt gradually stop noticing, then find their old food tastes oversalted.

One more result worth knowing, because it is usually reported the other way round. The DASH diet, which lowers sodium while raising potassium, calcium, magnesium and fibre, was reviewed by Cochrane in 2025 across five trials in 1,397 people. It lowers blood pressure, total cholesterol and triglycerides and raises HDL. Whether it prevents heart attacks and strokes remains inconclusive, because the trials were too small and too short to count those events at all.

Salt sensitivity changes with age, and so does the value of this advice

The usual framing is that some people are salt sensitive and most are not, and that you would somehow know which you were. Across 160 studies it turns out to track with age and with the pressure you already have, and it moves over a lifetime.

Established. Under 40, the salt-sensitive pattern appears in 12.5 per cent of people. At 60 and over it appears in 56.5 per cent. Sorted by blood pressure rather than age it is 16.3 per cent in people whose systolic pressure is below 120, and 54.1 per cent in people at 130 or above. Averaged across everyone the relationship is modest and precisely measured: about 1.76 mm Hg of systolic pressure per 50 mmol of sodium a day, with an interval from 1.54 to 1.98.

The same cut is worth fifteen times more to some people than to others

Millimetres of mercury are hard to feel. Count instead how many people would have to cut their sodium to prevent one cardiovascular event, and the answer depends completely on who you count.

Established. Among people aged 60 or over, with systolic pressure at 130 or above and a body mass index of 25 or more, it takes about 87 of them to prevent one event. Among people under 60, with systolic pressure below 130 and a body mass index under 25, it takes about 1,334. That is a fifteenfold gradient, from the same analysis, for the same piece of advice.

Two conclusions follow and they deliberately point opposite ways. If you are in the first group, cutting sodium is one of the highest-value things on offer to you, and it earns real effort. If you are in the second, it is a small good. It does not deserve anxiety, guilt, or a dinner you did not want to eat.

Not for everyone. Do not read that second sentence as permission to stop caring. Kidney disease, heart failure and several common medicines change the sums on their own, and the blood pressure you have at 35 is not the one you will have at 65. What these numbers argue against is fear, rather than against the advice itself.

Where most people actually sit

About two thirds of American adults are above the 2,300 mg ceiling, 77.8 per cent of men and 59.0 per cent of women, on a median intake of 2,884 mg a day. It is highest in the young, 74.8 per cent among 19 to 30 year olds, and falls to 58.8 per cent past 70. The ninety-fifth percentile is 6,248 mg, so the people at the top are eating nearly three times the ceiling.

Unsettled. On salt substitutes, which swap some sodium chloride for potassium chloride, the blood pressure result is solid and the survival result is thinner than it sounds. Thirty-four randomised trials across 15 countries were pooled. Substitutes probably cut deaths from any cause and from cardiovascular disease, 7 to 17 fewer deaths per 1,000 people, and lowered systolic pressure by about 4.4 to 4.6 mmHg. Remove one very large Chinese trial and the effect on deaths stops being statistically significant. Seventeen of the 34 trials were Chinese. Take the pressure result to the bank and treat the survival result as resting on one study.

Not for everyone. A salt substitute is a potassium product, which makes it a drug question rather than a shopping one. If your kidneys do not clear potassium well, or you take an ACE inhibitor, an ARB or a potassium-sparing diuretic, ask before buying rather than after.

Written from the evidence cards and the data on this page, then checked back against the sources. Bioma Learn has no human medical reviewer at this time and we say so rather than invent one. Information, not medical advice. How we verify.

Foods highest in sodium on this site

• Sunflower seed kernels from shell, dry roasted — 7690.2 mg per 1 cup (334% DV)
• Mackerel, salted — 3560 mg per 1 piece (5-1/2" x 1-1/2" x 1/2") (155% DV)
• Pasteurized process cheese, American, low fat — 2504.6 mg per 1 cup, diced (109% DV)
• Cheese spread, pasteurized process, American — 2275 mg per 1 cup, diced (99% DV)
• Bacon, meatless fortified — 2109.6 mg per 1 cup (92% DV)

All sodium sources ranked →

Full guide: Sodium: what lowering it actually does to blood pressure and to death rates