Salt and blood pressure: what the evidence actually shows

Hardly any dietary recommendation is as old and as often repeated as this one: eat less salt and your blood pressure will fall. And it is true. But that does not yet answer the really interesting question.

Because “lowers blood pressure” says nothing about by how much. And this is exactly where guides diverge: some suggest the salt shaker can replace a tablet, others declare the connection a myth outright. Neither does justice to the evidence. There are robust figures, and they lie in between.

What 34 controlled studies found

The most reliable summary comes from a Cochrane review. It analysed 34 randomised trials with 3,230 participants who reduced their salt intake for at least four weeks. The reduction actually achieved averaged 4.4 grams of salt per day – measured not by questionnaires but by sodium excretion in 24-hour urine.

The result across all participants:

• Systolic (upper value): −4.18 mmHg (95% confidence interval −5.18 to −3.18)
• Diastolic (lower value): −2.06 mmHg (−2.67 to −1.45)

So around four mmHg. That is less than the phrase “salt drives blood pressure up” leads you to expect – and more than a single number might suggest at first glance. More on why below.

Why the effect is not the same for everyone

The review analysed groups separately, and the difference is clear:

Group	systolic	diastolic
People with high blood pressure	−5.39 mmHg (−6.62 to −4.15)	−2.82 mmHg (−3.54 to −2.11)
People with normal values	−2.42 mmHg (−3.56 to −1.29)	−1.00 mmHg (−1.85 to −0.15)

People who already have raised values benefit about twice as much. This is a pattern seen again and again with blood-pressure-lowering measures: the higher the starting value, the more room there is to come down.

There is also an individual component. The German Hypertension League (Deutsche Hochdruckliga) points out that about one third of the population is salt-sensitive and reacts particularly strongly to salt. Whether you belong to this third cannot be determined at the dinner table – but it explains why two people make the same change and see very different results.

The bigger the cut, the bigger the effect

The Cochrane analysis contains a second, often overlooked figure: a dose-response relationship. A reduction of 100 mmol of sodium per day – equivalent to about 6 grams of salt – was associated with a systolic fall of 5.8 mmHg (95% confidence interval 2.5 to 9.2), after adjusting for age, ethnic group and baseline blood pressure.

This is the real reason why some studies report much larger figures: they reduced much more radically. Strictly controlled feeding studies, in which participants are provided with their food, achieve reductions that practically nobody sustains in everyday life. Their results show what is physiologically possible – not what a normal household achieves.

So if you read a much higher figure somewhere, it is worth asking: reduced from how much to how much?

How much salt is actually eaten in Germany

The WHO recommends that adults consume less than 2,000 milligrams of sodium per day – equivalent to less than 5 grams of salt, roughly one level teaspoon.

What was measured in Germany is different. In the German Health Interview and Examination Survey for Adults (DEGS1), sodium intake was estimated from urinary excretion. The median was 3.4 grams of sodium daily for women and 4.0 grams for men – in both cases above the recommendations of national and international organisations.

For context: sodium and salt are not the same thing. One gram of sodium corresponds to around 2.5 grams of table salt. The measured values are therefore roughly double what the WHO specifies. Anyone who went from 10 to 5 grams of salt would be exactly in the range for which the Cochrane data show around 5 to 6 mmHg.

Why the salt shaker is not the most important lever

This is the most important practical point of the whole topic. According to the German Hypertension League, around 70 per cent of salt consumption comes from processed foods.

That means: if you use less salt when cooking but continue to eat bread, cold meats, cheese, ready-made sauces and frozen pizza every day, you are only moving the smaller part. The large amounts are where you cannot taste them – in Germany, bread is one of the most significant sources of salt by quantity, even though it does not taste salty.

If you really want to achieve the effect seen in the studies, you will get there less through the shaker than through what goes into your shopping basket. The German Hypertension League puts it cautiously: “You don't have to give up salt, just reduce the amount.”

What 4 to 5 mmHg are worth

That leaves the honest question: is this a lot or a little?

For an individual it is a noticeable but limited shift. If you are at 155/95 mmHg, salt reduction alone will not get you into the target range. If you are at 138/86 mmHg, four to five mmHg can make the difference between “high normal” and the range in which treatment is discussed. What the numbers mean in detail is explained in our blood pressure values table.

At population level the assessment is different: shifting the entire distribution down by four mmHg changes the number of cardiovascular events considerably. That is why salt reduction is in the guidelines – not because it works miracles for individuals.

And one more point of context: salt reduction is one of the lifestyle measures. It sits alongside weight, exercise and alcohol, not in place of prescribed treatment. Never change existing medication on your own, even if your readings look better after changing your diet.

The comparison with other everyday measures is favourable. The effect of regular tea consumption, at around 1.8 mmHg systolic, is significantly lower – see Tea and blood pressure: what the research really shows. For breathing exercises, too, the evidence is weaker than the popularity of the advice suggests.

How you would even notice a change

Four mmHg is smaller than the normal range of variation between two measurements. A single measurement before and one after the change will therefore tell you nothing – the difference disappears in the noise. Why this is so is explained in Why your blood pressure is different every time you measure.

If you want to know, you need a series: seven days each before and after the change, two measurements each morning and evening under the same conditions, all values recorded. Only comparing two averages over one week each is informative enough to make a shift of this size visible. The details are in Measuring blood pressure correctly. And how much time your values spend in the intended corridor is another question in its own right – see Blood pressure target range.

Also give the change time. The studies in the Cochrane analysis ran for at least four weeks; shorter periods were not included at all.

When you should seek medical advice

• If your self-measured values are repeatedly at or above 135/85 mmHg
• If you have changed your diet and your values remain unchanged and high over several weeks
• If you are already taking blood pressure medication and are thinking about changing your diet – dosages belong in a doctor's hands
• If you have kidney disease, heart failure or are taking diuretic medication: in these situations salt and fluid intake are part of the treatment and should not be changed on your own

In case of sudden severe headache, chest pain, speech or vision problems or one-sided paralysis: call the emergency services on 112 immediately (the European emergency number).