This site is intended for health professionals only

At the heart of general practice since 1960

Is high salt intake really bad for you?

Dr Martin Dawes reviews the evidence on salt

and mortality, and asks whether your advice

can make people change their diets

Controversy surrounds the benefits of cutting salt consumption. I evaluated the available evidence by posing three questions:

 · Is there an association between high salt intake and mortality?

 · Does lowering salt intake reduce mortality (or morbidity or BP)?

 · Does advice to patients lead to a reduction in their salt intake?

Are high salt intake and mortality linked?

Surprisingly, the first question was hard to answer. How do you prove an association between an exposure and an outcome? Proving an association is fraught with difficulty as shown by the MMR debate. First you must find a way of measuring salt intake. Measuring intake by diary methods is open to considerable potential bias. Measuring urinary excretion is more reliable. Mean values of 24-hour sodium excretion are 216mmol for men and 162mmol for women. I restricted my search to papers that outlined the risk between salt excretion and mortality.

A recent study of 3,436 people in the Lancet showed salt was harmfu · 1. People recruited in most follow-up studies like this one are not the same age and have several important risk factors such as smoking, gender and cholesterol. A hazard ratio of mortality is produced to take account of these factors. This little bit of statistical magic, that needs to be taken with its own pinch of salt, roughly represents relative risk of dying if you belong to one group (high salt), compared with another (low salt), taking into account all other risk factors. The hazard ratio for overall mortality taking into account age, study year, smoking, serum total and HDL cholesterol, systolic blood pressure and body mass index for all-cause mortality was 1.22 (95 per cent confidence interval 1.02-1.47) for men and women together. For every 100mmol increase in urinary sodium excretion there was a 22 per cent additional risk of mortality entirely due to salt consumption after taking into account everything else. Wide confidence intervals indicated the risk increase may range from 2 to 47 per cent. The risk of men getting a fatal cardiovascular event is 55 per cent (95 per cent confidence interval 12 per cent to 113 per cent). This association was not seen in men of normal weight, suggesting only overweight men have the risk.

Does lowering salt intake reduce mortality?

Cynically, I wondered whether ­ if a tablet was on sale to reduce salt absorption ­ we would have several long-term mortality studies by now. There are only a couple of inconclusive systematic reviews of the effect of diet on mortality. There are many studies that show reducing salt reduces blood pressure. You cannot extrapolate from that and say definitely that lower salt intake reduces mortality but probably it does.

Does GPs' advice make people change their diets?

There is no evidence that advice on reducing salt intake is of benefit in the 'normal' population. However, it is probably wise to avoid excessive intake of salt2. For hypertensive patients, particularly those who are aged over 44 and overweight, I would recommend intake of dietary sodium be moderately restricted, to 3-7g of salt per day.

Research methodology

Source of Search term(s) Number of Date of

evidence trials found data in trial

Cochrane Sodium and diet One review in 1994

Library and mortality database of

Abstracts of

reviews of


Best Sodium and One systematic 1993

Evidence 5 diet and mortality review

Does dietary Three systematic 2001

advice to reduce reviews and one

sodium intake randomised

lead to a sustained controlled trial

fall in BP?

Clinical Does salt restriction One systematic 1997

Evidence reduce BP? review

Bandolier Can you reduce the One trial described3 1998

number of patients on

hypertensive treatment

by reducing salt?

evidence-based medicine

Rate this article 

Click to rate

  • 1 star out of 5
  • 2 stars out of 5
  • 3 stars out of 5
  • 4 stars out of 5
  • 5 stars out of 5

0 out of 5 stars

Have your say