Too much salt?

What a meaty subject to start with! This was the headline…

… and it was published this week on the Daily Mail (linked here). It was quite technical, even for the Daily Mail. It started off by referencing a new study from Professor Simon Capewell, who suggests that nobody is doing enough to lower the salt in our diets, and as a result, far more people are at risk of heart disease than they should be.

The original study is linked here. Simon Capewell was one of twelve authors on the paper, and not a lead author. FYI, the first and last authors on papers tend to be the big cheeses.

A few years back (2011), the UK government set up an initiative working with commercial food manufacturers to lower salt in foods. Capewell and chums investigated whether this initiative had actually worked to lower salt consumption in the general UK population. They also looked to see whether there had been any change in the amount of heart disease.

Now, how they went about their study is very interesting. They used computer modelling. To make a computer model, you first have to input lots of known data (from surveys etc) and from this you can create trends into the future – use the past to predict what will happen next. They used their model to estimate how salt intakes would have looked if the government initiative had not taken place, and compared it with a model of salt intake with the government initiative in place. Their findings suggest that before the government initiative was introduced, average daily intake of salt was reducing across the population at between 0.12 and 0.2 grams per day, each year. Unexpectedly, the model estimated that this reduction rate actually SLOWED after the government initiative was introduced, to between 0.07 and 0.11 grams per day, each year.

To measure the impact this slowing in salt reduction had on the amount of heart disease, the researchers used yet another computer model. This one looked at a fictional population based on lots of national statistics. Using these estimates, they suggest that the slow-down is responsible for up to ten thousand extra new cases of heart disease in the UK. That is, new cases that would not have occurred if the government initiative had not taken place. Of course, one of the problems with all this computer modelling is that you need to make an awful lot of assumptions and it is sometimes difficult to know if you’ve made the right ones.

The main conclusion of the research paper was that the government initiative to help lower salt intake didn’t really work. However, does this research tell us anything about how much salt in your diet is likely to lead to an increase of your risk in heart disease? No.

If you are interested in that, you need an up-to-date meta-analysis. A good meta-analysis gathers together all studies that look at your question, pools all the results together, and then figures out whether there is an overall effect or not. Here is a very recent one looking at the effect of dietary salt. The authors of this study found that there was a beneficial effect of reducing salt in the diet, but only if your blood pressure was already quite high, i.e., in the top 25% of the population. These things are never simple. Things like your own genetics will also play a role in whether you are likely to develop heart disease or not.

The Daily Mail article dived into this aspect with gusto. Apparently we were all sold a bunch of coconuts when we were told to cut down on our salt. Believing this message is dangerous as it generalises massively. When looking at the whole population bundled together, lowering dietary salt doesn’t seem to reduce the risk of heart disease that much. But this ignores a bigger effect in certain groups, like those with already increased blood pressure. It is not the same story for everyone.

I’m not here to give any health advice – just to unwrap what the science says. Your doctor is always the best person to ask about your own health questions.

Thanks for reading!

Image attributed to: By Mark Schellhase – Own work, CC BY-SA 3.0, https://commons.wikimedia.org/w/index.php?curid=4500571