Can it really be true that the Celtic nations of the UK could save almost 4,000 lives every year by eating like the English?

“You’ve heard of the Atkins, the South Beach, the Hollywood ... now try the English Diet” headlined the story in Friday’s Daily Record through gritted teeth. The claim comes from a study in BMJ Open that says small differences between the diet in England and that in Scotland, Wales and Northern Ireland can account for the higher rates of heart disease, stroke and cancer in the Celtic nations.

The conclusion has already met with some scepticism from a blogger who points out that the biggest percentage gains claimed for a change of diet would be in Wales, which the authors claim already has a diet closer to England’s than do Scotland or Northern Ireland.

Why? If you examine the basic dietary data used by the researchers, there’s an obvious blip in the 2008 Welsh data, which shows them eating more calories, fat and salt in 2008 than in any other year, by a considerable margin.The dietary patterns come from a DEFRA survey which seems to have generated some implausible results in Wales in 2008.

The sample size is small – fewer than 300 households in Wales every year – and the data are usually published by DEFRA as three-year rolling averages to improve reliability. The authors of the study calculate lives saved for three years, and also the total. The 2008 data are anomalous – 1,493 lives saved in Wales in 2008 against 591 in 2007 and 591 in 2009 – which clearly reflects the odd 2008 data.

As a result, they report a bigger percentage reduction of deaths “delayed or averted” in Wales (81 per cent of the mortality gap closed) despite pointing out that the Welsh diet is already quite close to the English one.  

The Scottish diet is worse than that in England or Wales, but very similar to that in Northern Ireland. Both consume fewer vegetables (potatoes excluded) than do England or Wales, and more salt. Yet the calculation shows a smaller proportion of lives would be saved in Scotland than in Northern Ireland by adopting the English diet – 40 per cent of the gap with England would be closed in Scotland, against 81 per cent of the gap in Northern Ireland.

The authors explain this by saying the gap with Scotland is wider, which is true but irrelevant. The real reason is not the size of the gap but the fact that diet and other risk factors do not fully explain it. The so-called “Scottish Effect”, or more narrowly the “Glasgow Effect”, means that death rates are higher than they ought to be in Scotland, once conventional risk factors are taken into account.

But if dietary factors don’t fully account for Scotland’s extra deaths, why should we assume they do for Wales and Northern Ireland? The dietary gap between England and Wales is barely perceptible (2008 excluded) yet it is claimed it accounts for 1,000 extra deaths a year in Wales (600 if you exclude 2008) – 81 per cent of which could be saved by making some very small changes.

It is likely, of course, that the deaths being recorded today are the result of differences in diet years or decades ago, when they may have been wider. An alternative explanation is that the model used, which derives from earlier studies, exaggerates the dietary contribution to heart disease, stroke, and cancer. This possibility is acknowledged by the authors.