Should you be worried about salt?

This post is may seem a little 'off-topic'. But please bear with me a moment. It's about 'evidence-based' public policy.

I find that working on international trade and trade policies places a premium on good evidence as the basis of good public policy. There is a lot of b.s. in the political economy of trade (no surprise). The very first thing to do is to look at the evidence—"who gains and who loses"—because a trade measure works by taxing one group of people to subsidize another, usually much smaller, group.

I'm interested in evidence-based policies in other public policy domains, too, such as climate policy and public health. As funding for public health expands in rich countries, so does the potential for 'noble cause corruption'; the hijacking of public policy by scientists or others who have an institutional interest in putting a particular 'spin' on the science of environment or health, for example.

Here's an article from today's Australian newspaper that caught my attention.

"AWASH chairman Bruce Neal, professor of medicine at the University of Sydney, said that despite entreaties to cut salt levels, the fast-food industry was still "fiddling around the edges". This was despite high blood pressure being a leading cause of death in Australia, alongside smoking." extract from The Australian

As someone diagnosed years ago with mild hypertension, that sort of statement sends me off to do a little research. It turns out that, even if we make allowances for the simplifications of a press release, this is pretty dubious stuff. A half hour of research (reproduced below), shows that it is simply not true that hypertension (high blood pressure) is a 'leading cause of death' in Australia and that it is a matter of some controversy whether reduced salt intake offers significant benefits for the control of blood pressure, particularly in the majority of the population who do not have high blood pressure.

I first checked the Australian Department of Health data (Australian Health and Welfare Institute's "Australia's health 2008") where hypertension is nowhere to be found among the top 'underlying causes'—the term used in the statistics for the main causes—of death. Hypertension does, however, figure on (without leading) the list of important 'contributory' causes of deaths in Australia. In 2005, according to the AIWH, "high blood pressure contributed to more than 14,500 deaths (almost 11% of all deaths, eighth in the ranks of contributory causes) but was the underlying cause in only 410 deaths" (in a higher proportion of female deaths than male, as it turns out).

So what is the evidence on the link between salt intake and the control of hypertension? To check, I consulted the excellent (and open access) Cochrane Library of clinical trials. Recent reviews ('meta-studies') of a number of high quality clinical trials point to at most a small effect on blood pressure due to reduced salt intake. This effect appears to be very small (around 1mg of mercury pressure in the diastolic pressure) in 'normal' people but greater for people with high blood pressure. One study of long-term effects in 2003 found no evidence of a 'dose effect' relationship between the amount of salt reduction and the size of the reduction in blood pressure.

"Intensive support and encouragement to reduce salt intake did lead to reduction in salt eaten. It also lowered blood pressure but only by a small amount (about 1 mmHg for systolic blood pressure, less for diastolic) after more than a year. This reduction was not enough to expect an important health benefit. It was also very hard to keep to a low salt diet. However, the reduction in blood pressure appeared larger for people with higher blood pressure. There was not enough information to assess the effect of these changes in salt intake on health or deaths. Evidence from a large and small trial showed that advice to reduce salt helps to maintain lower blood pressure following withdrawal of antihypertensive medication. If this is confirmed, with no increase in cardiovascular events, then comprehensive dietary and behavioural programmes in patients with elevated blood pressure requiring drug treatment would be justified." extract from Hoooper, Bartlett, Smith et al

Another study (also from 2003) summarizing the clinical trials of short term impacts of reducing sodium in the diet reached a similar conclusion:

"The magnitude of the effect in Caucasians with normal blood pressure does not warrant a general recommendation to reduce sodium intake. Reduced sodium intake in Caucasians with elevated blood pressure has a useful effect to reduce blood pressure in the short-term" extract from Jurgens, G et al

The Hooper study (first quotation above) of long-term impacts has been countered by a similar meta-study by one of the founders of AWASH's UK affiliate. MacGregor et al found, in 2004, that by choosing to look at a different group of trial results with double the average salt reduction of those considered by Hooper (from 2g/day to 4/g per day), they found reductions of 5mg/hg (systolic) and 2.7mg/hg (diastolic) in hypertensive participants. But they found only 2.03 mg/hg and 0.99 mg/hg falls in blood pressure in normotensive participants. They also found a dose-effect.

I am not qualified in medicine. I offer you no advice whatever about salt in your diet (ask your G.P). But, as a public policy analyst, I observe that even the study that is more favorable to the 'salt reduction' campaign, demonstrates very modest impacts from cutting down on salt in the diet. The average outcome for someone with normal diastolic (the lower) blood pressure readings is similar in all the studies and barely significant (at around 1mg/hg). Based on all three of these studies taken together, I do not see how a campaign to change population-wide dietary salt levels is justified, although someone with hypertension might be better advised to consider reducing salt intake.

I make no inferences about the bona fides of Professor Neal's campaign. I presume they are the best. But I question why he is targeting fast-food chains.

"A survey of salt levels in leading fast food brands, to be released today by Sydney-based health experts, found three-quarters of the burger and sandwich-style products sold by six companies provided more than 50 per cent of the suggested daily target. One product - KFC's Zinger Double BBQ Bacon & Cheese Burger - provided the highest amount of salt, with 2410mg of sodium per burger."

We know from international data as well as from the AIWH reports that the biggest 'risk factor' for hypertension is… grey hair. It seems unlikely that the part of the population most at risk of hypertensive problems, and therefore best able to take advantage of the modest benefits of salt reduction, is loading up on salt at these outlets.

Tags for this entry: policy evidence health obesity hypertension salt