Another one bites the dust

Another what bites the dust? Another one of the shibboleths of “healthy living” that the nutritional establishment has been pounding us over our heads with for decades: the idea that salt is bad for us.

Now, in the wake of the three Woman’s Health Initiative studies showing that fat doesn’t seem to cause heart disease nor cancers or the breast or colon, comes a study from the venerable NHANES II data showing that not only does salt intake (or to be more precise, sodium intake) not cause premature death from heart disease it actually seems to protect against it. And consuming more sodium appears to protect against premature deaths from not just heart disease but from all other causes as well. It’s been a bad couple of weeks for the holier-than-thou crowd.

In the current issue of The American Journal of Medicine is a study filled with much interesting information that gives a peephole into the way the pinheads in the government issue edicts that affect the health of all those ignorant enough to abide by them. This study shows that the subjects who consumed the least sodium in the late 1970s had greater rates of death from cardiovascular disease and from all other causes than those who consumed more sodium. Before we delve into all that, however, let’s look at what the paper shows about how government works.

The US Department of Health and Human Services and the US Department of Agriculture 2005 nutritional guidelines (click here to read in full in a large pdf download) recommend that Americans consume less than 2300 mg of sodium per day (which is less than the 2400 mg recommended in the 2000 guidelines) in order to “prevent or delay the onset of high blood pressure..” and “to lower elevated blood pressure” Seems rationale enough until one considers that there is really no good evidence that sodium intake causes blood pressure to increase other than that shown in short-term clinical trials, a number of which are inconclusive or contradictory. It’s just like with the idea of low-fat: somewhere, sometime, someone got it into his or her head that dietary sodium is bad, the word spread, and researchers start doing studies to prove it. As long as a study here or there confirms this bias, then the idea is held in the minds of many people not simply as an hypothesis, but as a truth. In the case of the nutritional guidelines, the scientific committee making recommendations did so

largely based on the blood pressure reduction associated with lower sodium in short-term clinical trials. However, these trials could not assess the long-term cardiovascular morbidity and mortality consequences of lower sodium. Of concern is that lower sodium intake can generate increased activity of the renin-angiotensin and sympathetic nervous systems, and possibly increased insulin resistance, and each of these could have adverse cardiovascular effects. Morbidity and mortality outcomes will be influenced by unfavorable and favorable effects, as well as the unknown consequences of a diet altered to achieve lower sodium intake. In the absence of clinical trial data, several observational studies, with contradictory results, are available.

So, it’s not even really a case of unintended consequences. The scientific committee chose to overlook evidence clearly showing that there could easily be a downside to sodium restriction in favor of their built-in bias against salt. In fact, based on no good evidence they lowered the recommendation from that of the time before. Gives one a lot of faith in the nutritional guidelines, doesn’t it?

Basically here is how the study was done. Researchers used the data from NHANES II study (a large pdf of the NHANES II study can be downloaded by clicking here) to determine if sodium intake correlated with premature death. The NHANES II researchers interviewed and examined participants and collected data in 1976-1980. The nutrient intake data came from one 24 hour recall done by trained interviewers, which isn’t as good as a 3 or 4 day food diary, but is better that a food frequency questionnaire. The study was followed up by evaluating the mortality statistics as of December 31, 1992 to determine the numbers of deaths in the study subjects and their causes. As the researchers put it in reference to these study subjects without a hint of tongue in cheek:

Those not found to be deceased were assumed alive at that date.

Indeed. I wonder if there were another choice.

The researchers set the breakpoint of their data analysis at the 2300 mg of sodium recommended in the nutritional guidelines. After analyzing the nutritional and mortality data on this basis it turned out that those subjects who consumed less than 2300 mg of sodium per day had a 1.37 times increased risk (95% CI 1.03-1.81, P=.033) of dying from heart disease and a 1.28 times increased risk (95% CI 1.1-1.5, P=.003) of dying from all causes as compared to those who consumed more than 2300 mg of sodium per day.

As the authors of the study put it:

The principal finding in this representative sample of US adults is that sodium intake, measured as a continuous variable and adjusted for calories by each of three distinct methods, had a statistically significant and inverse association with CVD mortality, independent of known cardiovascular risk factors. Results were consistent for all-cause, CHD, and cerebrovascular-specific mortality, although these latter associations were not statistically significant. In addition, individuals reporting consumption of sodium consistent with the most recent US dietary guidelines of

Makes you glad you spent all that time watching your salt doesn’t it?

I have a little different take on the results of this study from a statistical standpoint. Harkening back to the long and complex post about confidence intervals a while back you might remember that the parenthetical expression after the risk ratio shows what the range is statistically with 95% confidence that it will actually fall in that range. In the case above for the cardiovascular mortality risk of 1.37 what those figures actually mean is that there is a 95% probability (not 100%) that the actual risk will fall into the range of 1.03 to 1.81. In other words we can say that with 95% confidence if you consume less than 2300 mg of sodium per day you have somewhere in the range of 1.03 to 1.81 times greater risk of dying prematurely from heart disease than if you consume more than 2300 mg of sodium per day.

Because of the 5% uncertainty (100% minus 95%) and the fact that it could just as easily be the 1.03 figure as the 1.81 figure for risk I don’t like to take these things as gospel unless the risk ratio is at least 2.0. But that’s just me. Others who are less statistically nit picky are more than happy to go with the lower risk ratios.

Based on my more statistically rigid interpretation of the data, I can conclude that at the very least it doesn’t make any difference how much sodium you consume. To me, that’s the take home message.

Oh, one other take home message: don’t believe scientific committees. In the discussion part of this paper the authors laid out a sampling of studies looking at sodium intake verses disease:

Eight previous observational studies examined clinical outcomes associated with sodium levels. Sodium intake was inversely and significantly associated with higher CVD mortality for the entire sample of NHANES I, with myocardial infarction among male participants in a prospective cohort study of treated hypertensives and with all-cause mortality in men in the Scottish Heart Study. By contrast, a statistically significant direct association of sodium with CVD and all-cause mortality was observed in a Finnish community sample, among the overweight subset of NHANES I, with CHD incidence among women in the Scottish study, and with stroke in a community sample in Japan. No statistically significant associations were reported either among Japanese-American men of the Honolulu Heart Study or in the MRFIT cohort.

All these studies were done and published long before the scientific committee made their recommendations for the nutritional guidelines and were inconsistent and inconclusive at best, so how could the recommendation to lower sodium even further be even contemplated much less done? Bias. Or its synonym prejudice.

Ambrose Bierce (one of my favorite writers) defined prejudice as:

A vagrant opinion without visible means of support.

Please pass the salt.

Monthly Book Reviews


I have been writing a series of book reviews each month that I email to subscribers. If you're interested and want to get on the list, sign up here (or above where it says Get free email alerts in the upper right). I'll send you an email notice of all new blog posts plus all my monthly book reviews. Also, you will get a link to all the previous month's book reviews I've sent. Hope to see you aboard.

Powered by ConvertKit

Please note: I reserve the right to delete comments that are offensive or off-topic.

Leave a Reply

Your email address will not be published. Required fields are marked *

4 thoughts on “Another one bites the dust

  1. Those not found to be deceased were assumed alive at that date.

    This strikes me as a statistical fudge to increase the significance of the results. The best way to deal with people lost to follow-up is to drop them from all statistical calculations and therefore results.
    The next best, in my opinion, is to assume that they had a similar mortality rate to the population with follow-up. It is unreasonable to assume that everybody lost to follow-up at that point is still alive. Approximately 19% (1343 out of 7154) of the rest of this sample were deceased at the time of the analysis.
    This would seem to create a negative bias (i.e. that the study is understating the significance of the data) to the results because it is clear that there would have been deaths in this subgroup. If (admittedly a very big if) they were similar to the larger group, it would strengthen the association between lower salt intake and total mortality and possibly the other mortalities. In my quick glance at the paper, it is unclear how many were lost to follow-up.
    However or possibly in addition to the negative bias, it also needs to be taken into consideration that counting the missing data points as having determinate outcomes (i.e. they are alive) unfairly strengthens the statistical power of the results, creating a positive bias to the results,(i.e. the study is overstating the significance of the data).
    Thanks in advance for allowing me the space to use this blog to express my viewpoint or to stimulate my thought processes if you don’t choose to publish it.

  2. Severe salt restriction has been something I’ve been fighting for years!!!! I can’t tell you how many people I know that treat salt as if it’s poison….and they try to get as low an intake as possible. Pointing out the dangers of low serum sodium seems to fall on deaf ears. I’ve even heard people say they’ve been told they have low serum sodium by their docs, but still refuse to increase their intake!

    I fear though, that this report, like the ones on low fat, will be disputed as being flawed and the medical community will continue to tell people to avoid salt at all costs!