High Salt Intake Health Risk Debunked?

Would you please pass the salt?

High-salt diets may not increase the risk of death, contrary to long-held medical beliefs, according to investigators from the Albert Einstein College of Medicine of Yeshiva University.

They reached their conclusion after examining dietary intake among a nationally representative sample of adults in the U.S. The Einstein researchers actually observed a significantly increased risk of death from cardiovascular disease (CVD) associated with lower sodium diets.

The researchers analyzed data from the Third National Health and Nutrition Examination Survey (NHANES III), which was conducted by the federal government among a nationally representative sample of U.S. adults. These data were then compared against death records that had been collected by the government through the year 2000. The sample of approximately 8,700 represented American adults who were over 30 years of age at the time of the baseline survey (1988-1994) and were not on a special low-salt diet.

After adjusting for known CVD risk factors, such as smoking, diabetes and blood pressure, the one-fourth of the sample who reported consuming the lowest amount of sodium were found to be 80% more likely to die from CVD compared to the one-fourth of the sample consuming the highest level of sodium. The risk for death from any cause appeared 24% greater for those consuming lower salt, but this latter difference was not quite large enough to dismiss the role of chance.

“Our findings suggest that for the general adult population, higher sodium is very unlikely to be independently associated with higher risk of death from CVD or all other causes of death,” says Dr. Hillel W. Cohen, associate professor of epidemiology and population health at Einstein.

So does this mean that previous results of poorly constructed statistical analysis has caused a generation of patients to needlessly suffer through bland diets and salt deprivation? Makes me wonder how much other medical and nutritional advice routinely given also isn't valid?