Interesting questions, all, but only one approaches the basic question related to improving health and that does not address directly the impact of the intervention under consideration.  Unfortunately, the public has had no opportunity to review the analyses themselves to offer comments to assist the DGAC.  We hope the DGAC has at ready reference our earlier letter with references to all the relevant studies, ten in all.  They are also found at http://www.saltinstitute.org/healthrisk.html for your convenience.

It has been only a few weeks since the DGAC met, but we now have two additional sets of recommendations concerning intake levels of dietary salt.  They differ dramatically.

The week after the DGAC meeting, the Canadian Journal of Cardiology published the recommendations of The Canadian Hypertension Society, The Canadian Coalition for High Blood Pressure Prevention and Control, The College of Family Physicians of Canada, The Heart and Stroke Foundation of Canada, and the Chronic Disease Prevention Division, Centre for Chronic Disease Prevention and Control, Health Canada.[4]  These groups unitedly recommended against a universal reduction of dietary salt.  In fact, their recommendation is that a majority of Canadians do not need to reduce salt intake. Salt reduction should be limited to hypertensives and those at “high risk” – defined to mean those that are both salt-sensitive and have DBPs in the 80-89 mmHg range.

The next week, the FNB recommended that every American try to reduce salt to 1,500 mg/day – a 60% reduction from the current average.

When medical experts in two major, similar countries in two successive weeks come to very different conclusions about whether their entire populations should undertake a major dietary change, it should provoke a critical examination of the process that those experts used to review scientific evidence to reach their conclusions.  We hope the DGAC will conduct that critical examination. 

After all, virtually everyone supports “evidence-based” public health policy.  But, remember, so too does “everyone” support highway traffic safety, efficient government and healthy lifestyles.  Everyone doesn’t always mean the same thing when they say they favor safe roads, less government waste or wholesome habits of exercise and eating – or, we might add, when they say they endorse “evidence-based” medicine. 

We need to ask whether the process is truly “evidence-based” or whether it relies, as does the FNB report, on experts’ views of what the science means.  Expert opinion, of course, is considered “Class D” evidence – the lowest type of evidence used in making general recommendations.  What we really need are recommendations based on Class A evidence, evidence of whether the intervention recommended actually improves human health.  There is no evidence that asking everyone to reduce dietary salt improves population health outcomes.  None. Impacts of reduced sodium diets on intermediate variables such as blood pressure, insulin resistance, sympathetic nerve response and plasma renin activity show evidence of both benefit and risk in lowering salt, depending on the variable and the individual. 

Dr. Lawrence M. Resnick, executive editor of the American Journal of Hypertension, and a member of the FNB panel, disputed the published conclusions.  He stated for the record:  “The recommendations of the panel to further reduce dietary sodium intake beyond that already recommended in the past is not justified by a fair analysis of the evidence quoted in the report.  A fair discussion of the data does not allow us to come to a universal recommendation about dietary sodium for the population as a whole.”  He went even further in an interview published by Fox News, calling the recommendations “nonsense.”[5]  We agree.

Dr. Alexander G. Logan, co-author of the Canadian recommendations, hit the nail on the head when he said after the FNB report was published: “Our process in Canada tracks the internationally-accepted approach of the Cochrane Collaboration.  Obviously, the U.S. report has not adopted this rigorous scientific and widely accepted method of assessing evidence to make dietary recommendations.” 

Like Dr. Logan, we believe the evidence-based procedures favored by the Cochrane Collaboration should be used to evaluate the evidence on the question of the possible health consequences of reducing dietary salt.  That is the position of the U.S. Preventive Services Task Force[6] as well, a unit of the Agency for Healthcare Research and Quality of the U.S. Department of Health and Human Services.  Both the Cochrane Collaboration and the U.S. Preventive Services Task Force analyses found insufficient evidence to recommend universal sodium reduction.

We say: follow the evidence.  Use the rules of evidence-based policy-making.  Please.

In the Canadian process, the committee first applied a pre-determined study method to assign grades to intervention studies.  The “rules” gave greater weight to evidence from adequate randomized controlled trials (RCT), quality subgroup analysis and systematic reviews of RCT data than data from case series or observational studies.  The committee next used pre-determined rules to consider the precision of the results, with substantially more weight being given to statistically significant results in positive studies, high-quality negative studies that ruled out an important difference and systematic reviews with homogeneous results.  Finally, the Canadian process valued the applicability of studies, weighing more heavily studies with important health outcomes and a representative study population.  Only then were recommendations made by content experts independently reviewed and graded by experts in the process of health policy formulation.  This process ensured that high-grade recommendations are internally valid, have a high degree of precision and are widely applicable. 

Is this the approach being followed by the salt subcommittee to try to sift through the myriad of opinion to evaluate the data?  Is the effort to find Class A data and “settle” for Class B studies or to give up on “evidence-based” determinations and make recommendations on Class D expert opinion.  With Dr. Appel on the committee it is difficult to imagine a greater conflict of  interest. Our Dietary Guidelines deserve a foundation of strong evidence.

Fortunately, the Cochrane review has done the heavy process-related work to provide an exceptionally-sound scientific foundation.  The hard part for the DGAC will be assembling the verbiage extolling an evidence-based conclusion that is a marked departure from past “expert” reviews (including both the most recent DGAC report and the FNB report).

I’ve enclosed a discussion about evidence-based dietary guidelines published in the Journal of the American Dietetic Association discussing the imperatives of process discipline in evaluating evidence for population recommendations.[7]

Several other points are worth mentioning.

The DGAC received some very useful input on January 28^th from Dr. Frank Hu.  Dr. Hu pointed out that the “Healthy Eating Index” (which included sodium reduction) was not a good predictor of chronic disease.  His alternative index did better – and excised sodium reduction from among the index variables; the revised index was associated with reduced CV risk.  This is a nice post-facto confirmation of the results of the process advocated by the leading heart and stroke organizations in Canada, by the Cochrane Collaboration and by the U.S. Preventive Services Task Force.  We commend it as a model for the DGAC.

Since the FNB report has put the 1,500 mg/day sodium figure on the table, we feel it would be appropriate to comment on the December 2003 article by He and MacGregor[8] advocating general reduction to 3 grams of salt per day (1,200 mg/day).  This recommendation is based on having the population reduce sodium by 150 mmol/day.  Not only is the methodology ridiculous (the projection is based on drawing a regression line through two points, one for each of the two referenced, short-term studies with no justification offered justifying the extrapolation), but Americans only consume a total of 150 mmol day so this would mean the recommendation is to have them cut out salt use altogether.  No doubt the DGAC already reached that conclusion, but I thought it worth mentioning.

As mentioned above, the impact of dietary salt on blood pressure is important, but only one of the intermediate variables leading to the health impact of the intervention.  In that regard, the DGAC will be interested to read the report published last year in the Journal of the American Society of Nephrology by Drs. Vallon, Blantz and Thomson[9] on the “salt paradox” in diabetes.  The paper documents an increase in plasma renin activity on low salt diets.  The elevated PRA, in turn, contributes to vasoconstriction of the afferent blood vessel and damage to the diabetic kidney.  Thus, salt restriction is an added risk for diabetics.  This is another, largely unexplored, impact of salt reduction.  Studies like this remind us of the imperative of finding Class A evidence (or at least not relying on Class D evidence) when considering advocating a population-wide intervention.

Finally, we would remind the DGAC of its obligation to provide recommendations to HHS and USDA that are actionable.  Not only does this mean that the recommendations must be scientifically-based and capable of being implemented in the real world, but the process and scientific evidence employed must satisfy the new Data Quality Act.  Though an outstanding quality study, the DASH-Sodium Study, at present, fails the DQA standard and incorporation of it – or the FNB report that utilizes its data and analysis – will jeopardize the utility of the DGAC report.  Since the DGAC includes the principal investigator, we recommend he be required to divulge these data if they are to be considered in fashioning the DGAC recommendations.

Sincerely,

Richard L. Hanneman
President