Shaking out the truth about salt

For decades, the question of how much sodium people should consume has been roiled by controversy and soiled by special interests. Most research has supported consuming 2300 milligrams (mg)—or even 1500 mg—per day to reduce blood pressure and the risk of cardiovascular disease (CVD). However, some research has suggested that levels below 4000 mg—the amount Americans currently consume1—might actually increase CVD risk. Sodium consumption is critically important to public health, given that almost half of US adults have hypertension, and that reducing sodium intake by about one-third could save as many as 44 000-92 000 lives and $10 billion-$24 billion in health care costs annually.2 In March, the National Academy of Medicine published an authoritative, but little-publicized, report, “Dietary Reference Intakes for Sodium and Potassium,” which will serve as a foundation for nutrition policies in the United States and Canada.3 The report introduced a new term: Chronic Disease Risk Reduction Intake (CDRR) level, saying, “The sodium CDRR is the lowest level of intake for which there was sufficient strength of evidence to characterize a chronic disease risk reduction.” The committee set the sodium CDRR at 2300 mg for adults, based on the “moderate to high strength of evidence for both a causal relationship and an intake-response relationship between sodium and several interrelated chronic disease indicators: cardiovascular disease, hypertension, systolic blood pressure, and diastolic blood pressure.” It also noted “Further reductions in sodium intake below the CDRR have demonstrated a lowering effect on blood pressure.” (The committee said it had insufficient evidence to characterize the impact of reductions below 2300 mg on CVD.) The previous DRI report (2005), relying only on evidence that reducing sodium intake lowers blood pressure, set a Tolerable Upper Intake Level (UL) at the identical 2300 mg. The American Medical Association, American Heart Association, Health Canada, World Health Organization, and other authorities also recommend that people consume no more than 2000-2400 mg of sodium per day. The new committee also set an Adequate Intake (AI) level, a level that would ensure that people get sufficient sodium and other nutrients. The AI was set at 1500 mg for adults—the same as in 2005. The DRI Committee squarely addressed several observational studies, such as PURE4 and some analyses of NHANES data,5 that reported J- or U-shaped curves indicating that sodium intakes based on spot urines or dietary recalls greater than about 5000 mg and levels below 3000 or 4000 mg were associated with increased CVD risk. Acknowledging that observational studies are not definitive, some of their authors have called for large RCTs to assess the health impact of sodium consumption in the range of 2300 mg/d or less before population-wide advice is issued. The DRI committee, however, concluded that studies that estimated sodium intakes using spot urines, instead of multiple 24-hour urine collections, had a “high risk of bias.” The Kawasaki formula, which has been used to convert overnight or other spot urines to daily excretion, overstates 24-hour urinary sodium excretion when intakes are low and understates excretion when intakes are high.6 Possible reverse causality (illness caused persons to consume less food, and therefore, less sodium) and residual confounding were cited as additional limitations of observational studies. Other studies finding J- or U-shaped relationships between sodium and CVD mortality relied on 24-hour dietary recalls, which the DRI committee considered a “major limitation.” Two recent studies add empirical evidence to strengthen concerns about the observational studies. Using multiple 24-hour urines from the Trials of Hypertension Prevention, researchers found a linear relationship between sodium and total mortality.6 However, when they used the Kawasaki formula to estimate 24-hour sodium, a J-shaped curve appeared. By another analysis, even a single 24-hour urine is insufficient to characterize long-term sodium intake. A long-term observational study found a linear relationship between sodium intakes and CVD outcomes when based on multiple 24-hour urines, but a U-shaped relationship occurred when based on only an initial 24-hour urine sample.7 Such studies should further allay the concerns, widely publicized in lay media, that lowering sodium intake to 2300 mg would be harmful. The DRI report might break those political logjams. Last year, FDA commissioner Scott Gottlieb said that FDA would finalize the 2-year targets in 2019. But by coincidence, on the day that the DRI report was released, Gottlieb announced his resignation. While that might delay finalization of the targets, one impediment to implementing policies was eliminated when several days earlier the Salt Institute, the industry's lobbying and PR organization, unexpectedly announced that it would dissolve on March 31. For years, that organization vociferously opposed efforts to reduce sodium consumption. However, other trade groups have begun to fill the vacuum left by the Salt Institute's demise.8 SNAC, the trade association of snack-food manufacturers, and at least four other trade associations commissioned an econometric study that may have been intended to show that the high costs of meeting FDA's targets justified allowing companies more time to meet them. Additional means of encouraging lower sodium diets are to require warning labels on high-sodium packaged foods (as Chile and Israel are doing) and on chain-restaurant menus next to high-sodium meals (as New York City and Philadelphia have done). While the DRI report may not end the controversy over sodium, it provides a solid basis for local and national efforts to lower sodium consumption. It is time for the United States to move forward promptly with effective sodium-lowering initiatives so that Americans can begin to derive the benefits.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.