Insights From the Institute of Medicine Committee
Brian L. Strom, MD, MPH; Cheryl A. M. Anderson, PhD, MPH, MS; Joachim H. Ix, MD, MAS
The recent Institute of Medicine (IOM) report regarding dietary sodium1 has generated considerable interest and debate, as well as misinterpretation by advocates on both sides. Further discussion is necessary to inform the public and the health care community and to inform public health strategies for sodium reduction.
CURRENT PUBLIC HEALTH RECOMMENDATIONS REGARDING DIETARY SODIUM
Dietary sodium intake averages approximately 3400 mg/d in US adults, far in excess of the Dietary Guidelines for Americans (DGA) recommendation of less than 2300 mg/d for those older than 2 years and less than 1500 mg/d for certain high-risk subgroups, including African Americans, individuals with hypertension, diabetes, or chronic kidney disease (CKD), or those older than 50 years.2 In contrast, the 2005 IOM Panel on Dietary Reference Intakes (DRI) for Water, Potassium, Sodium, Chloride, and Sulfate3 found insufficient evidence to derive a “recommended dietary allowance” for sodium. Instead, an “adequate intake” of 1500 mg/d of dietary sodium was determined, reflecting the minimum needed to achieve a diet adequate in essential nutrients and to cover sweat losses. Additionally, the 2005 IOM panel established a “tolerable upper intake level,” using projections from available data on the effects on blood pressure, that consumption up to 2300 mg/d was unlikely to cause harm.
Based on the strength of the blood pressure data, various US (eg, American Heart Association [AHA]) and international (eg, World Health Organization [WHO]) organizations published recommendations for sodium consumption.4- 5 Although these recommendations were somewhat different from the DGA, there was general agreement that sodium consumption is excessive worldwide and should be reduced. Despite these recommendations, more than 90% of US adults consume more than 2300 mg of sodium per day, and among the high-risk subgroups more than 98% consume more than 1500 mg of sodium per day.6
A substantial body of evidence supports efforts to reduce sodium intake. This evidence links excessive dietary sodium to high blood pressure, stroke, and cardiovascular disease (CVD).1 However, effects of sodium on blood pressure cannot always be disentangled from effects of total dietary modification, and effects of other electrolytes on blood pressure remain unresolved.7 Concerns have been raised that a very low sodium intake may adversely affect lipids, insulin resistance, renin, and aldosterone levels and potentially may increase risk of CVD and stroke. Some studies link sodium intakes of less than 2300 mg/d to increased risk of CVD, at least in subpopulations. Thus, debate emerged about the sodium intake target that best improves health outcomes.
In response, the US Centers for Disease Control and Prevention commissioned the IOM to convene an expert committee to examine the designs, methods, and conclusions of literature published since the 2005 DRI report.3 Specifically, the committee was asked to review and assess potential benefits and adverse outcomes of reducing sodium intake in the population, particularly in the range of 1500 to 2300 mg/d, with emphasis on the high-risk subgroups. The committee was asked to focus on studies of direct health outcomes (vs surrogate end points such as blood pressure), to comment on implications for population-based strategies to reduce sodium intake, and to identify methodologic gaps and ways to address them. The committee's full report is published elsewhere.
SODIUM AND DIRECT HEALTH OUTCOMES
The committee searched literature published through 2012 for relevant publications. Information also was gathered from an open public workshop. Although not its primary emphasis, the committee summarized studies published since 2003 evaluating intermediate markers, particularly blood pressure. Focusing on CVD outcomes, the committee's assessment of evidence was guided by factors such as study design, quantitative measures of dietary sodium intake, confounder adjustment, and number and consistency of available studies.
FINDINGS AND CONCLUSIONS
General US Population. Studies linking dietary sodium intake with direct health outcomes were highly variable in methodological quality; limitations included overreporting or underreporting of sodium intake. However, when considered collectively, the evidence on direct health outcomes indicates a positive relationship between higher levels of sodium intake and risk of CVD, consistent with the known effects of sodium intake on blood pressure. Furthermore, in some studies, the association between sodium and CVD outcomes persisted after adjusting for blood pressure, suggesting that associations between sodium and CVD may be mediated through other factors (eg, effects of other electrolytes) or through pathways other than blood pressure.
Studies evaluating sodium intake in the range of 1500 to 2300 mg/d demonstrate evidence of blood pressure lowering, but no studies have examined sodium intake in that range in the general population and direct CVD outcomes. The committee found that studies on direct health outcomes were of inconsistent quality and insufficient quantity to conclude whether sodium intake of less than 2300 mg/d was associated with either a greater or lesser risk of CVD.
Population Subgroups. The committee reviewed multiple randomized trials conducted by a single team that indicated low sodium intake (up to 1840 mg/d) may lead to greater risk of adverse events in patients with heart failure (HF) with reduced ejection fraction who received aggressive therapeutic regimens. Because these therapeutic regimens were different from standard US practice, trials using regimens that more closely resemble standard US clinical practice are needed. Of note, due to allegations of duplicate publication in 2 of these trials, a meta-analysis including them was recently retracted, after the IOM report's completion.8 Another recently published small randomized trial involving patients with acute decompensated HF showed no benefit on weight or clinical stability from a combination of sodium and fluid restriction.9
The committee reviewed 2 related studies in individuals with prehypertension that suggested benefit from lowering sodium intake to 2300 mg/d and perhaps lower, although these studies were based on small numbers of persons with sodium intake in the less than 2300 mg/d range. In contrast, for patients with diabetes, CKD, or preexisting CVD, the committee found no evidence of benefit and some evidence suggesting risk of adverse health outcomes at sodium intake of 1500 to 2300 mg/d. In studies that explored statistical interactions, race, age, hypertension, and diabetes did not modify associations of sodium with health outcomes. The committee concluded that, with the exception of heart failure, evidence of both benefit and harm is not strong enough to indicate that these subgroups should be treated differently from the general US population. Thus, the committee also concluded that evidence on direct health outcomes does not support recommendations to lower sodium intake within these subgroups to or even less than 1500 mg/d.
IMPLICATIONS FOR POPULATION-BASED STRATEGIES TO GRADUALLY REDUCE SODIUM INTAKE
Although not asked to specify targets for dietary sodium, the committee noted factors that precluded establishing these targets. These include lack of consistency in methods for defining sodium intakes at both high and low ends of typical intakes and extreme variability in intake levels across studies. The committee could only consider sodium intake levels within the context of each individual study because there were impediments to calibrating sodium assessment measures across studies.
After release of the IOM report, several news outlets highlighted disagreement among health agencies about targets for dietary sodium intake and reported that experts disagreed about the importance of blood pressure. Focusing the debate on specific targets misses the larger conclusion with which all are in agreement and may hinder implementation of important public health policy. Rather than focusing on disagreements about specific targets that currently affect less than 10% of the US population (ie, sodium intake of <2300 mg/d vs <1500 mg/d), the IOM, AHA, WHO, and DGA are congruent in suggesting that excess sodium intake should be reduced, and this is likely to have significant public health effects. Accomplishing such a reduction will require efforts to decrease sodium in the food environment10 and provide individual consumers more choice in their dietary consumption of sodium.
