Introduction

High sodium intake is the leading dietary risk for death globally and its attributable disease burden is estimated at 1.8 million or more deaths and close to 45 million disability-adjusted life years (DALYs) in 2019 [1••]. Hence, reducing sodium intake is a global population health priority. Analyses indicate reducing dietary sodium is one of the most cost-effective interventions to enhance population health, labeled a “best buy” by the World Health Organization [2]. Unlike many clinical interventions, reducing dietary sodium is cost-saving for governments with returns on investment of between 12 and 78$ for every dollar invested [3, 4].

Current best evidence includes randomized controlled trials that show reducing dietary sodium linearly reduces blood pressure in those with normal or increased blood pressure, in different ethnicities and different age groups with greater blood pressure reductions in those who are older, those who are black, and those with raised blood pressure [5••, 6•, 7•]. Randomized trials examined sodium intake levels as low as 400 mg/day and show a reduction in sodium lowers blood pressure to the same or a greater degree when the initial sodium intake is low and that the reduction in blood pressure is approximately twice as large in studies that are 2 weeks or longer compared to those of shorter duration [6•, 7•]. Approximately, 1/3rd of global hypertension is attributed to high dietary sodium [8, 9]. A meta-analysis of randomized controlled trials shows that a reduction in dietary sodium from 3646 to 2690 mg/day causes a linear 26% and 15% reduction in cardiovascular disease (CVD) and mortality, respectively [5••]. The reductions in CVD are also consistently found in meta-analyses of cohort studies that use quality criteria to exclude studies with low-quality designs and methods likely to produce spurious findings but not in meta-analyses without quality criteria [10,11,12,13,14].

There are multiple independent scientific reviews on dietary sodium overseen by major governmental organizations (e.g., World Health Organization, USA, UK and Ireland, Australia and New Zealand, European Union) [5••, 15]. Most major non-governmental hypertension recommendation processes also review the evidence on dietary sodium and are often overseen by major hypertension and cardiovascular organizations. The World Health Organization and other governmental review processes that we are aware of are exhaustive, comprehensive, and conducted over multiple years; many exclude individuals with conflicts of interest, have rigid evidence-based processes, and have an external review before release. The non-governmental reviews are of variable quality, although several are strongly evidence-based high-quality processes. These reviews are conducted regularly with, in general, at least one major report published in most years. Nearly all independent scientific reviews, whether governmental or nongovernmental, recommend reductions in dietary sodium to be under 2400 mg/day for adults as a minimum; the World Health Organization recommends reductions in dietary sodium to less than 2000 mg/day for adults [5••, 15].

The evidence supporting dietary sodium reduction is one of the strongest available for a dietary intervention given the difficulties in performing very large long-term interventions that alter diet, the ubiquitous addition of sodium to the food supply and salt (and other sodium-containing ingredients used at home), and the consensus that it is unethical to intervene to provide increased dietary sodium long term. Nevertheless, many view that reducing dietary sodium is still controversial [16••]. There are several potential reasons for this. One reason is the lack of large outcome-driven randomized controlled trials which are unlikely to ever be conducted [17]. Also, low-quality observational studies finding a J-shaped relationship between sodium intakes and cardiovascular disease have been widely publicized in major media. However, it is important to note that other nutritional recommendations are supported by less strong evidence and are not viewed as controversial (e.g., fruits and vegetables, reducing dietary sugar, eliminating industrial trans-fat, and increasing dietary potassium).

Health and scientific organizations have expressed concerns that the controversy is largely generated by low-quality research, while others are concerned about conflicts of interest contributing much to the controversy [5••, 18,19,20,21,22,23,24,25,26,27,28,29,30,31,32]. The reality is that the controversy has been led by a relatively small number of highly vocal scientists and clinicians. The health and scientific opponents to reducing dietary sodium, together with the food and salt industries and their various umbrella and associated organizations, have used the aura of controversy to impede progress in implementing policies to reduce dietary sodium at a population level, a strategy commonly used in the past for other issues of public health relevance (e.g., tobacco smoking) [33]. In this article, we provide some of our personal experiences in interactions with people and organizations opposed to dietary sodium reduction and propose that broad steps be taken to protect public health policies in general and sodium reduction policies specifically. We accept that differing scientific views are important to foster healthy debates for scientific advancement. However, scientific advancement is driven by striving for truth, using rigorous research methods and designs unimpeded by flawed research bias and commercial influence.

Concerns Relating to Advocacy Against Reducing Dietary Sodium

Table 1 indicates some concerns about advocacy against reducing dietary sodium. We have categorized the concerns regarding (1) use of inappropriate research methods and lack of rigor in research; (2) conflicts of interest with, or biases for, commercial organizations; (3) lack of public access to data supporting research; (4) scientific and professional conduct; (5) journal responsibility in publishing low-quality controversial research and inadequate review processes; (7) grants and research and ethics committee approvals for inappropriate study designs and methods; (8) health and scientific organization issues; and (9) lack of government oversight.

Table 1 Concerns with research quality and organizational policies related to dietary sodium reduction
Full size table

Use of Inappropriate Research Methods and Lack of Rigor in Research

A major problem with research is the use of low-quality or inappropriate methods and research designs prone to spurious results. For example, it is now known that the use of estimating formulas that convert spot urine sodium values to 24-h urine sodium estimates produces biased estimates of sodium excretion that change the linear association between sodium intake from multiple 24-h urines samples (gold standard estimate of sodium intake) and mortality to a J-curve [34••, 35]. Several methods, such as food recall surveys, have been demonstrated to be invalid estimates of individuals’ customary sodium intake [36,37,38,39]. Even using a single 24-h urine sodium collection to estimate intake either attenuates the association with outcomes or can cause a spurious J-curve, relative to multiple 24-h urine samples collected on nonsequential days [40]. Resolving these issues is the responsibility of granting agencies, research committees, ethics committees and journals, and the investigators. Nevertheless, studies of low quality are numerous and often published in journals with high impact factors, partially due to the large sample size of some of these studies [41,42,43]. However, a large sample size cannot make up for erroneous estimations of sodium intake [44]. These low-quality studies seem to bypass critical reviews in journal processes and they even feature editorials by other dissenting scientists and, in one case, by a consultant/advisor to the Salt Institute and a past witness for the tobacco industry [45] (https://www.industrydocuments.ucsf.edu/tobacco/results/#q=Oparil&h=%7B%22hideDuplicates%22%3Atrue%2C%22hideFolders%22%3Atrue%7D&subsite=tobacco&cache=true&count=918, accessed March 17, 2021; https://www.cspinet.org/new/200610022.html, accessed March 18, 2021).

Conflicts of Interest and Commercial Bias

In several instances, scientists with conflicts of interest have denied or been inconsistent in disclosing their conflict of interest [23, 46, 47] https://www.cspinet.org/new/200610022.html. Accessed March 18, 2021); http://www.megaheart.com/pdf/beard140.pdf accessed March 9, 2021). In one particularly grievous example, an American College of Nutrition’s entire journal supplement was supported by the International Life Sciences Institute (ILSI), a non-profit research organization funded by industry. The supplement theme opposed dietary sodium reduction and used many authors, including the guest editor, that were advisors or consultants to the food industry and/or Salt Institute (a now-defunct industry trade association), without providing any disclosures [48,49,50]. Although these omissions overtly violated journal policy and were identified, no corrective action was taken by the journal other than an after-the-fact disclosure by the publisher [51]. The ILSI distributed the supplement free of charge for many years after its publication and still features the supplement on its website in 2021 (https://ilsi.org/search/?kp=2&k=dietary+sodium&ks=1&kt=any). The ILSI still claims that all its scientific activities have a primary public purpose and benefit, that it has a commitment to achieve and maintain the highest standards of scientific integrity, and that all interests, conflicts, and biases are declared. It is notable that there is a close relationship between the published findings of review manuscripts on nutrition topics and the presence of conflicts of interest among authors [43, 52, 53].

Lack of Public Access to Data Supporting Research

We also note that some cohort databases that find adverse effects of reducing dietary sodium have data that are not in the public domain, and hence, the results cannot be easily independently verified [54,55,56]. This is important for several reasons. In cohort studies, different adjustments for confounding factors can alter the association of sodium to outcomes and cause false J-curves [57]. There is one example of analyses from two public databases, finding adverse outcomes associated with lower sodium intake where independent analyses of the databases did not confirm any harm from lower sodium intake but found harmful effects only from high intake [58,59,60]. There is at least one instance where a standard formula for assessing the completeness of urine collection was altered in a fashion that enhanced the proportion of complete 24-h urines and important data on a validation study was not disclosed even after repeated requests [23]. Furthermore, it is often possible to test if the analyses showing harm from lower sodium consumption are independent of sodium by inserting constant sodium values into analyses [35]. The controversial studies have not assessed the independence of their findings from sodium values in their analyses to date.

Scientific and Professional Conduct

It is inconceivable that responsible granting agencies and scientific and ethics review processes would approve all the weak and inappropriate research designs that have been published. Controversial results may have derived from post hoc secondary analyses, not originally planned and reviewed by granting bodies and ethics committees. This raises serious questions about the lack of scientific rigor and conduct of authors, given journal requirements for disclosures of research funding and ethics review. In one case, there was a series of single-center randomized controlled trials finding harm from reducing their sodium intake in people who had heart failure [23]. In a meta-analysis, it was noted that there were identical data from different trials with the investigator claiming it was coincidental [23, 61]. When the journal requested the supporting data, the investigator indicated the data (from multiple trials) was lost in a computer failure. Although this led to the withdrawal of the meta-analysis, the journals with the primary publications did not withdraw those publications, knowing there was no data to confirm the study results, even when requested to do so by multiple scientists. These publications are still cited in support of controversial reviews, even after the retraction date (https://pubmed.ncbi.nlm.nih.gov/22914535/ accessed March 18, 2021).

Additional concerns arise when publications and presentations have erroneous or misleading statements or take scientific reviews out of context [23, 46]. This represents a failure of the peer-review system of journals and raises questions of professional conduct. A few examples of documented erroneous and misleading statements are provided in Table 1, under scientific and professional conduct. One recent false statement, repeatedly published, is that there is no evidence that reducing dietary sodium below 3000 mg/day reduces CVD [16••]. The National Academy of Sciences, however, published a meta-analysis of randomized controlled trials which found that reducing sodium on average from 3646 to 2690 mg/day reduced CVD by 26% and mortality by 15%, respectively [5••]. Arguably, the highest quality cohort study conducted to date also found reducing dietary sodium to under 2300/day reduces CVD [62]. Recent reviews also claim there are no successful programs to reduce dietary sodium [16••, 63••]. On the contrary, there are several population interventions that not only reduced dietary sodium significantly but also there were associated reductions in CVD mortality [64, 65, 66•].

Journal Issues

Recently, we have experienced rejections of “letter to the editors” that point out serious methodologic flaws and false statements in published manuscripts. We also note that some editors have made public statements against reducing dietary sodium, a bias against very consistent evidence reviews on a major public health issue [67].

Granting Agencies, Research and Ethics Committees, Health and Scientific Organizational Issues

On preliminary investigation, one of the authors was unable to confirm or refute if a granting agency had reviewed and approved specific secondary analyses on dietary sodium from a large cohort study. This raises concern as to whether these analyses were reviewed and funded as the publications suggest or were carried out post hoc without ethics and scientific review.

Health and Scientific Organization Issues

Health and scientific organizations often feature debates on dietary sodium although their own evidence reviews, and recommendations, support reducing dietary sodium. These debates often have featured low-quality research, as well as erroneous and misleading statements to support that low dietary sodium may be harmful [23, 46]. The debate formats also provide a sense of false equipoise and credibility for dissenting scientists and offer new platforms for the presentation of low-quality research methods creating an aura of controversy. Similar to modern media formats, by allowing “equal representation,” these debates “…only succeed in creating ‘false equivalence’ between two sides of an issue even when there is only one credible side” [68]. We also note that several openly dissenting scientists have achieved high-level positions in scientific and health organizations (e.g., Presidents of the American Society of Hypertension, International Society of Hypertension, American Heart Association, World Heart Federation) and at least two have used the platforms they are provided by their position to advocate against dietary sodium policies even in the face of their organizations support for dietary sodium reduction. In one case, a president even created a committee dominated by dissenting scientists to create a document that did not support reductions in dietary sodium [67].

Government Issues

Governments have largely been complacent about these threats to evidence-based public policy positions. There are several cases where industry lobbying has significantly delayed or deterred policies from being implemented, despite the government having already described the established evidence in support of the policy.

Action to Support Evidence-Based Dietary Sodium Public Policies

In nutritional science, there has been a long-standing lack of ethical guidance for the industry, health and scientific organizations, health care providers, journalists, and scientists [52, 76,77,78,79]. With many millions of lives at stake, solutions can be at the level of providing guidance on standards of acceptable and unacceptable conduct. Very high standards of conduct and rigorous research designs and methods should be expected for everyone involved in policy research, advocacy, and design.

In our view, being able to obtain research funding or personal financial gain from opposing public policy is concerning. Given that definitive research has established that a large proportion of CVD and hypertension have high dietary sodium as a cause [1••], potential conflicts of interest with the pharmaceutical industry, which profits from treating CVD, should be disclosed at a minimum when presenting and publishing research on dietary sodium policy. Opponents of public policy designed to reduce sodium consumption should be held accountable for the accuracy and truthfulness of their statements, in light of the compelling scientific evidence supporting such policies, in the same way as society as a whole today would require from those denying the ill-health effects associated with tobacco smoking or opposing vaccination. Table 2 provides the authors’ suggestions for potential policies and actions that could protect dietary sodium public policies from low-quality research, commercial interests, and scientists and clinicians who have displayed questionable conduct.

Table 2 Potential solutions to increase protection for healthy dietary sodium policies
Full size table

Academic organizations should have strict guidelines on obtaining research funding or personal financial gain from opposing important public policies such as reducing dietary sodium and reducing tobacco use. Where faculty oppose public policy based on their research findings, academic institutions should ensure there was a full scientific and ethical review of the protocols used and that the research methods and designs were rigorous. Similarly, academic organizations should establish and implement enforcement protocols for false or misleading statements in important public health areas.

Funding bodies should ensure the research they fund that might impact important areas of public policy, such as dietary sodium, and use rigorous designs and methods. They should clarify that when they fund a primary study with specified analyses, that the scientific and ethical review does not cover unspecified secondary analyses. Secondary previously unspecified analyses should require separate scientific and ethical review. Funding bodies should require the databases that are created through research funding are put in an accessible domain once the analyses in the grant are conducted by the investigators, or after a given reasonable time has passed after the study is complete (e.g., 2 years). This is particularly important for research in major public policy areas. Funding bodies can also develop calls for research on the scientific impact of financial interests, low research quality, misinformation, and professional conduct. Currently, financial interests are often disclosed decades after the fact and in the case of tobacco was only through legal actions.

In our opinion, editors have sometimes been negligent in publishing low-quality research on dietary sodium. Large studies using methods highly likely, and now proven, to provide spurious results were and continue to be published and cited, often without acknowledging the weaknesses of the research [16••, 63••, 80]. Low-quality previously published research is used in the introduction and discussion sections of manuscripts to justify further use of low-quality methods and to validate findings. Higher quality research and major scientific reviews are often ignored, dismissed, or taken out of context [16••, 65].

A consortium of international health and scientific organization formed to develop minimum standards for the conduct of clinical and population research on dietary sodium (https://warwick.ac.uk/fac/sci/med/staff/cappuccio/who/true/, accessed March 11, 2021). To date, editors have not implemented the recommendations, and journals continue to publish manuscripts based on substandard research methods. Journals should not accept manuscripts that have findings that are at odds with established evidence and where the data is not publicly accessible for verification or will be in a defined time period. Importantly, editors should require documentation of funding and ethical and scientific review for the specific sodium-related analyses, especially where low-quality research methods are used.

Governments need to protect policy from low-quality research and researchers who have commercial financial interests. Requiring overt public disclosure of commercial financing to the health and scientific sector including individuals, organizations, and “shell” companies would be of great assistance. Government funders can develop investigative bodies that whistleblowers can anonymously submit concerns, for investigation.

Conclusions

In this manuscript, we discuss experiences and observations that relate to undermining the evidence that supports reductions in dietary sodium. Similar issues have been reported with other dietary health concerns and public policies and also plague science outside of the health field (e.g., where the oil industry has paid scientists to dispute climate change. https://www.bbc.com/news/stories-53640382; https://www.scientificamerican.com/article/tobacco-and-oil-industries-used-same-researchers-to-sway-public1/). Science is a never-ending quest for truth. The issues we raise are critical for society to address and to support scientific and public policy advancement and maintain public trust.

We find that the opponents of sodium reduction often use low-quality research methods and designs prone to spurious results and databases that are not accessible (or easily accessed) for independent verification; they also may fail to disclose research findings that undermine their contentions. Some attain high-level influential positions in major health and scientific organizations and have used the powers of those positions to advocate against reducing dietary sodium over the organizations stated positions supporting sodium reduction. Several have strong commercial bias, if not frank direct conflicts of interest. These observations suggest a seriously flawed system that threatens not only dietary sodium policies but also the integrity of evidence-based medicine and science itself. It is paradoxical that opponents to a population reduction in dietary sodium often ask for rigorous randomized controlled trials on dietary sodium, while often conducting and citing very low-quality research and making erroneous and misleading statements to support the need for such trials.

We hope this commentary stimulates discussion and interest not just in dietary sodium policies but also in society’s approach to protecting healthy nutrition policy and other public policies more broadly. Individual scientists and clinicians, academic institutions, funding bodies, and journals can all play a role to safeguard science and promote its advancement. Major governmental organizations (e.g., World Health Organization) need to collaborate with major ethics groups to develop, implement and enforce higher standards to safeguard public policy (https://apps.who.int/iris/handle/10665/274165, accessed March 18, 2021; https://apps.who.int/iris/bitstream/handle/10665/206554/9789241510530_eng.pdf?sequence=1, accessed March 18, 2021).