High Na intake is associated with high blood pressure and increased risk of stroke and heart disease – the leading cause of death in the USA(1). The US Dietary Guidelines recommend limiting Na intake to 2300 mg daily(2); however, average daily Na intake among US adults is currently above 3500 mg(Reference Quader, Zhao and Gillespie3).
More than 25 % of Na intake among Americans comes from food purchased at restaurants, and restaurant foods have higher Na density (milligrams of Na per 1000 calories) compared with food obtained from other sources(Reference Quader, Zhao and Gillespie3). While in recent years, some reductions in Na content have been seen in newly introduced menu items, these changes have not been consistent across menu food categories and restaurant service types(Reference Wu and Sturm4,Reference Wolfson, Moran and Jarlenski5) . Additionally, research has found a wide range of Na levels in similar restaurant food items across restaurants (e.g. vastly different Na content in a medium order of French fries in restaurant A v. restaurant B)(Reference Urban, Roberts and Fierstein6), which makes it difficult for consumers to estimate Na content without a label.
One policy strategy to assist consumers in understanding the nutrition content of restaurant food is providing nutrient information through menu labelling. Most US federal, state and local laws requiring menu labelling have focused on calories, for example, the federal menu labelling law (as part of the Affordable Care Act (ACA) in 2010) requires chain restaurants to label menu items with calories and provide other nutrition information, including Na, upon request(7). The federal menu labelling law preempted many local and state Na labelling laws requiring Na content labelling on menus in restaurants with twenty or more locations; therefore, more recently, Na warning labels have emerged as a policy focus. Some states and localities have enacted Na menu labelling laws; yet, we know of no systematic compilation of Na menu labelling laws. While several localities and states have demonstrated interest in Na menu labelling laws, only two real-world studies have been published evaluating US legislation – both studies evaluated laws enacted prior to the ACA that required menu items’ Na content to be labelled on the menu(Reference Bruemmer, Krieger and Saelens8,Reference Auchincloss, Mallya and Leonberg9) . Bummer et al. found a statistically significant decrease in entrée’s Na content comparing 6 to 18 months post-menu labelling implementation in King County, WA, and Auchincloss et al. found no significant difference in the Na content of menu items purchased comparing restaurants with Na labelling in Philadelphia, PA, to control restaurants outside of Philadelphia.
While studies have examined the impact of Na menu labelling on the Na content of restaurant foods purchased by consumers, no reviews have synthesised the evidence on Na menu labelling’s effects on the Na content of restaurant foods offered to or purchased by consumers. It is important to evaluate the Na content of menu items offered by restaurants and the Na content of menu items purchased by consumers – this allows evaluation of reformulation by restaurants and impacts on consumer behaviour, as has been evaluated in response to calorie labelling(Reference Bleich, Economos and Spiker10).
To understand policies being considered to address Na content of restaurant food, we reviewed proposed and enacted Na menu labelling laws at the local and state levels. We also reviewed and summarised the available evidence on the relationship between Na menu labelling and Na content of menu items offered by restaurants or purchased by consumers and conclude by identifying areas for future research and policy.
Methods
Local and state sodium labelling laws
We identified local- and state-level proposed and enacted Na menu labelling laws (i.e. legislation and regulation). While bills that would have required Na menu labelling have been introduced unsuccessfully at the federal level, for this review, we focused on policy activity at the state and local levels. Given this potential for local and state action, we sought to better understand the types of Na menu labelling laws that these jurisdictions are introducing and enacting.
Proposed and enacted state-level Na menu labelling laws were identified between the years 2000 and 2020 (search updated on 27 April 2020) using the LexisNexis State Capital, Nexis Uni and UConn Rudd Center Legislative databases with a combination of the keywords ‘menu’, ‘sodium’, ‘salt’ and ‘restaurant’ and verified using state governmental websites. We selected this time frame to ensure inclusion of the earliest Na menu labelling laws, which appeared in the early 2000s. We included both proposed and enacted laws to identify the types of policies that jurisdictions are considering in this area. Because bills are often introduced multiple times before they become law, proposed bills provide a window into policies that may become law in the future.
As there is no central repository of proposed and enacted local Na menu labelling laws, we searched Google using the terms ‘menu’, ‘label’, ‘policy’, ‘sodium’ and ‘restaurant’ and identified additional publicly available compilations of Na menu labelling laws collected through 2012 from the Centers for Science in the Public Interest(11), the National Conference of State Legislatures(12) and the Centers for Disease Control and Prevention’s Public Health Law Program(Reference Ramanathan13). Findings were validated through local jurisdictions’ websites.
All identified local- and state-proposed and enacted laws were reviewed for inclusion by EA, and thirty-six proposed and enacted laws were included in the final data set. Laws were included if they addressed required Na menu labelling in the restaurant setting. Laws were excluded if: (1) they did not require Na menu labelling (e.g. Na information must be available upon request) or (2) they sought to prevent a jurisdiction from engaging in Na menu labelling. For each law, the following data were abstracted: year introduced; jurisdiction; current status and primary focus. Laws were reviewed and summarised to describe the status: successfully enacted compared with not enacted; focus area: Na menu labelling within broader menu labelling or Na menu warning label; and level of government proposed: state or local.
Evaluations of sodium menu labelling
A narrative review, which comprehensively synthesises previously published literature using a narrative format(Reference Green, Johnson and Adams14), was conducted to evaluate the impact of Na menu labelling on two outcomes: Na content of menu items offered by restaurants or purchased by consumers in real-world settings in the USA. Evaluations conducted outside of the USA and online were excluded from the review. We created a search strategy to identify articles indexed in PubMed (Appendix A). The search occurred on 27 March 2019 and was updated on 27 April 2020. Twenty articles were retrieved from this search. Then, reference lists from menu labelling review articles(Reference Bleich, Economos and Spiker10,Reference Long, Tobias and Cradock15–Reference Kiszko, Martinez and Abrams20) were examined for additional articles to include. Following this search, articles were reviewed for inclusion by EA.
The review was guided by the Population, Intervention, Comparison, Outcome, Time (PICOT) framework (Appendix A). The PICOT framework helps to find the best evidence by defining key components of the research question (who, what, compared with whom, etc.), which identify study inclusion criteria(Reference Echevarria and Walker21). Two studies not set in the USA were excluded from the analysis. A flow diagram describing the literature search and filtering results is included in Appendix Fig. 1.
Fig. 1 Philadelphia Na warning label
Results
Sodium menu labelling laws
Since 2000, thirty-eight laws – eleven at the local level and twenty-seven at the state level – have been proposed to require Na labelling of restaurant menu items (Table 1). Na menu labelling laws require Na content labelled (in milligrams) or a warning label. More laws were enacted at the local level than the state level – by 2020, seven local laws were enacted to require chain restaurants to label the Na content of menu items (San Francisco City and County, CA (2008)(22); San Mateo County, CA (2008)(23); Santa Clara County, CA (2008)(24); King County, WA (2008)(25); Philadelphia, PA Code § 6-102 (2008)(26); New York, NY (2015)(27); Philadelphia, PA (2018)(28), compared with one state law (California 2008)(29) (Table 1). Thirty of the thirty-eight proposed and enacted local and state laws required Na labelling within a broader menu labelling law that also requires calorie labelling (Table 1).
Table 1 Local- and state-level proposed laws related to restaurant menu labelling for Na content
Note: ‘Adopted’ is used for law made within a government’s executive branch and ‘passed’ is used for law made within a government’s legislative branch.
* The Affordable Care Act (ACA) preempted many local and state menu labelling laws (e.g. laws applying to chains with twenty or more locations are preempted; laws applying to chains with nineteen or fewer locations can still be enforced).
Local laws
In 2003, the first local Na menu labelling bill was proposed in the District of Columbia as part of a broader menu labelling bill requiring nutrition information (specifically, calories, grams of saturated fat, grams of trans-fat, grams of carbohydrates and milligrams of Na) labelling for food and drinks. The bill was not successful, and similar bills in the District of Columbia were unsuccessfully introduced in subsequent years.
The first enacted local Na labelling laws, which required Na menu labelling within a broader menu labelling law, were introduced in 2008 in two cities and four counties: San Francisco City and County, CA (2008)(22), San Mateo County, CA (2008)(23), Santa Clara County, CA (2008)(24), Philadelphia, PA (2008)(26) and King County, WA (2008)(25) (Table 1). In 2015, the New York City Board of Health adopted regulations to require chain restaurants to include a warning label on the menu next to items containing at least 2300 mg of Na and post a warning statement on the menu(27,30) (Table 1). Most recently, in 2018, the Philadelphia mayor signed a law requiring chain restaurants to include a warning label on the menu next to items containing >2300 mg of Na and post a warning statement on the menu that defines the warning label(28,31) (Fig. 1).
State laws
Between 2003 and 2008, bills requiring Na menu labelling were proposed in California, Kentucky and New York as a component of a broader menu labelling law (Table 1). In 2007, California passed a Na menu labelling law within a broader menu labelling law(32), but the law was vetoed by the Governor. The only state-level Na menu labelling law to be enacted was signed by the California Governor in 2008(29). The law required labelling of items’ Na content and other nutrients (calories, carbohydrates and saturated fat) on menus, menu inserts, brochures or table tents.
In 2010, the federal menu labelling law was passed as part of the ACA(33), which requires calorie labelling at restaurants with twenty or more locations and preempted many local and state menu labelling laws that required labelling of nutrient data within covered establishments under the law (i.e. laws applying to chains with twenty or more locations are preempted; laws applying to chains with nineteen or fewer locations can still be enforced). At least one court has determined that a graphical Na warning label is not preempted by the federal menu labelling law because it is a warning rather than a nutrient content label (Fig. 1)(34). Following the ACA, some states, including New York and New Jersey, then began introducing stand-alone bills focused on Na menu labelling specifically. None of these bills has been passed into law.
Summary of sodium menu labelling research evidence
Five studies met inclusion criteria (Table 2)(Reference Bruemmer, Krieger and Saelens8,Reference Auchincloss, Mallya and Leonberg9,Reference Ge, Behnke and Almanza35–Reference Pulos and Leng37) . Two studies evaluated local laws in Philadelphia (2008)(26) and King County, WA (2008)(25) requiring Na menu labelling(Reference Bruemmer, Krieger and Saelens8,Reference Auchincloss, Mallya and Leonberg9) , and three studies assessed Na menu labelling interventions unrelated to state or local law(Reference Ge, Behnke and Almanza35–Reference Pulos and Leng37). Four of the five studies assessed the effects of Na menu labelling on the Na content of menu items ordered/purchased by consumers(Reference Auchincloss, Mallya and Leonberg9,Reference Ge, Behnke and Almanza35–Reference Pulos and Leng37) , and one study assessed the Na content of menu items offered by restaurants(Reference Bruemmer, Krieger and Saelens8). Articles were published between 1986 and 2014.
Table 2 Studies of the impact of Na menu labelling on Na content of menu items offered by restaurants or purchased by consumers
* Exact number(s) not published.
Evaluations of local sodium menu labelling laws on restaurant offerings
One study evaluated the impact of King County’s, WA, Na labelling law on Na content of menu items offered by restaurants(Reference Bruemmer, Krieger and Saelens8). Bruemmer et al. examined mean entrée’s Na content in menu items offered by twenty-six fast-food and eleven full-service restaurants in King County, WA, at six and 18 months post-implementation of Na menu labelling (pre-implementation entrée’s Na content was not measured). The authors found a statistically significant decrease of 108 mg of Na per entrée from 6 to 18 months post-implementation(Reference Bruemmer, Krieger and Saelens8).
Evaluations of local sodium menu labelling laws on consumer purchases
One study evaluated the impact of Philadelphia’s, PA, Na menu labelling law on Na content of items purchased by consumers(Reference Auchincloss, Mallya and Leonberg9). Auchincloss et al. compared the Na content of menu items purchased in two full-service restaurants that required Na labelling in Philadelphia, PA, to five control restaurants outside of Philadelphia not requiring Na labelling in August 2011 and found no significant difference in the Na content of menu items purchased in the labelling v. comparison groups(Reference Auchincloss, Mallya and Leonberg9). However, the investigators found that among the subgroup of customers purchasing food at Philadelphia restaurants with Na labelling who reported that nutrition information affected their order (26 % of customers), 370 mg of less Na was purchased relative to everyone else.
Voluntary sodium menu labelling interventions unrelated to state or local law
Three studies assessed the impact of voluntary Na labelling interventions on Na content of menu items purchased in settings including restaurants and cafeterias. All of the studies had an observational design. The earliest study, published in 1986, was set in a California corporate cafeteria that displayed cards labelling and comparing the Na content, as well as other nutrients, of two items(Reference Schmitz and Fielding36). The cards stated ‘HEP [health enhancement program] suggests you compare the difference’ and displayed the milligrams of Na for each item and a pie chart of maximum recommended Na intake(Reference Schmitz and Fielding36). This study by Schmitz and Fielding found that consumers purchased food with an average of 413 mg of less Na per person comparing 6 months following labelling to 6 d pre-labelling(Reference Schmitz and Fielding36). Ge et al. evaluated Na labelling interventions, in addition to other nutrient labelling interventions, at a full-service restaurant at Purdue University in Indiana(Reference Ge, Behnke and Almanza35). The researchers found no change in entrée’s Na content purchased in response to Na labelling with Na and other nutrients, comparing purchases the week of the labelling intervention to baseline(Reference Ge, Behnke and Almanza35). Pulos and Leng did not find a significant difference in the Na content of entrées sold comparing 30 d pre-labelling to 30 d post-labelling in six full-service restaurants in Pierce County, Washington(Reference Pulos and Leng37). These restaurants voluntarily labelled their menus with SmartMenu labelling, which includes labelling menu items’ Na (mg) content, as well as labelling items’ other nutrients.
Discussion
Sodium menu labelling laws
Na menu labelling policy has been implemented at the local and state levels, with seven local laws and one state law enacted of thirty-eight proposed. Thirty of the local and state Na menu labelling policies included labelling within a broader menu labelling law that included calorie labelling, and some recently proposed policies have focused on Na menu warning labels that highlight items with high Na content.
While evidence on the impact of Na menu labelling is limited(Reference Musicus, Moran and Lawman38), the policy implications from calorie menu labelling evaluations may help to inform policymakers’ decisions about Na menu labelling policy. Reviews of the effects of calorie menu labelling conclude that overall, there is limited evidence supporting a significant association between menu item’s calorie labelling and reductions in the calories of menu items ordered/purchased(Reference Long, Tobias and Cradock15,Reference Sinclair, Cooper and Mansfield16,Reference Swartz, Braxton and Viera18–Reference Kiszko, Martinez and Abrams20) , although coffee shops, full-service restaurants and cafeterias have been identified as settings in which calorie menu labelling has been associated with reduced calories purchases(Reference Bleich, Economos and Spiker10,Reference VanEpps, Roberto and Park17) . Given the lack of consistent evidence suggesting that calorie menu labelling reduces calories purchased, policymakers may want to take a different approach to Na menu labelling, such as the Na warning menu labels currently implemented in New York, NY, and Philadelphia, PA. The Na menu warning labels highlight menu items to avoid, rather than simply informing consumers of a nutrient count that requires interpretation. Additionally, while this review focused on the USA, future research is needed to review the international landscape of Na menu labelling policies. Learnings may be informative for policymakers in the USA.
Providing Na information puts the onus on the consumer to make a lower Na choice; yet, 30 % of US adults are not interested in changing their salt intake(Reference Newson, Elmadfa and Biro39). Therefore, it is important to consider policy options that reduce the Na content of restaurant items. For example, in 2010, New York State proposed a prohibition on the use of salt in food preparations by restaurants(40). Lessons from the trans-fat policy ban may also prove instructive. Before the FDA banned trans-fats in foods in 2016 (with implementation in 2018), trans-fat was banned in New York City effective in 2008 (New York Health Code Section 81.08 [2006]). Evaluation of the New York City regulation found a significant trans-fat reduction of 2·4 g per fast-food purchase following implementation, compared with before the regulation(Reference Angell, Cobb and Curtis41). This suggests that laws requiring restaurants to reduce Na content of menu items may be an effective strategy for reducing the amount of Na consumed by customers. However, unintended consequences may be associated with reduced Na consumption, if consumers or restaurants replace Na with other unhealthy nutrients such as fat and sugar; therefore, changes in nutrients of menu items offered should be monitored to identify unhealthy changes.
Voluntary reductions in Na content of menu items is an opportunity for the food industry. The US FDA has released guidance for voluntary Na reduction to encourage restaurants and manufacturers to reduce Na in foods(42). Several examples highlight successful Na reduction at the community level(Reference Riordan, Zeitz and Fulton43,Reference Long, Rowland and Langston44) . Additionally, several restaurants have committed to voluntarily reduce Na in menu items as part of the National Salt and Sugar Reduction Initiative, a programme convened by the New York City Health Department(45). An evaluation of the National Salt Reduction Initiative found modest progress by packaged food companies(Reference Curtis, Clapp and Niederman46), although progress among restaurants is not well understood. Restaurants can also introduce new menu items with reduced Na, remove items with high Na, reformulate existing menu items to reduce Na or reduce portion sizes, but voluntary initiatives cannot be enforced and may be difficult to evaluate.
Sodium menu labelling research evidence
It is important to understand the impact of Na menu labelling on changes in both the Na content of menu items purchased by consumers and the Na content of menu items offered by restaurants. The shift from Na numeric labelling laws to Na warning laws in recent years highlights the impact of preemption on local and state policy. At the local level, the New York, NY (2015)(27) and Philadelphia, PA (2018)(28) laws requiring Na warning labels have not been evaluated, and Na warning labels have also been proposed in New York State and New Jersey. Future research is needed to understand the impact of these laws on the Na content of available and purchased restaurant menu items, as this evidence could inform policymakers about the effectiveness of such policy.
Research has shown that adding calorie and Na information may lead consumers to order items with more or less Na, depending on taste preferences(Reference Byrd, Almanza and Ghiselli47). Na menu labelling may change consumer behaviour, encouraging consumers to use the Na menu label(Reference Auchincloss, Young and Davis48) and increase purchases of menu items with reduced Na, which may contribute to reduced Na intake. Further, Na menu labelling may encourage restaurants to reduce Na content in menu items. Of the five studies evaluating Na menu labelling, only one assesses the Na content of menu items offered as an outcome(Reference Bruemmer, Krieger and Saelens8). This study found a significant decrease in entrée’s Na content comparing 6- and 18-month post-menu labelling implementation. Additional studies are needed to understand whether Na menu labelling has an effect on the Na content of menu items offered. Additionally, qualitative research that evaluates the barriers restaurants face in reducing Na in menu items (e.g. lack of lower Na ingredients from suppliers and limited research and development(R&D) budgets for Na reduction) may inform researchers, public health practitioners and other stakeholders of resources that could support restaurants’ efforts to reduce Na content of menu items.
Considering Na menu labelling impact on consumer purchases, currently only four US-based real-world studies have been published evaluating the effects of Na menu labelling on Na content of menu items purchased by consumers in food service settings and the results are mixed(Reference Auchincloss, Mallya and Leonberg9,Reference Ge, Behnke and Almanza35–Reference Pulos and Leng37) . Online experiments have been conducted to evaluate consumer response to Na menu labelling(Reference Musicus, Moran and Lawman38,Reference Byrd, Almanza and Ghiselli47,Reference Hobin, Lillico and Zuo49) , with significant results in some interventions – for example, compared with control, traffic light and red stop sign warning labels significantly reduced Na content of menu items ordered. Qualitative research focused on consumers is needed to explore how they view and understand Na menu labelling and consumers’ perceptions of whether and how Na menu labelling does or does not influence their purchasing decisions.
Evaluation of trends in the Na content of restaurant menu items has found that calorie-adjusted Na content in new restaurant menu items declined from 2012 to 2016, but reductions are inconsistent across menu categories and Na content of menu items remains high(Reference Wolfson, Moran and Jarlenski5). Given that average daily Na intake among US adults is over 1000 mg more than recommended, additional reductions in Na intake may be needed to reach clinical significance. Further, the accuracy of Na labelling may be an issue; research evaluating the accuracy of nutrient labelling in university food service settings found significant differences in the Na content, among other nutrients, of sampled items(Reference Feldman, Murray and Chavarria50).
Additional studies with strong designs are needed to evaluate Na menu labelling. All of the studies evaluated in this analysis had an observational design. Three used single-group, pre-/post-study designs, measuring outcomes before and after a Na menu labelling intervention in the setting(s) that implemented the intervention(Reference Ge, Behnke and Almanza35–Reference Pulos and Leng37). One study measured post-Na menu labelling outcomes in an intervention v. a comparison group but did not measure pre-intervention baseline outcomes and so was unable to assess differences in the changes in Na content of menu items in the intervention group relative to the comparison group(Reference Auchincloss, Mallya and Leonberg9). Finally, one study used a single-group, post-test-only design, where outcomes were measured at two time points after Na menu labelling was implemented(Reference Bruemmer, Krieger and Saelens8). While a randomised controlled trial design is not feasible to evaluate laws, as laws cannot be applied to randomised groups, evaluations using rigorous quasi-experimental approaches, such as difference-in-differences or synthetic control analyses, to evaluate changes in the Na content of menu items offered or purchased before and after Na menu labelling laws/interventions, relative to a control group, are needed. Further, a systematic review of the literature is needed to evaluate the existing evidence on Na menu labelling, including critical review of the study methodology.
Limitations
Given there is no central repository of local laws, it is possible that not all local laws were identified through the search. Studies evaluating Na menu labelling have also been conducted in international settings and may provide useful insights for US policy. For example, Vanderlee and Hammond compared Na consumed between participants at a hospital cafeteria in Canada with a Hospital Check programme that labelled nutrients including Na at the point of purchase, with a control hospital that provided limited nutrition information for a selection of items(Reference Vanderlee and Hammond51). Participants at the hospital with the Hospital Check programme reported consuming fewer 314 mg of Na than participants at the control cafeteria. The goal of this paper is to understand the different Na menu labelling approaches that have been considered by policymakers, rather than to understand the current legal environment. Identifying whether the laws were enacted is beyond the scope of this paper. It is possible that some of the laws identified were subsequently nullified due to preemption. The functional properties of Na in food science (e.g. texture, structure and food safety) were not discussed as they are outside the scope of this article.
Conclusion
Na -focused local and state menu labelling laws have been increasingly proposed and enacted since 2003. Understanding the impact of Na menu labelling is important to inform policymakers about whether this approach is effective. The current evidence evaluating Na menu labelling in the USA shows mixed results, with reductions in Na content of menu items offered by restaurants or purchased by consumers seen in some settings. More research is needed to evaluate the impact of Na menu labelling on Na content of menu items and purchasing behaviour. Policymakers should also consider laws that limit the Na content of restaurant foods.
Acknowledgements
Acknowledgements: N/A. Financial support: This research received no specific grant from any funding agency, commercial or not-for-profit sectors. Conflicts of interest: None. Authorship: All authors contributed to the design of the study. E.A. led the review of laws, the narrative review of evidence and manuscript writing. E.E.M., L.R., K.G. and J.E.C. contributed revisions to the manuscript and approved the final manuscript for submission. Ethics of human subject participation: This study was deemed Exempt by the Johns Hopkins Bloomberg School of Public Health Institutional Review Board.
Supplementary material
For supplementary material accompanying this paper visit https://doi.org/10.1017/S1368980020003961
