Abstract
Background
Experiments have shown that added-sugar warning labels on restaurant menus reduce the added-sugar content of meal selections. Consequently, New York City mandated the first added-sugar warning labels on chain restaurant menus, and other United States jurisdictions are pursuing similar policies. These policies could be especially beneficial for priority populations at higher risk of diet-related diseases. However, little is known about public support for restaurant added-sugar warning label policies among the populations that might benefit the most from the labels.
Objectives
This research aimed to assess support by demographic and health-related characteristics for an added-sugar menu-labeling policy.
Methods
We assessed support for a policy requiring added-sugar warning labels on restaurant menus among 15,275 United States adults in an online cross-sectional convenience sample recruited to match national sociodemographic distributions. We described and examined differences in support by participant characteristics using bivariate and multivariable regressions (multivariable results described below).
Results
Across all characteristics examined, there was majority support (range: 54%–87%) for a potential added-sugar warning label policy for chain restaurant menus, followed by no opinion (range: 8%–32%) with few opposing (range: 0%–19%). There was significantly higher support among those with prediabetes (82% compared with 71% among those without prediabetes) and those trying to reduce their added-sugar consumption (81% compared with 54%; all P < 0.05). In contrast, individuals with lower compared with higher income (66% compared with 79%) and lower compared with higher educational attainment (67% compared with 78%) showed lower levels of support (all P < 0.05). Some groups had large shares of “no opinion” responses, such as those not trying to reduce added-sugar intake (32%).
Conclusions
Across all subpopulations examined, majority supported an added-sugar warning label policy. Support was higher among those with prediabetes and those trying to reduce their added-sugar consumption. Added-sugar restaurant menu-warning labels are widely supported across groups defined by demographic and health-related characteristics.
Keywords: nutrition labeling, food labeling, policy support, sociodemographic differences, key populations
Introduction
With >60% of United States adults exceeding the 2020–2025 Dietary Guidelines for Americans recommended limit of 10% of calories from added sugar [1] and more than half of adults in the United States having prediabetes (44%) or type 2 diabetes (14%) [2], there is a clear need for interventions that reduce consumption of added sugar. Restaurants are a promising avenue for addressing added-sugar overconsumption as they are a major source of foods and beverages for United States adults [3]; restaurant food typically has low nutritional quality with high amounts of added sugar [[4], [5], [6]]; and frequent restaurant food consumption is a risk factor for developing diet-related chronic conditions [[7], [8], [9], [10], [11], [12], [13], [14], [15]]. Additionally, restaurant menus rarely display nutritional information beyond calories (the disclosure of which is mandated by federal regulations [16]), potentially hindering consumers from making healthier food choices. Added-sugar warning labels on restaurant menus (i.e., labels indicating that a menu item contains a high amount of added sugar) are a promising policy intervention for increasing consumer understanding of the nutritional content of restaurant foods [[17], [18], [19], [20]], reducing the amount of added sugar ordered [[18], [19], [20]], correcting misperceptions about sugary items [21,22], and potentially encouraging restaurants to reduce the added-sugar content of menu items [23,24].
In the United States, 2 cities have implemented warning labels on restaurant menus. New York City (NYC) and Philadelphia implemented policies (NYC in 2016, Philadelphia in 2019) requiring sodium warning labels next to menu items containing ≥100% of the daily recommended sodium limit (2300 mg) [25,26]. NYC now additionally requires an added-sugar warning label for items with ≥100% of the daily recommended added-sugar limit (50 g) [27,28]. Other menu-labeling policy initiatives, which include both added sugar and sodium labels, have been proposed in multiple United States jurisdictions at the local [29,30] and state [[31], [32], [33], [34], [35]] levels.
Online randomized experiments demonstrate that added-sugar warning labels provide easy-to-understand nutrition information and promote healthier food choices, which may improve population-level dietary patterns [[17], [18], [19]]. Specifically, these experiments have found that added-sugar menu labels are perceived as effective [17], improve added-sugar-content knowledge, [17,19] and, in hypothetical ordering tasks, reduce both the risk of selecting high-added-sugar items and the total amount of added sugar in selected items [18,19]. Importantly, online experiments have found that added-sugar warning labels are consistently effective at improving knowledge and encouraging healthier choices across diverse groups [[18], [19], [20],36], suggesting that these labels do not exacerbate health disparities likely because they do not require high numeracy or health literacy to understand and use. With evidence that well-designed labels are effective at improving consumer understanding and promoting healthier choices, assessing public support becomes an important next step for advancing policy.
Polls indicate bipartisan support for sugary drink and added-sugar warning label policies among the general population [[37], [38], [39]], and we had previously observed 80% [17] and 72% [18] overall support for an added-sugar warning label policy, but there is a lack of peer-reviewed research on the levels of support among key populations such as those with prediabetes, type 2 diabetes, and populations disproportionately affected by diet-related chronic diseases [40,41] (e.g., low-income populations, Hispanic, and non-Hispanic Black populations). Understanding support among these groups is particularly important because they may especially benefit from added-sugar warning labels [42]. Examining support levels is also key for policy engagement strategies; depending on the context, low support might indicate an opportunity for education and outreach, whereas high support might suggest an opportunity for coalition building. With added-sugar warning label policies proposed across the country and the growing evidence of their efficacy, there is a need to understand public support for this type of policy. Thus, this study aimed to assess support for a potential added-sugar restaurant menu-warning label policy across key characteristics in a large national sample of United States adults.
Methods
Recruitment
Using the data platform, Dynata panels, a national cross-sectional convenience sample of 26,752 English-speaking United States adults were recruited for a parent study [18]. Quota-matching was used to obtain a sample that represented approximate American Community Survey 2018 5-y estimates [43] for age (18–34 y, 35–54 y, and ≥55 y), sex (male or female, estimated by self-reported gender), race/ethnicity (Hispanic any race, non-Hispanic of the following: American Indian or Alaskan Native, Asian, Black, Pacific Islander or Native Hawaiian, White, or Multiracial), and education [up to high school diploma or General Educational Development (GED), some college or Associate’s degree, or Bachelor’s degree or higher]. After participants were provided with an electronic consent form and indicated informed electronic consent by checking a box, they were assessed for eligibility, which included: currently living in the United States, being 18 to 99 y old, having purchased food from a restaurant at least once per month, and passing a bot-detection task (CAPTCHA). Participants were not explicitly told the aim of the research but were informed they would be asked to “view restaurant menus and provide your opinions about them.” Participants were compensated for their time (∼United States $2.25–$2.50) by the panel company as part of the parent study [18]. Data were collected May–June 2021. All aspects of this study were approved by the University of California, Davis Institutional Review Board (#1641776).
Procedures and measures
Prior to indicating levels of policy support, as part of the parent study, [18] participants were randomly assigned to view menus with either no label (control) or an upside-down triangle icon with an exclamation mark over a spoon next to menu items containing >50% of the daily recommended limit (25 g) for added sugar. Participants then placed hypothetical orders from a fast-food and full-service menu and responded to other items (e.g., knowledge of items’ added-sugar content, perceptions of label). This label design was selected from a collection of icons that were all perceived as significantly more effective than a control label [17] and was the option most similar to NYC’s label design. The label used herein differs from NYC’s warning label design (NYC uses a pointed-up triangle and no exclamation point) and was applied to items using a lower added-sugar threshold (NYC labels items with ≥100% of the daily recommended limit for added sugar).
Then, to assess support for a theoretical added-sugar warning label policy, all participants (regardless of assigned condition) were shown the added-sugar warning icon used in the intervention group in isolation and on a menu excerpt (Supplemental Figure 1). While viewing the icon and menu excerpt, participants were asked, “Some cities are considering a law that requires that chain restaurants display this label next to items that are high in added sugars. Would you...?” Response options were, “Strongly support this law,” “Somewhat support this law,” “No opinion,” “Somewhat oppose this law,” and “Strongly oppose this law.”
Age, gender, race/ethnicity, education, and state of residence were assessed during eligibility screening to ensure participants met eligibility criteria and to monitor demographic quotas; other demographics were assessed at the end of the survey. Additional demographics included household income before taxes, dietary restrictions, diet-related chronic disease diagnoses, intentions to reduce added-sugar intake, and frequency of consuming restaurant foods for both fast-food and full-service venues. These participant characteristics were selected because they represent important behaviors and risk factors for diet-related chronic diseases [[7], [8], [9], [10], [11]] as well as populations that disproportionately experience higher burdens of these diseases [40,41]. Exact item wording can be found in Supplemental Table 1.
The policy support measure and the measure assessing “trying to reduce added-sugar intake” were novel items developed by the research team and adapted from similar items we have used previously [17].
Policy support was examined by these demographic variables to understand the extent to which policy perceptions varied across population groups, especially those living with diet-related conditions and those disproportionately affected by unhealthy food environments, food marketing, and diet-related chronic diseases, such as prediabetes and diabetes [[44], [45], [46], [47]].
Statistical analysis
Support for an added-sugar warning label policy was dichotomized as “support” (strongly or somewhat support) and “no opinion/oppose” (no opinion, somewhat or strongly oppose) to focus on distinguishing support from all other responses. Exact item wording, response options, and how responses were categorized are shown in Supplemental Table 1.
Differences in policy support by individual characteristics (e.g., income group) were examined using Poisson regression with robust SEs to directly estimate probability ratios (PRs) [48]. Support was regressed on each characteristic in separate bivariate models as well as on all characteristics in a single multivariable model. We used bivariate regressions to identify potential differences in support by each characteristic, and we used the multivariable regression to examine the overall relationship between participant characteristics and support, accounting for relationships among predictors. In addition to participant characteristics, we also adjusted for assigned condition in the parent study in the multivariable model, as there was a small but significant difference in support by assigned condition in the parent study. Those in the intervention group (i.e., saw the label in a menu ordering task) were slightly less likely to support the policy (by 2 percentage points) compared with the control group. However, due to randomization, assigned condition was not associated with any participant characteristic. All models used complete case analysis. We report percentages, PRs from the multivariable model, and note when there are differences between the bivariate and multivariable models. In addition to regression results, we present the proportions of all 5 response options in Table 1 [18] and report the proportions of “no opinion” and oppose in the results. Post hoc pairwise comparisons between all predictor levels in the multivariable model were also conducted. We note significant differences in the results but report all comparisons in Supplemental Table 2. All tests (2-sided α = 0.05) were conducted using Stata/MPv16.1 (StataCorp LLC).
TABLE 1.
Predictors of support for restaurant menu added-sugar warning labels
| n | Strongly support |
Somewhat support |
No opinion | Somewhat oppose |
Strongly oppose |
Support vs. no opinion/oppose |
||||
|---|---|---|---|---|---|---|---|---|---|---|
| Multivariable (n = 14,971) |
Bivariate (n ≥ 15,081) |
|||||||||
| PR (95% CI) | P | PR (95% CI) | P | |||||||
| Overall1 | 15,275 | 5955 (39%) | 5002 (33%) | 2888 (19%) | 836 (5%) | 594 (4%) | —— | — | — | — |
| Diagnosed with prediabetes | ||||||||||
| Not diagnosed | 13,758 | 5180 (38%) | 4543 (33%) | 2715 (20%) | 771 (6%) | 549 (4%) | ref | — | ref | — |
| Diagnosed | 1392 | 726 (52%) | 416 (30%) | 147 (11%) | 60 (4%) | 43 (3%) | 1.04 (1.01, 1.072) | 0.0032 | 1.16 (1.13, 1.19) 2 | <0.001 |
| Diagnosed with type 2 diabetes | ||||||||||
| Not diagnosed | 13,158 | 4970 (38%) | 4324 (33%) | 2597 (20%) | 747 (6%) | 520 (4%) | ref | — | ref | — |
| Diagnosed | 1992 | 936 (47%) | 635 (32%) | 265 (13%) | 84 (4%) | 72 (4%) | 1.00 (0.98, 1.03) | 0.712 | 1.12 (1.09, 1.15) 2 | <0.001 |
| Diagnosed with obesity | ||||||||||
| Not diagnosed | 13,050 | 4965 (38%) | 4310 (33%) | 2511 (19%) | 736 (6%) | 528 (4%) | ref | — | ref | — |
| Diagnosed | 2100 | 941 (45%) | 649 (31%) | 351 (17%) | 95 (5%) | 64 (3%) | 1.02 (0.99, 1.05) | 0.180 | 1.07 (1.04, 1.09) 2 | <0.001 |
| Trying to reduce added-sugar intake | ||||||||||
| Not trying to reduce | 5030 | 1106 (22%) | 1585 (32%) | 1603 (32%) | 391 (8%) | 345 (7%) | ref | — | ref | — |
| Trying to reduce | 10,198 | 4830 (47%) | 3398 (33%) | 1276 (13%) | 445 (4%) | 249 (2%) | 1.46 (1.42, 1.50) 2 | <0.0012 | 1.51 (1.47, 1.55) 2 | <0.001 |
| Reported dietary restrictions | ||||||||||
| No restrictions | 11,902 | 4156 (35%) | 4049 (34%) | 2516 (21%) | 688 (6%) | 493 (4%) | ref | — | ref | — |
| At least 1 dietary restriction | 3179 | 1720 (54%) | 891 (28%) | 334 (11%) | 139 (4%) | 95 (3%) | 1.08 (1.06, 1.11) 2 | <0.0012 | 1.19 (1.17, 1.22) 2 | <0.001 |
| Fast-food ordering frequency | ||||||||||
| Low ordering frequency | 5130 | 2028 (40%) | 1700 (33%) | 950 (19%) | 269 (5%) | 183 (4%) | ref | — | ref | — |
| Moderate ordering frequency | 6505 | 2242 (34%) | 2216 (34%) | 1398 (21%) | 398 (6%) | 251 (4%) | 0.96 (0.94, 0.99) 2 | 0.0042 | 0.94 (0.92, 0.97) 2 | <0.001 |
| High ordering frequency | 3631 | 1681 (46%) | 1084 (30%) | 537 (15%) | 169 (5%) | 160 (4%) | 1.02 (0.99, 1.06) | 0.156 | 1.05 (1.02, 1.07) 2 | <0.001 |
| Full-service ordering frequency | ||||||||||
| Low ordering frequency | 7792 | 3032 (39%) | 2555 (33%) | 1513 (19%) | 416 (5%) | 276 (4%) | ref | — | ref | — |
| Moderate ordering frequency | 5248 | 1781 (34%) | 1798 (34%) | 1108 (21%) | 335 (6%) | 226 (4%) | 0.94 (0.92, 0.97) 2 | <0.0012 | 0.95 (0.93, 0.97) 2 | <0.001 |
| High ordering frequency | 2226 | 1138 (51%) | 647 (29%) | 264 (12%) | 85 (4%) | 92 (4%) | 1.02 (0.98, 1.05) | 0.374 | 1.12 (1.09, 1.15) 2 | <0.001 |
| Age (y) | ||||||||||
| 18–34 | 4503 | 1788 (40%) | 1435 (32%) | 867 (19%) | 248 (6%) | 165 (4%) | 1.02 (0.99, 1.05) | 0.237 | 1.02 (0.99, 1.04) | 0.199 |
| 35–54 | 4688 | 2038 (43%) | 1411 (30%) | 872 (19%) | 197 (4%) | 170 (4%) | 1.02 (0.99, 1.04) | 0.270 | 1.04 (1.02, 1.07) 2 | <0.001 |
| 55+ | 6084 | 2129 (35%) | 2156 (35%) | 1149 (19%) | 391 (6%) | 259 (4%) | ref | — | ref | — |
|
N |
Strongly support |
Somewhat support |
No opinion |
Somewhat oppose |
Strongly oppose |
Support vs. no opinion/oppose | ||||
| Multivariable (n = 14,971) |
Bivariate (n ≥ 15,081) |
|||||||||
| PR (95% CI) |
P |
PR (95% CI) |
p |
|||||||
| Gender3 | ||||||||||
| Man | 7010 | 2635 (38%) | 2298 (33%) | 1321 (19%) | 420 (6%) | 336 (5%) | ref | — | ref | — |
| Woman | 8189 | 3285 (40%) | 2685 (33%) | 1552 (19%) | 412 (5%) | 255 (3%) | 1.07 (1.05, 1.09) 2 | <0.0012 | 1.04 (1.02, 1.06) 2 | 0.001 |
| Nonbinary/gender nonconforming | 76 | 35 (46%) | 19 (25%) | 15 (20%) | 4 (5%) | 3 (4%) | 1.04 (0.91, 1.20) | 0.554 | 1.01 (0.87, 1.17) | 0.896 |
| Race and ethnicity4 | ||||||||||
| Hispanic any race | 2379 | 1036 (44%) | 746 (31%) | 402 (17%) | 118 (5%) | 77 (3%) | 1.02 (1.00, 1.05) | 0.081 | 1.05 (1.02, 1.07) 2 | 0.001 |
| NH American Indian or Alaskan Native alone | 96 | 36 (38%) | 23 (24%) | 19 (20%) | 11 (11%) | 7 (7%) | 0.83 (0.71, 0.97) | 0.0162 | 0.86 (0.73, 1.01) | 0.059 |
| NH Asian alone | 919 | 324 (35%) | 356 (39%) | 185 (20%) | 40 (4%) | 14 (2%) | 1.00 (0.96, 1.05) | 0.885 | 1.03 (0.99, 1.08) | 0.113 |
| NH Black alone | 2061 | 877 (43%) | 527 (26%) | 449 (22%) | 112 (5%) | 96 (5%) | 0.95 (0.92, 0.98) 2 | 0.0032 | 0.95 (0.92, 0.98) 2 | 0.002 |
| NH Multiracial | 24 | 72 (40%) | 52 (29%) | 38 (21%) | 7 (4%) | 9 (5%) | 1.00 (0.91, 1.11) | 0.987 | 0.97 (0.88, 1.07) | 0.579 |
| NH Pacific Islander or native Hawaiian alone | 9618 | 12 (50%) | 8 (33%) | 2 (8%) | 1 (4%) | 1 (4%) | 1.15 (0.96, 1.38) | 0.121 | 1.16 (0.97, 1.39) | 0.098 |
| NH White alone5 | 178 | 3598 (37%) | 3290 (34%) | 1793 (19%) | 547 (6%) | 390 (4%) | ref | — | ref | — |
| Education | ||||||||||
| Up to high school diploma or GED | 5626 | 2047 (36%) | 1723 (31%) | 1281 (23%) | 303 (5%) | 272 (5%) | 0.92 (0.90, 0.95) 2 | <0.0012 | 0.86 (0.84, 0.88) 2 | <0.001 |
| Some college or associate’s degree | 4937 | 1844 (37%) | 1666 (34%) | 949 (19%) | 298 (6%) | 180 (4%) | 0.96 (0.94, 0.98) 2 | 0.0012 | 0.91 (0.89, 0.93) 2 | <0.001 |
| Bachelor's degree or higher | 4712 | 2064 (44%) | 1613 (34%) | 658 (14%) | 235 (5%) | 142 (3%) | ref | — | ref | — |
| Region | ||||||||||
| West | 3144 | 1273 (40%) | 1015 (32%) | 573 (18%) | 162 (5%) | 121 (4%) | 1.00 (0.97, 1.02) | 0.865 | 1.03 (1.00, 1.06) 2 | 0.044 |
| Midwest | 3128 | 1138 (36%) | 1056 (34%) | 640 (20%) | 179 (6%) | 115 (4%) | 1.01 (0.98, 1.03) | 0.630 | 0.99 (0.96, 1.02) | 0.520 |
| Northeast | 3014 | 1208 (40%) | 1025 (34%) | 522 (17%) | 154 (5%) | 105 (3%) | 1.03 (1.01, 1.06) 2 | 0.0162 | 1.05 (1.02, 1.07) 2 | 0.001 |
| South | 5974 | 2329 (39%) | 1900 (32%) | 1151 (19%) | 341 (6%) | 253 (4%) | ref | — | ref | — |
| US Territory | 15 | 7 (47%) | 6 (40%) | 2 (13%) | 0 (0%) | 0 (0%) | 1.17 (0.94, 1.46) | 0.159 | 1.22 (1.00, 1.49) 2 | 0.046 |
| Income | ||||||||||
| ≤$35,000/y | 4106 | 1467 (36%) | 1247 (30%) | 980 (24%) | 228 (6%) | 184 (4%) | 0.91 (0.88, 0.94) 2 | <0.0012 | 0.84 (0.82, 0.87) 2 | <0.001 |
| $35,001–$65,000 | 4052 | 1513 (37%) | 1337 (33%) | 807 (20%) | 238 (6%) | 157 (4%) | 0.95 (0.92, 0.98) 2 | <0.0012 | 0.90 (0.88, 0.92) 2 | <0.001 |
| $65,001–$95,000 | 2802 | 1012 (36%) | 1014 (36%) | 503 (18%) | 161 (6%) | 112 (4%) | 0.96 (0.94, 0.99) 2 | 0.0102 | 0.92 (0.90, 0.95) 2 | <0.001 |
| $>95,000 | 4227 | 1932 (46%) | 1375 (33%) | 576 (14%) | 206 (5%) | 138 (3%) | ref | — | ref | — |
| Assigned condition6 | ||||||||||
| Control | 7663 | 2963 (39%) | 2615 (34%) | 1293 (17%) | 488 (6%) | 304 (4%) | ref | — | ref | — |
| Intervention | 7612 | 2992 (39%) | 2387 (31%) | 1595 (21%) | 348 (5%) | 290 (4%) | 0.97 (0.95, 0.99) | 0.002 | 0.97 (0.95, 0.99) | 0.004 |
Missing data were not included in the denominator for calculating percentages.
Abbreviation: NH, non-Hispanic.
The parent study [18] previously reported overall support results. Support by participant characteristics was not previously reported.
Values are significant.
Gender was assessed and served as a proxy for ensuring a similar distribution of sex to American Community Survey estimates.
Race and ethnicity were combined because of the high proportion of Hispanic residents in the United States who identify as Hispanic or Latino/a/x for both their ethnic and racial background (https://pewrsr.ch/3rQUlj9).
Includes Middle Eastern and North African for consistency with United States census.
Assigned condition of parent study [18].
Analytic sample
After screening, 10,044 participants were deemed ineligible (373 did not provide age; 268 were aged <18 or >99 y; 5804 reported purchasing from restaurants <1 time/mo; 3599 were screened out due to full quotas or failing CAPTCHA). A total of 16,708 participants were eligible and proceeded to the parent study [18]. An additional 1433 responses were excluded from the analytic sample: 795 did not provide a response to the policy support question, 96 participants were excluded for completing the survey too quickly (i.e., <30% median completion time), 536 were excluded for failing the attention check (i.e., incorrectly answering a question that asked what the current month was), and 6 were outside the United States. These exclusions resulted in a final analytic sample of 15,275 participants (Supplemental Figure 2).
Results
Participant characteristics can be found in Table 2, and national distributions of these characteristics can be found in Supplemental Table 3. Nine percent of participants reported a prediabetes diagnosis, 13% reported a type 2 diabetes diagnosis, 14% reported an obesity diagnosis, 67% reported trying to reduce their added-sugar consumption, and 21% reported at least one dietary restriction. One-in-four participants (24%) reported high fast-food ordering frequency (ordering a few times a week or ordering every day), and 15% reported high full-service ordering frequency (ordering a few times a week or ordering every day). The distributions of quota-matched characteristics (age, gender, race and ethnicity, and education) were similar to national estimates (Supplemental Table 3). The largest departure (+4 percentage points) was for those aged ≥55 y (sample = 40%, national = 36%). We had a larger proportion of women (sample = 54%, national = 51%) and a lower proportion (˗3 percentage points) of those with up to a high school diploma or GED (sample = 37%, national = 40%) compared with national distributions. All other sample proportions for age, race and ethnicity, and education were within 2 percentage points of national estimates. Half of the participants (50%) were assigned to the intervention in the parent study.
TABLE 2.
Participant characteristics
| Characteristics | n (%) |
|---|---|
| Diagnosed with prediabetes | |
| Not diagnosed | 13,758 (91%) |
| Diagnosed | 1392 (9%) |
| Diagnosed with type 2 diabetes | |
| Not diagnosed | 13,158 (87%) |
| Diagnosed | 1992 (13%) |
| Diagnosed with obesity | |
| Not diagnosed | 13,050 (86%) |
| Diagnosed | 2100 (14%) |
| Trying to reduce added-sugar intake | |
| Not trying to reduce | 5030 (33%) |
| Trying to reduce | 10,198 (67%) |
| Reported dietary restrictions | |
| No restrictions | 11,902 (79%) |
| At least 1 dietary restriction | 3179 (21%) |
| Fast-food ordering frequency | |
| Low ordering frequency | 5130 (34%) |
| Moderate ordering frequency | 6505 (43%) |
| High ordering frequency | 3631 (24%) |
| Full-service ordering frequency | |
| Low ordering frequency | 7792 (51%) |
| Moderate ordering frequency | 5248 (34%) |
| High ordering frequency | 2226 (15%) |
| Age (y) | |
| 18–34 | 4503 (29%) |
| 35–54 | 4688 (31%) |
| 55+ | 6084 (40%) |
| Gender1 | |
| Man | 7010 (46%) |
| Woman | 8189 (54%) |
| Nonbinary/gender nonconforming | 76 (1%) |
| Race and ethnicity2 | |
| Hispanic any race | 2379 (16%) |
| NH American Indian or Alaskan Native alone | 96 (1%) |
| NH Asian alone | 919 (6%) |
| NH Black alone | 2061 (13%) |
| NH Pacific Islander or native Hawaiian alone | 24 (<1%) |
| NH White alone3 | 9618 (63%) |
| NH Multiracial | 178 (1%) |
| Education | |
| Up to high school diploma or GED | 5626 (37%) |
| Some college or associate's degree | 4937 (32%) |
| Bachelor's degree or higher | 4712 (31%) |
| Income | |
| ≤$35,000/y | 4106 (27%) |
| $35,001–$65,000 | 4052 (27%) |
| $65,001–$95,000 | 2802 (18%) |
| $>95,000 | 4227 (28%) |
| Region | |
| West | 3144 (21%) |
| Midwest | 3128 (20%) |
| Northeast | 3014 (20%) |
| South | 5974 (39%) |
| United States territory | 15 (<1%) |
| Assigned condition4 | |
| Control | 7663 (50%) |
| Intervention | 7612 (50%) |
Missing data were not included in the denominator for calculating percentages.
Abbreviation: NH, non-Hispanic.
Gender was assessed and served as a proxy for ensuring a similar distribution of sex to American Community Survey estimates.
Race and ethnicity were combined due to the high proportion of Hispanic residents in the United States who identify as Hispanic or Latino/a/x for both their ethnic and racial background (https://pewrsr.ch/3rQUlj9)
Includes Middle Eastern and North African for consistency with United States census.
Assigned condition of parent study [18].
As previously reported [18], the majority of participants (72%) overall supported an added-sugar warning label policy, with 39% strongly supporting and 33% somewhat supporting the policy. Across all subgroups of participants (Table 1), there was majority support for the added-sugar warning label policy (54%–87%). Nineteen percent of participants responded, “no opinion,” ranging from 8% to 32% by subgroup. Overall opposition was 9% (5% somewhat oppose, 4% strongly oppose), ranging from 0% to 19% by subgroup.
Regression results are shown in Table 1. For diet- and health-related variables, there was a significantly higher likelihood of support among those with a prediabetes diagnosis than those without prediabetes (82% compared with 71%; PR = 1.04; 95% CI: 1.01, 1.07; P = 0.003), among those with a dietary restriction (83% compared with 69%; PR = 1.08; 95% CI: 1.06, 1.11; P < 0.001), and among those trying to reduce their added-sugar consumption (81% compared with 54%; PR = 1.46; 95% CI: 1.42, 1.50; P < 0.001) in multivariable and bivariate models. Type 2 diabetes diagnosis and obesity diagnosis were only significant predictors of higher support in the bivariate model. The groups with higher support also exhibited higher levels of “strongly support” (e.g., 52% of those with prediabetes “strongly supported” compared with 38% without). Across these characteristics, opposition ranged from 7% to 15%, and “no opinion” ranged from 11% to 32%.
Compared with the lowest frequency of ordering restaurant food, those with moderate ordering frequency for both fast-food (73% compared with 69%; PR = 0.96; 95% CI: 0.94, 0.99; P = 0.004) and full-service restaurants (72% compared with 68%; PR = 0.94; 95% CI: 0.92, 0.97; P < 0.001) were less likely to support the policy. In the multivariable model, there was no difference in support between the highest and lowest ordering frequency groups for both fast-food (73%) and full-service (72%) ordering (all P > 0.05). However, in the bivariate model, the highest ordering frequency groups for both fast-food and full-service ordering were significantly more likely to support the policy than the lowest ordering frequency groups. In post hoc comparisons, the highest ordering frequency groups had higher support compared with the moderate ordering frequency group for both restaurant types (all P < 0.001). Across restaurant type and ordering frequency, opposition ranged from 8% to 10%, and “no opinion” ranged from 12% to 21%.
There was no difference in support by age in the multivariable model (all P > 0.05); however, in the bivariate model, participants aged 35 to 54 y were significantly more likely to support the policy than participants aged 55+ y. Compared with men, women were more likely to support the policy in both the bivariate and multivariable models (73% compared with 70%; PR = 1.07; 95% CI: 1.05, 1.09; P < 0.001). Across age and gender categories, opposition ranged from 8% to 11%, and “no opinion” ranged from 19% to 20%.
Non-Hispanic Black (68% compared with 72%; PR = 0.95; 95% CI: 0.92, 0.98) and non-Hispanic American Indian or Alaskan Native (61% compared with 72%; PR = 0.83; 95% CI: 0.71, 0.97) participants were significantly less likely to support the policy compared with non-Hispanic White participants (all P < 0.05). The difference for non-Hispanic Black participants was driven by a larger percentage with “no opinion,” although non-Hispanic Black participants were also more likely than non-Hispanic White participants to “strongly support” the policy (43% compared with 37%). Hispanic, non-Hispanic Asian, non-Hispanic Pacific Islander or Hawaiian, and non-Hispanic multiracial participants did not differ in support compared with non-Hispanic White participants (all P > 0.05), although these groups indicated higher levels of “strongly support.” There were 2 differences between these multivariable results and the bivariate model. In the bivariate model, Hispanic participants were significantly more likely to support the policy, but there was no significant difference in support for non-Hispanic American Indian or Alaskan Native participants. In post hoc comparison, Hispanic participants (75%) were more likely to support policy compared with non-Hispanic American Indian or Alaskan Native and non-Hispanic Black participants (all P < 0.05). Other significant post hoc differences included non-Hispanic Asian (74%) and non-Hispanic Multiracial (69%) participants having higher support than non-Hispanic Black and non-Hispanic American Indian or Alaskan Native participants (all P < 0.05). For race and ethnicity categories, opposition (6%–19%) and “no opinion” (8%–22%) had a similar range of proportions.
There was a higher likelihood of support with higher levels of educational attainment in both the multivariable and bivariate models. Compared with those with a Bachelor’s degree or higher, support was lower among those with up to a high school diploma or GED (67% compared with 78%; PR = 0.92; 95% CI: 0.90, 0.95; P < 0.001) and among those with some college or an associate’s degree (71% compared with 78%; PR = 0.96; 95% CI: 0.94, 0.98; P = 0.001). In a post hoc comparison, support among those with up to a high school diploma or GED was also significantly lower than those with some college or an associate’s degree (P = 0.003). Across education levels, opposition ranged from 8% to 10% whereas a larger proportion expressed “no opinion” (14% to 23%).
A similar pattern was observed regarding income; there was a higher likelihood of support with higher income levels in the multivariable and bivariate models. Compared with the highest income group (≥$95,000/y), support was lower among those with ≤$35,000/y (67% compared with 78%; PR = 0.91; 95% CI: 0.88, 0.94; P < 0.001), those with $35,001 to $65,000/y (70% compared with 78%; PR = 0.95; 95% CI: 0.92, 0.98; P < 0.001), and those with $65,001 to $95,000/y (72% compared with 78%; PR = 0.96; 95% CI: 0.94, 0.99; P = 0.009). In post hoc comparisons, support was significantly higher among those with $35,001 to $65,000/y and those with $65,001 to $95,000/y than those with ≤$35,000/y (all P < 0.01). Similarly to education, across income categories, opposition ranged from 8% to 10%, whereas a larger percentage expressed “no opinion” (14% to 24%).
Compared with participants in the South, there was significantly, albeit modestly, higher support among participants in the Northeast (71% compared with 74%; PR = 1.03; 95% CI: 1.01, 1.06; P = 0.015) and no differences in support for participants in the West, Midwest, or United States territories (all P > 0.05). However, participants in the West and United States territories were significantly more likely to support the policy in the bivariate model. There was also higher support in the Northeast compared with the West (72%) in post hoc comparisons (P = 0.022). Opposition ranged from 0% to 10% with larger percentages expressing “no opinion” (13% to 20%) across regions.
In general, we observed that for participant characteristics associated with lower support for the policy (e.g., those not trying to reduce added-sugar intake), there were higher proportions of participants responding “no opinion.” For example, about one-third (32%) of participants who were not trying to reduce their added-sugar intake and 21% of participants without a dietary restriction responded “no opinion” (Table 1). We also observed that the proportion of “no opinion” decreased as income and educational attainment increased.
Discussion
In a large online cross-sectional convenience sample, we found that majorities of participants across all subpopulations examined (54% to 87%) supported a policy requiring added-sugar warning labels on restaurant menus. Support was particularly high in key groups, such as those with prediabetes. This is consistent with research suggesting that consumers are generally supportive of nutrition intervention policies, with labeling policies, specifically, receiving high support [[49], [50], [51], [52]]. The observed range of support herein, including the previously reported 72% overall support in this sample, [18] is comparable with national [39] and New York state [38] polls on a restaurant menu added-sugar warning labels, which found 70% and 78% support, respectively, as well as prior research showing 80% overall support [17] for such a policy. There was very high support among those with a prediabetes diagnosis and among those trying to reduce their added-sugar intake, possibly because reducing added-sugar intake can be particularly challenging [53]. A key reason it can be challenging in restaurant environments is that the United States Food and Drug Administration does not currently require that restaurants disclose added-sugar information, making it nearly impossible to determine the added-sugar content of menu items. We also observed high support among those with high restaurant food ordering frequency. This pattern of support is encouraging, as it suggests an opportunity to engage frequent restaurant patrons in policy advocacy efforts and potentially indicates that policymakers would encounter minimal opposition from those who would be most affected by these types of policies.
We observed slightly higher support among women, in line with previous research indicating that women are more likely to be supportive of nutrition policies [52,[54], [55], [56]]. Also consistent with the literature, we observed higher levels of support among higher levels of income and higher levels of educational attainment [52]. The association between higher socioeconomic status and higher nutrition policy support is possibly because higher socioeconomic status is associated with both greater political engagement [57] and health literacy [58], although we did not assess political engagement or health literacy. We also observed high support among participants who reported living in the Northeast region of the United States, which could be because this region has already implemented restaurant menu-warning labeling policies [25,26,28]. Differences in support by regions may also reflect underlying regional norms and behaviors regarding sugary items, such as variations in sugar-sweetened beverage intake [59].
The meaningful proportion of participants responding “no opinion,” which was about double the proportion of those opposing, could be a reflection of limited familiarity with labeling policies and highlight opportunities for educational outreach. However, even with limited familiarity, support remained high across demographics, indicating the potential for coalition building and engagement across communities for policy introduction and implementation. For example, the Interfaith Public Health Network played a significant role in convening community organizations to build a coalition increasing public support of NYC’s Sweet Truth Act, which introduced restaurant menu added-sugar warning labels [60].
Although the multivariable model disentangles characteristics that drive support, the bivariate models are still informative and provide valuable insights by descriptively identifying characteristics associated with increased support; some differences between the multivariable and bivariate models could be worth further exploration. For example, in the bivariate model, Hispanic participants were significantly more likely to support than non-Hispanic White participants, but this difference was not significant in the multivariable model. The same pattern, significantly higher support in the bivariate model but no significant difference in the multivariable model, was observed for those diagnosed with obesity and those diagnosed with type 2 diabetes. These bivariate model results may identify groups that could be important for coalition building and advocacy efforts. Future policy work could engage with these key populations to better understand which aspects of labeling policies are most helpful and to help increase awareness of policies within their communities.
Strengths and limitations
Strengths of this study include a large national sample and inclusion of key groups. Limitations of this study include using quota, not probability, sampling and an online convenience sample. However, quota sampling can be a viable alternative to probability sampling [61,62] and, although it limits generalizability, quota sampling is useful for detecting differences by sociodemographics [63]. Additionally, the 1212 excluded participants (7%) differed modestly from the final analytic sample in the distributions of race/ethnicity, education, and age; however, distributions of these characteristics in the analytic sample closely approximate national distributions. Furthermore, support was assessed using a specific warning label (i.e., an upside-down triangle icon with an exclamation mark over a spoon applied to items with >25 g of added sugar); support may differ for other label designs or thresholds. These findings are also limited in generalizability for non–English-speaking and low–English-literacy populations, as the survey was only provided in English. Lastly, specifics of the policy, such as label size or the definition of “chain restaurant,” were not described to the participants, possibly leading some participants to respond “no opinion.”
In conclusion, in this large national sample, the majority of participants across all subgroups examined were supportive of a policy that would require added-sugar warning labels on restaurant menus. Support was notably high among those with prediabetes and those trying to reduce added-sugar intake—populations that could potentially benefit the most from this type of policy. High support among those who frequently order restaurant food and low opposition across groups indicates that policymakers would likely encounter minimal resistance from constituents. Overall, these results suggest that a restaurant menu added-sugar warning label policy is a popular and promising intervention for advocates and policymakers to pursue, although future studies should confirm high support in nationally representative samples and in the specific jurisdictions considering these policies. These results also indicate that the United States Food and Drug Administration should mandate that chain restaurants disclose the added-sugar content of menu items, as is currently required for nutrients such as sodium, total sugar, and saturated fat. Such disclosure would facilitate the implementation of local and state added-sugar labeling laws in restaurants.
Author contributions
The authors’ responsibilities were as follows – BL: formal analysis, data curation, visualization, writing – original draft, writing – reviewing & editing; AER: methodology, formal analysis, writing – reviewing and editing; AAM, CAR, MGH: methodology, writing – reviewing and editing; SS, DN, SDB, LEA: writing – reviewing and editing; JF: conceptualization, methodology, formal analysis, investigation, data curation, supervision, funding acquisition, writing – review and editing; and all authors: have read and approved the final version of this manuscript.
Data availability
Data described in the manuscript will be made available upon request for approved noncommercial research purposes only.
Funding
The questions used in this survey were added to a study supported by Bloomberg Philanthropies (2019-71208). SDB received support from National Institutes of Health (NIH) grants K26 DK138246 and P30 DK092924. JF received support from the United States Department of Agriculture/National Institute of Food and Agriculture (USDA/NIFA) Hatch project 7005204. This research was also supported in part by the American Diabetes Association grant CDTR-19.
Declaration of generative AI and AI-assisted technologies in the manuscript preparation process
The author(s) declare that no generative AI or AI-assisted technologies were used in the writing of this manuscript.
Conflict of interest
The authors report no conflicts of interest. The content of this article is solely the responsibility of the authors and does not necessarily represent the official views or policy of Bloomberg Philanthropies, USDA/NIFA, ADA, or the NIH. Bloomberg Philanthropies played no role in the study design, analytic plan, data collection, data curation or analysis, or the drafting or revising of the manuscript.
Footnotes
Supplementary data to this article can be found online at https://doi.org/10.1016/j.cdnut.2026.109442.
Appendix A. Supplementary data
The following is the Supplementary data to this article:
References
- 1.U.S. Department of Agriculture & U.S. Department of Health and Human Services . 2020. Dietary Guidelines for Americans 2020–2025 [Internet]https://www.dietaryguidelines.gov/about-dietary-guidelines/previous-editions/2020-dietary-guidelines Available from: [Google Scholar]
- 2.U.S. Centers for Disease Control and Prevention . 2023. National Diabetes Statistics Report [Internet]https://usdss.cdc.gov/diabetes/report.html Available from: [Google Scholar]
- 3.Hoy K., Steinfeldt L., Clemens J., Moshfegh A. Restaurant dining and energy intake among adults in the U.S., what we eat in America, NHANES 2015–2018. Curr. Dev. Nutr. 2021;5(Supplement_2):134. doi: 10.1093/cdn/nzab035_042. [DOI] [Google Scholar]
- 4.Liu J., Rehm C.D., Micha R., Mozaffarian D. Quality of meals consumed by US adults at full-service and fast-food restaurants, 2003–2016: persistent low quality and widening disparities. J. Nutr. 2020;150(4):873–883. doi: 10.1093/jn/nxz299. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Vercammen K.A., Frelier J.M., Moran A.J., Dunn C.G., Musicus A.A., Wolfson J.A., et al. Calorie and nutrient profile of combination meals at US fast food and fast casual restaurants. Am. J. Prev. Med. 2019;57(3):e77–e85. doi: 10.1016/j.amepre.2019.04.008. [DOI] [PubMed] [Google Scholar]
- 6.Yang Y., Weippert M., Ahmed M., L'Abbé M.R. Cross-sectional nutritional information and quality of Canadian chain restaurant menu items in 2020. Am. J. Prev. Med. 2023;64(1):42–50. doi: 10.1016/j.amepre.2022.07.015. [DOI] [PubMed] [Google Scholar]
- 7.Gesteiro E., García-Carro A., Aparicio-Ugarriza R., González-Gross M. Eating out of home: influence on nutrition, health, and policies: a scoping review. Nutrients. 2022;14(6):1265. doi: 10.3390/nu14061265. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Wen H., Li S., Hun M., Yang Y., Xiong L., Wen H., et al. Takeaway food consumption, dietary inflammatory index, and cardiometabolic risk factors in US adults: findings from NHANES (2009–2018) Food Sci. Nutr. 2025;13(12) doi: 10.1002/fsn3.71316. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Godbharle S., Jeyakumar A., Giri B.R., Kesa H. Pooled prevalence of food away from home (FAFH) and associated non-communicable disease (NCD) markers: a systematic review and meta-analysis. J. Health Popul. Nutr. 2022;41(1):55. doi: 10.1186/s41043-022-00335-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Wellard-Cole L., Davies A., Allman-Farinelli M. Contribution of foods prepared away from home to intakes of energy and nutrients of public health concern in adults: a systematic review. Crit. Rev. Food Sci. Nutr. 2022;62(20):5511–5522. doi: 10.1080/10408398.2021.1887075. [DOI] [PubMed] [Google Scholar]
- 11.Du Y., Rong S., Sun Y., Liu B., Wu Y., Snetselaar L.G., et al. Association between frequency of eating away-from-home meals and risk of all-cause and cause-specific mortality. J. Acad. Nutr. Diet. 2021;121(9):1741–1749.e1. doi: 10.1016/j.jand.2021.01.012. [DOI] [PubMed] [Google Scholar]
- 12.Cahill L.E., Pan A., Chiuve S.E., Sun Q., Willett W.C., Hu F.B., et al. Fried-food consumption and risk of type 2 diabetes and coronary artery disease: a prospective study in 2 cohorts of US women and men. Am. J. Clin. Nutr. 2014;100(2):667–675. doi: 10.3945/ajcn.114.084129. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Krishnan S., Coogan P.F., Boggs D.A., Rosenberg L., Palmer J.R. Consumption of restaurant foods and incidence of type 2 diabetes in African American women. Am. J. Clin. Nutr. 2010;91(2):465–471. doi: 10.3945/ajcn.2009.28682. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Odegaard A.O., Koh W.P., Yuan J.-M., Gross M.D., Pereira M.A. Western-style fast food intake and cardiometabolic risk in an Eastern country. Circulation. 2012;126(2):182–188. doi: 10.1161/CIRCULATIONAHA.111.084004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Pereira M.A., Kartashov A.I., Ebbeling C.B., Van Horn L., Slattery M.L., Jacobs D.R., Jr., et al. Fast-food habits, weight gain, and insulin resistance (the CARDIA study): 15-year prospective analysis. Lancet. 2005;365(9453):36–42. doi: 10.1016/S0140-6736(04)17663-0. [DOI] [PubMed] [Google Scholar]
- 16.U.S. Department of Health and Human Services Food and Drug Administration Food Labeling; Nutrition Labeling of Standard Menu Items in Restaurants and Similar Retail Food Establishments. Federal Register [Internet] 2014 https://www.federalregister.gov/d/2014-27833 Available from: [PubMed] [Google Scholar]
- 17.Sigala D.M., Hall M.G., Musicus A.A., Roberto C.A., Solar S.E., Fan S., et al. Perceived effectiveness of added-sugar warning label designs for U.S. restaurant menus: an online randomized controlled trial. Prev. Med. 2022;160 doi: 10.1016/j.ypmed.2022.107090. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Falbe J., Musicus A.A., Sigala D.M., et al. Online randomized controlled trial of icon added-sugar warning labels for restaurant menus. Am. J. Prev. Med. 2023;65(1):100–111. doi: 10.1016/j.amepre.2023.02.007. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Huang Y., Lemmon B., Musicus A.A., Roberto C.A., Hall M.G., Falbe J. Effects of added-sugar warning labels on restaurant menus in the USA: an online randomised controlled trial. Lancet Public Health. 2026;11(6):e375–e385. doi: 10.1016/S2468-2667(26)00074-5. [DOI] [PubMed] [Google Scholar]
- 20.Lemmon B., Gil A., Musicus A.A., Hall M.G., Roberto C.A., Falbe J. Optimal design of an added-sugar menu warning label: a randomized experiment. Am. J. Prev. Med. 2025;69(6) doi: 10.1016/j.amepre.2025.108038. [DOI] [PubMed] [Google Scholar]
- 21.Jensen M.L., Choi Y.Y., Fleming-Milici F., Harris J.L. Caregivers' understanding of ingredients in drinks served to young children: opportunities for nutrition education and improved labeling. Curr. Dev. Nutr. 2022;6(1) doi: 10.1093/cdn/nzab151. nzab151. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Moran A.J., Roberto C.A. Health warning labels correct parents' misperceptions about sugary drink options. Am. J. Prev. Med. 2018;55(2):E19–E27. doi: 10.1016/j.amepre.2018.04.018. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Grummon A.H., Petimar J., Zhang F., Rao A., Gortmaker S.L., Rimm E.B., et al. Calorie labeling and product reformulation: a longitudinal analysis of supermarket-prepared foods. Am. J. Prev. Med. 2021;61(3):377–385. doi: 10.1016/j.amepre.2021.03.013. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Grummon A.H., Petimar J., Soto M.J., Bleich S.N., Simon D., Cleveland L.P., et al. Changes in calorie content of menu items at large chain restaurants after implementation of calorie labels. JAMA Netw. Open. 2021;4(12) doi: 10.1001/jamanetworkopen.2021.41353. e2141353–e2141353. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.New York City Department of Health and Mental Hygiene. Sodium Warning Rule in NYC Chain Restaurants [Internet], 2022. Available from: https://www.nyc.gov/assets/doh/downloads/pdf/cardio/sodium-warning-rule.pdf [March 31, 2026; date cited].
- 26.City of Philadelphia . American Legal Publishing [Internet]; In: Philadelphia: 2019. The Philadelphia Code Health Code. 6-310: Sodium SafetyWarning Labeling for Chain Establishments.https://codelibrary.amlegal.com/codes/philadelphia/latest/philadelphia_pa/0-0-0-272915 [Google Scholar]
- 27.The New York City Council . City Hall; New York, NY: 2026. A local law to amend the administrative code of the city of New York, in relation to requiring added sugar notifications in chain restaurants. Int. 0454-2026.https://legistar.council.nyc.gov/LegislationDetail.aspx?ID=7862996&GUID=7D642A46-6A49-4469-B6F2-88412EDD35A5 Available from: [Google Scholar]
- 28.M. Sheldon, “Sweet Truth Act” Requires High-Sugar Warnings on Menus at NYC Chain Restaurants [Internet], 2021. Available from: https://www.nycfoodpolicy.org/food-policy-snapshot-nyc-sweet-truth-act/.
- 29.C. Wilder, ‘Cut the Salt’ coalition asks Cleveland City Council to require restaurants to label high-sodium foods. Signal Cleveland [Internet], 2024. Available from: https://signalcleveland.org/cut-the-salt-coalition-asks-cleveland-city-council-to-require-restaurants-to-label-high-sodium-foods/.
- 30.Davis J.N., Goon S., Gouck J., Solar S.E., Mancini S., Hinojosa A.M.O., et al. An inventory of proposed and enacted sugar-sweetened beverage policies at the state, local, and tribal levels in the United States, 2014‒2023. Am. J. Public Health. 2024;114(12):1344–1353. doi: 10.2105/ajph.2024.307855. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.An act to amend the public health law, in relation to requiring chain restaurants to label menu items that have a high content of sodium, Senate Bill S4890A, State of New York Senate (Rivera G 2024) [Internet], 2023. Available from: https://www.nysenate.gov/legislation/bills/2023/S4890/amendment/A. (Accessed 31 March 2026).
- 32.The New York State Senate An act to amend the public health law, in relation to requiring chain restaurants to label menu items that have a high content of added sugars. Assembly Bill A6546B, State of New York Assembly [Internet] 2024 https://www.nysenate.gov/legislation/bills/2023/A6546/amendment/B Available from: [Google Scholar]
- 33.Legiscan . 2026. Restaurant menus: added sugar warnings, SB869, California State Legislature, 2025–2026 Regular sess [Internet]https://legiscan.com/CA/text/SB869/id/3398854 Available from: [Google Scholar]
- 34.Public Health – Chain Restaurants – Disclosure and Notice to Customers of Sodium and Added Sugars, SB0866, Maryland General Assembly [Internet] 2026. https://mgaleg.maryland.gov/mgawebsite/Legislation/Details/SB0866?ys=2026RS Available from: [Google Scholar]
- 35.An Act for healthy kids, healthy futures, SB1571, Senate and House of Representatives of the Commonwealth of Massachusetts [Internet] 2025. https://legiscan.com/MA/text/S1571/2025 Available from: [Google Scholar]
- 36.Grummon A.H., Reimold A.E., Hall M.G. Influence of the San Francisco, CA, sugar-sweetened beverage health warning on consumer reactions: implications for equity from a randomized experiment. J. Acad. Nutr. Diet. 2022;122(2):363–370.e6. doi: 10.1016/j.jand.2021.07.008. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Lagos M. 2014. Soda tax, warning labels backed by voters, poll finds. SF Gate. [Internet]https://www.sfgate.com/bayarea/article/soda-tax-warning-labels-backed-by-voters-poll-5251547.php Available from: [Google Scholar]
- 38.Center for Science in the Public Interest . 2021. New Yorkers support added sugar warnings on chain restaurant menus [Internet]https://www.cspi.org/resource/new-york-added-sugars-warning-label-poll Available from: [Google Scholar]
- 39.Center for Science in the Public Interest . 2026. Serving up the truth: Poll shows strong support for added sugars info at chain restaurants.https://www.cspi.org/resource/serving-truth Available from. [Google Scholar]
- 40.Chen Y., Zhou X., Bullard K.M., Zhang P., Imperatore G., Rolka D.B. Income-related inequalities in diagnosed diabetes prevalence among US adults, 2001−2018. PLoS One. 2023;18(4) doi: 10.1371/journal.pone.0283450. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Yedjou C.J., Sims J.N., Njiki S., Chitoh A.M., Joseph M., Cherkos A.S., et al. Health and racial disparities in diabetes mellitus prevalence, management, policies, and outcomes in the United States. J. Community Med. Public Health. 2024;8(3):460. doi: 10.29011/2577-2228.100460. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.McClinchy J., Dickinson A., Wills W. ‘Feta is obviously very dangerous stuff looking at all those reds’: can nutrition labelling help people living with type 2 diabetes to self-manage their condition? J. Hum. Nutr. Diet. 2023;36(4):1131–1143. doi: 10.1111/jhn.13152. [DOI] [PubMed] [Google Scholar]
- 43.U.S. Census Bureau. 2014–2018 American Community Survey [Internet], 2019. Available from: https://www.census.gov/acs/www/data/data-tables-and-tools/data-profiles/2018/ [May 8, 2021, date cited]. (Accessed 31 March 2026).
- 44.Marinello S., Powell L.M., Falbe J. Neighborhood sociodemographic characteristics and healthfulness of store checkouts in Northern California. Prev. Med. Rep. 2023;35 doi: 10.1016/j.pmedr.2023.102379. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Barnhill A., Ramírez A.S., Ashe M., Berhaupt-Glickstein A., Freudenberg N., Grier S.A., et al. The racialized marketing of unhealthy foods and beverages: perspectives and potential remedies. J. Law Med. Ethics. 2022;50(1):52–59. doi: 10.1017/jme.2022.8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Harris J.L., Fleming-Milici F., Kibwana-Jaff A., Phaneuf L. UConn Rudd Center for Food Policy and Health, University of Connecticut; CT, United States: 2020. Sugary Drink FACTS 2020. Sugary Drink Advertising to Youth: Continued Barrier to Public Health Progress. [Google Scholar]
- 47.Cooksey Stowers K., Jiang Q., Atoloye A.T., Lucan S., Gans K. Racial differences in perceived food swamp and food desert exposure and disparities in self-reported dietary habits. Int. J. Environ. Res. Public Health. 2020;7(19):7143. doi: 10.3390/ijerph17197143. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48.Zou G.Y. A modified Poisson regression approach to prospective studies with binary data. Am. J. Epidemiol. 2004;159(7):702–706. doi: 10.1093/aje/kwh090. [DOI] [PubMed] [Google Scholar]
- 49.Pettigrew S., Booth L., Dunford E., Scapin T., Webster J., Wu J., et al. An examination of public support for 35 nutrition interventions across seven countries. Eur. J. Clin. Nutr. 2023;77(2):235–245. doi: 10.1038/s41430-022-01211-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 50.Lobstein T., Neveux M., Landon J. Costs, equity and acceptability of three policies to prevent obesity: a narrative review to support policy development. Obes. Sci. Pract. 2020;6(5):562–583. doi: 10.1002/osp4.423. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 51.Espinosa R., Nassar A. The acceptability of food policies. Nutrients. 2021;13(5):1483. doi: 10.3390/nu13051483. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 52.Scheidmeir M., Kubiak T., Luszczynska A., Wendt J., Scheller D.A., Meshkovska B., et al. Acceptability of policies targeting dietary behaviours and physical activity: a systematic review of tools and outcomes. Eur. J. Public Health. 2022;32(Supplement_4):iv32–iv49. doi: 10.1093/eurpub/ckac053. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 53.Skoglund G., Nilsson B.B., Olsen C.F., Bergland A., Hilde G. Facilitators and barriers for lifestyle change in people with prediabetes: a meta-synthesis of qualitative studies. BMC Public Health. 2022;22(1):553. doi: 10.1186/s12889-022-12885-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 54.Fleming-Milici F., Harris J.L., Liu S. Race, ethnicity, and other factors predicting U.S. parents' support for policies to reduce food and beverage marketing to children and adolescents. Health Equity. 2018;2(1):288–295. doi: 10.1089/heq.2018.0048. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 55.Andreas M., Kaiser A.K., Dhami, Brugger V., Sniehotta F.F. How beliefs and policy characteristics shape the public acceptability of nutritional policies—a survey study in Germany. Health Policy. 2025;152 doi: 10.1016/j.healthpol.2024.105221. [DOI] [PubMed] [Google Scholar]
- 56.Lambert-De Francesch J., Saint-Onge K., Muhajarine N., Gauvin L. Sociodemographic characteristics help predict Canadian urbanites’ acceptability of restaurant food environment policies. Front. Nutr. 2024;11 doi: 10.3389/fnut.2024.1360360. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 57.Laurison D. Social class and political engagement in the United States. Sociol. Compass. 2016;10(8):684–697. doi: 10.1111/soc4.12390. [DOI] [Google Scholar]
- 58.Stormacq C., Van den Broucke S., Wosinski J. Does health literacy mediate the relationship between socioeconomic status and health disparities? Integrative review. Health Promot. Int. 2019;34(5):e1–e17. doi: 10.1093/heapro/day062. [DOI] [PubMed] [Google Scholar]
- 59.Andreyeva T. Large state variation in sugar-sweetened beverage purchases: what we learn from the beverage industry data. Curr. Dev. Nutr. 2021;5(12) doi: 10.1093/cdn/nzab128. nzab128. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 60.Gottlieb A. 2023. The Not So Sweet Truth [Internet]https://www.iphnetwork.org/the_not_so_sweet_truth Available from: [Google Scholar]
- 61.Cumming R.G. Is probability sampling always better? A comparison of results from a quota and a probability sample survey. Aust. N. Z. J. Public Health. 1990;14(2):132–137. doi: 10.1111/j.1753-6405.1990.tb00033.x. [DOI] [PubMed] [Google Scholar]
- 62.Yang K., Banamah A. Quota sampling as an alternative to probability sampling? An experimental study. Sociol. Res. Online. 2014;19(1):56–66. doi: 10.5153/sro.3199. [DOI] [Google Scholar]
- 63.Bornstein M.H., Jager J., Putnick D.L. Sampling in developmental science: situations, shortcomings, solutions, and standards. Dev. Rev. 2013;33(4):357–370. doi: 10.1016/j.dr.2013.08.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Data described in the manuscript will be made available upon request for approved noncommercial research purposes only.