Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2024 Jun 1.
Published in final edited form as: Am J Prev Med. 2023 Feb 14;64(6):805–813. doi: 10.1016/j.amepre.2023.01.026

Naming Matters: Prompting Smaller Portions in an Online Randomized Controlled Trial

Sophia V Hua 1, Erica L Kenney 1, Jeffrey M Miller 2, Aviva A Musicus 3, Christina A Roberto 4, Anne N Thorndike 5, Eric B Rimm 1,6
PMCID: PMC10200744  NIHMSID: NIHMS1868778  PMID: 36792450

Abstract

Introduction:

Large portions, which can lead people to eat more, are becoming increasingly common in U.S. restaurants. This study tested whether portion-size descriptions on menus and different pricing strategies influence selection of smaller portion sizes.

Study Design:

This was a 4×2 between subjects online randomized controlled experiment.

Setting/participants:

This was an online simulated menu-ordering study conducted in 2021 among 2,205 U.S. adults.

Intervention:

Adults viewed a fast-casual and a full-service menu with entrées available in two sizes and ordered an entrée from each. Participants were randomized to view one of four portionsize descriptors (Smaller/Larger portion): 1) no descriptor/“Large” (control); 2) “Standard”/“Large”; 3) “Just Right”/“Large”; and 4) no descriptor/“Hearty”. Participants were also randomized to either linear (i.e., reduced price=50% larger portion’s price) or non-linear pricing (i.e., reduced price=70% larger portion’s price) (4×2 factorial design).

Main outcome measures:

In 2022, logistic regression models were used to analyze whether the interventions increased the likelihood of choosing a reduced portion.

Results:

Regardless of pricing scheme, participants in the “Standard/Large” condition selected reduced portions by 10 (95% CI: 0.04, 0.16) and 13 (95% CI: 0.07, 0.18) percentage points more than those in the control condition (fast-casual and full-service menus, respectively). Selection of reduced portions in the “Just Right”/“Large” condition increased by 9 (95% CI: 0.04, 0.15) and 8 (95% CI: 0.02, 0.14) percentage points. For the fast-casual menu, holding portion-size descriptors constant, participants ordered a reduced portion by 5 percentage points more with non-linear pricing compared to linear pricing.

Introduction

Portion sizes for foods and beverages in the U.S. have increased over time, and this parallels an increase in dining out.1–3 When served large portions, people consume in excess without realizing it, and in the restaurant setting, this means consuming more foods that tend to be higher in calories, sodium, and saturated fats than foods cooked at home.4–6 Excess weight and obesity are driven in part by larger portion sizes, which have reset norms around reasonable serving sizes and contribute to overeating.7–9

Developing low-cost interventions to guide consumers to choose smaller portions has the potential to help people stay within calorie recommendations and potentially shift perceived norms about portion sizes towards appropriate servings to support a healthy, active life. Hollands et al. found that portion size reduction has the potential to reduce average daily energy consumption among adults by up to 16% and many studies have shown that larger portions lead to increased consumption.7,10,11 Researchers have found that when people are served less food, they eat less without feeling less sated,12–14 though it is not fully documented in the literature if people compensate for these calories later in the day.14,15 In one of the few studies conducted in a real-world setting, nearly one-third of consumers in a Chinese take-out restaurant opted to receive reduced side dish portions for the same cost as the regular portions.16 Downsizing a part of their meal led people to consume fewer calories and did not lead to overconsumption of other portions of the meal.

Current restaurant pricing strategies also play an important role in encouraging the selection of large portions when dining out.17 Many restaurants use “non-linear pricing,” in which the price per ounce is cheaper as the portion size increases, thus incentivizing customers to order larger sizes.18 This is in contrast to linear pricing, in which the price per ounce stays the same regardless of size. Despite models suggesting that non-linear pricing should increase the ordering of larger-sized meals,18 experimental studies that have examined the effects of linear and non-linear pricing on portion size choice have found no significant effects.19–22 For example, a study by John et al. showed that beverage selection from restaurant menus offering three different beverage sizes was not modified by the pricing scheme used.23 Because the restaurant industry operates on thin margins, it is important to understand how portion size reduction interventions may be influenced by non-linear or linear pricing schemes.

In recognition of the role large portions play in promoting overconsumption of unhealthy foods, New York City attempted and failed in 2013 to implement a cap on the size of sugary drink portions.24 This highlights a need to explore low-cost portion size reduction strategies that restaurants can voluntarily implement. Therefore, the goal of this study was to test the effects of manipulating portion-size descriptions on fast-casual and full-service restaurant menus on consumers’ choice of reduced portions. The study also examined if the portion-size description manipulation was dependent on non-linear or linear pricing schemes.

The primary hypotheses of the study were that a reduced portion entrée described as the “Standard” size or “Just Right” (compared to having no descriptor) would increase the odds of customers choosing the smaller portion size, regardless of whether linear or non-linear pricing was used. These hypotheses were informed by the psychological insight that people like to do what they think other people will do in situations of uncertainty.25 For this reason, we predicted that calling the smaller size “Standard” or “Just Right” would lead consumers to choose reduced portions because they think that size is what others typically select. As for pricing, we hypothesized that using linear pricing would increase the odds that participants will select the smaller portion size compared to the non-linear pricing condition, regardless of portion size descriptors.

Methods

Study Sample

A pre-registered (AsPredicted #72850) online randomized controlled experiment with a 4×2 factorial design was conducted. Data were collected September–October 2021. This study was determined exempt by the Harvard T.H. Chan School of Public Health IRB. The sample consisted of U.S. adults who spoke English and reported eating out or ordering from a restaurant at least twice in the past four weeks. Participants were recruited by Dynata, a research firm with an online panel of U.S. consumers, to take an online Qualtrics survey about consumer preferences when dining out. Dynata compensated participants with an internal point system based on the length of the survey and requested sample characteristics. Accumulated points can be used to redeem gift cards, charitable contributions, and partner products. Quotas were used to recruit a sample with an educational attainment that aligned with the U.S. distribution according to the 2020 Census.26 In total, 2,785 participants were randomized and completed the survey. Participants were excluded from primary analyses if they met the following pre-specified criteria: 1) completed the survey in less than one-third of the median completion time (<118 seconds; n=23); or 2) did not pass the attention check question asking, “What month are we were in now?” (n=404). Although not pre-specified, participants were further excluded from the primary analyses if they omitted race (n=5), BMI (n=58), or numeracy (n=42) information, or had a reported BMI <13 kg/m2 (n=48). Since the secondary analyses controlled for these demographic variables, excluding those missing the relevant information allowed the sample size to be the same for both primary and secondary analyses. The final analytic sample included 2,205 participants (See Figure 1 for CONSORT diagram).

Figure 1:

Figure 1:

Open in a new tab

CONSORT diagram for an online randomized controlled experiment examining selection of a reduced portion entrée on menus with different portion-size descriptor

Intervention

Participants were randomized to see one of eight sets of menus, with each set including a menu inspired by a fast-casual restaurant (e.g., Panera) and a menu inspired by a full-service chain restaurant (e.g., TGI Fridays), shown in a random order (Figure 1). The authors reviewed existing restaurant menus to come up with plausible menu items for the study. All menu items were labeled with their calorie content as required by national law. Each menu showed entrées available in two sizes: a reduced portion and a larger portion. The calorie contents for the larger entrées were modeled after standard sizes in restaurants, and the reduced-portion entrées had half as many calories. The mean number of calories per order in the fast-casual menu assuming the side salad was 100 calories and the beverage was 150 calories was 665 calories; in the full-service menu, it was 770 calories.

Participants were randomized to see menu sets with one of four pairs of portion-size descriptors (reduced portion and larger portion): no descriptor and “Large”; “Standard” and “Large”; “Just Right” and “Large”; and no descriptor and “Hearty”. The no descriptor and “Large” condition was the control based on the labeling used at several popular chain restaurants that offer multiple entrée sizes.27,28 Participants were additionally randomized to one of two pricing conditions: linear pricing (i.e., the reduced portion was 50% of the larger entrée’s price) and non-linear pricing (i.e., the reduced portion was 70% of the larger entrée’s price). The larger portion entrée prices were held constant while the reduced portion entrée prices changed.

After providing informed consent, participants viewed the two menus one at a time and were instructed to imagine they were eating out on a typical night and to order one entrée from each menu. Participants clicked on the menu item they wanted to select, similar to how they might for an online ordering platform.

Measures

The primary outcome was a dichotomous variable indicating whether or not a participant chose a reduced portion entrée from each restaurant menu (fast-casual and full-service chain). Participants were additionally asked questions about their menu choices, followed by a series of demographic questions, and a 3-question, validated numeracy scale.29 (See Appendix A for sample menus and Appendix B for the survey.) Numeracy was measured because the only indicator that the reduced portion was half the size of the larger portion was the calorie labels; measurement of numeracy allowed it to be included in the analytic model if it was not balanced across conditions.

Statistical Analysis

Analyses were conducted separately for the fast-casual and full-service restaurant menus because portion sizes are usually larger in full-service restaurants, a pattern reflected in the study menus. Primary analyses used logistic regression models to assess how naming and pricing affected choice of a reduced portion entrée. Each model contained a categorical variable for portion-size-description condition (reference=control), a binary variable for pricing condition (non-linear vs. linear), and an interaction between the two variables. The interactions were not significant in any model, so they were removed from the final models. Differences across conditions by demographic variables such as age and gender identity were tested using ANOVA. Because conditions were randomized, demographic variables were balanced across conditions, so the primary analyses did not control for other variables. Sensitivity analyses for primary outcomes were performed on: 1) the full sample without any of the exclusions based on time to completion, accuracy of attention check answer, or missing information; and 2) the prespecified sample, with exclusions for time to completion and accuracy of attention check answer.

Secondary analyses were conducted to develop more precise estimates and to demonstrate that the results were not due to confounding. Models adjusted for age (categorical by decade; 18- and 19-year-olds were included in the <30 group), gender identity (categorical; female, male, other), BMI (categorical; underweight, normal weight, overweight, obese), race (categorical; White, Black, other), numeracy (categorical; score of 0–3 depending on number of accurate answers), and educational attainment (categorical; less than high school, high school/GED, some college, 2-year/Associate degree, 4-year degree, graduate degree), in addition to price and portion size variables.

Exploratory analyses tested for effect modification of the portion description by educational attainment and numeracy as both have been found to affect the efficacy of nutrition labels.30 Though the study was not explicitly powered to test for these interaction effects, these analyses allowed exploration of whether these factors could similarly affect the efficacy of portion-size descriptors. Finally, exploratory chi square tests were run to examine whether or not participants’ reasons for selecting a specific entrée differed by condition collapsed across the two types of pricing.

Odds ratios and predicted probabilities are presented. A sample size of 2200 provided 80% power to detect an odds ratio of 1.3 (α = 0.05, two tailed).16 Although 2,358 completed survey responses were collected after the pre-specified exclusions, a further 153 responses were excluded for missing demographic information included in the secondary analyses, which became the sample size for the primary analyses as well. All analyses were conducted using Stata version 17 (StataCorp LLC, College Station, TX).

Results

In the final analytic sample (n=2,205), the average age was 59.1 years, 53.0% identified as being female, and the majority of participants identified as White (82.5%). Participants had an average BMI in the overweight category (28.0 kg/m2). These variables did not differ by condition. See Table 1.

Table 1:

Participant demographics by condition for an online randomized controlled experiment examining selection of reduced portion entrées (n=2,205)

Participant Characteristic Linear Pricing (n=1,097) Non-Linear Pricing (n=1,108) p-valuea
no descriptor and “Large”
n=265		 “Standard” and “Large” n=282 “Just Right” and “Large”
n=281		 no descriptor and “Hearty”
n=269		 no descriptor and “Large”
n=278		 “Standard” and “Large” n=278 “Just Right” and “Large”
n=286		 no descriptor and “Hearty”
n=266
Age (years), mean (SD) 59.5 (18.2) 59.6 (18.1) 59.2 (17.9) 57.9 (19.1) 59.6 (18.8) 58.1 (18.2) 59.6 (18.3) 58.9 (17.9) 0.89
Gender Identity, n(%) 0.85
Female 148 (55.9) 148 (52.5) 150 (53.4) 136 (50.5) 148 (53.2) 155 (55.8) 145 (50.7) 138 (51.9)
Male 116 (43.8) 130 (46.1) 130 (46.3) 132 (49.1) 129 (46.4) 121 (43.5) 138 (48.3) 127 (47.7)
Other 1 (0.4) 4 (1.4) 1 (0.4) 1 (0.4) 1 (0.4) 2 (0.7) 3 (1.1) 1 (0.4)
Race, n(%) 0.93
Black 27 (10.2) 33 (11.7) 30 (10.7) 32 (11.9) 29 (10.4) 36 (13.0) 36 (12.6) 34 (12.8)
White 221 (83.4) 230 (81.6) 236 (84.0) 219 (81.4) 229 (82.4) 226 (81.3) 239 (83.6) 220 (82.7)
Other b 17 (6.4) 19 (6.7) 15 (5.3) 18 (6.7) 20 (7.2) 16 (5.8) 11 (3.9) 12 (4.5)
BMI (kg/m2), mean (SD) 27.8 (6.2) 28.2 (6.8) 27.7 (9.0) 27.8 (6.4) 27.7 (6.2) 28.0 (6.4) 28.2 (6.9) 28.8 (7.5) 0.56
Educ Attainment, n(%) 0.73
Less than HS 18 (6.8) 26 (9.2) 29 (10.3) 24 (8.9) 28 (10.1) 24 (8.6) 27 (9.4) 26 (9.8)
High school/GED 80 (30.2) 80 (28.4) 79 (28.1) 68 (25.3) 87 (31.3) 81 (29.1) 84 (29.4) 74 (27.8)
Some college 55 (20.8) 43 (15.3) 50 (17.8) 53 (19.7) 50 (18.0) 35 (12.6) 46 (16.1) 53 (19.9)
2-yr/Associate 32 (12.1) 34 (12.1) 19 (6.8) 33 (12.3) 24 (8.6) 36 (13.0) 27 (9.4) 27 (10.2)
4-yr/University 56 (21.1) 59 (20.9) 60 (21.4) 56 (20.8) 59 (21.2) 65 (23.4) 71 (24.8) 58 (21.8)
Graduate 24 (9.1) 40 (14.2) 44 (15.7) 35 (13.0) 30 (10.8) 37 (13.3) 31 (10.8) 28 (10.5)
Ordered reduced portion, n(%)
Fast Casual 140 (52.8) 184 (65.3) 165 (58.7) 144 (53.5) 159 (57.2) 179 (64.4) 200 (69.9) 155 (58.3) <0.001
Sit Down 148 (55.9) 193 (68.4) 178 (63.4) 155 (57.6) 163 (58.6) 197 (70.9) 193 (67.5) 148 (55.6) <0.001
Open in a new tab
a

ANOVA used to test for significant differences among conditions

b

Includes American Indian, Asian, and Native Hawaiian

For the fast-casual menu, holding pricing condition constant, the predicted probability of participants in the control condition (no descriptor and “Large”) choosing the reduced portion was 55% (95% CI: 0.51, 0.59). Participants who viewed a menu that referred to the smaller portion as “Standard” and the larger portion as “Large” had 1.5 (95% CI: 1.2, 2.0) times the odds—or an increase of 10 percentage points in predicted probability (95% CI: 0.04, 0.16)—of choosing the smaller size compared to the control condition. Participants who viewed the “‘Just Right’ and ‘Large’” menu had 1.5 (95% CI: 1.2, 1.9) times the odds and an increase of 9 percentage points (95% CI: 0.04, 0.15) of choosing the smaller portion size. The menu that displayed the name of the larger portion as “Hearty” while keeping the smaller portion unlabeled was not associated with increased odds of choosing a smaller portion (OR = 1.0 [95% CI: 0.8, 3]) compared to the control. See Figure 2 and Appendix C.

Figure 2:

Figure 2:

Open in a new tab

Predicted probabilities for selecting a reduced portion entrée by portion-size descriptors and menu type for an online randomized controlled experiment (n=2,205)

Footnotes:

**p<0.01; ***p<0.001 compared to the control condition

aLogistic regression models controlled for pricing scheme (non-linear vs. linear pricing)

For the full-service menu, similar patterns existed (Figure 2). Those who viewed the “‘Standard’ and ‘Large’” descriptors on the menu had 1.7 (95% CI: 1.3, 2.2) times the odds and an increase of 13 percentage points (95% CI: 0.07, 0.18) of choosing a reduced portion entrée, and those who viewed the “‘Just Right’ and ‘Large’” labeled menu had 1.4 (95% CI: 1.1, 1.8) times the odds and an increase of 8 percentage points (95% CI: 0.02, 0.14) of ordering a reduced portion entrée as compared to the control. As with the fast-casual menus, the odds of consumers choosing a reduced portion entrée when viewing the “no descriptor and ‘Hearty’” menu (OR = 1.0 [95% CI: 0.8, 1.2]) did not statistically differ from the control menu.

For the fast-casual menu, holding portion-size descriptors constant, customers had 1.2 (95% CI: 1.0, 1.5) times the odds of ordering a reduced portion entrée with non-linear pricing compared to linear pricing. For the full-service menu, the odds of choosing a reduced portion entrée did not differ between linear and non-linear pricing (OR = 1.1 [95% CI: 0.9, 1.3]) (Figure 3).

Figure 3:

Figure 3:

Open in a new tab

Predicted probabilities for selecting a reduced portion entrée by pricing scheme and menu type for an online randomized controlled experiment (n=2,205)

Footnotes:

**p<0.01 compared to the control condition

aLogistic regression models controlled for portion size descriptors (4 levels: no descriptor and “Large”; “Standard” and “Large”; “Just Right” and “Large”; and no descriptor and “Hearty”

Based on chi square tests, we did not observe any statistically significant differences among the portion size descriptor groups (collapsed across pricing conditions) in the reasons participants gave for selecting their entrée. The majority of participants gave the reason that, “The entrée sounded delicious” (fast-casual range: 37.8% - 40.8%; full-service range: 36.8% - 44.8%), followed by, “It was a healthy choice” (fast-casual range: 14.9% - 20.6%; full-service range: 11.6% - 18.1%). Participants who chose, “The calories seemed right for me” ranged from 3.2% - 5.5% (fast-casual menu) and 2.7% - 5.8% (full-service menu). “The portion size seemed right for me” was selected by 9.4% - 15.5% participants in the fast-casual setting, and 10.4% - 15.2% in the full-service setting.

Exploratory analyses were run examining effect modification of the portion description by education and numeracy. For the full-service menu, compared to participants with a high school degree, participants with “some college” or a “4-year degree” were significantly more likely to order reduced portions in the “no descriptor and ‘Hearty’” condition compared to the control “no descriptor and ‘Large’” condition (OR = 2.7 [95% CI: 1.3, 5.8]; OR = 2.7 [95% CI: 1.3, 5.6], respectively). No other significant education or numeracy interactions with portion-size descriptors were observed for portion size choice on either restaurant menu.

In secondary models which took the primary analyses’ models and additionally controlled for demographic characteristics, results were generally similar across both menus (Appendix D). Results were similar to the primary models for the sensitivity analyses with all participants who completed the survey (ignoring exclusion criteria) and with the pre-specified sample of participants with exclusions for time to completion and accuracy of attention check answer (Appendices E and F).

Discussion

This study found that naming the smaller portion “Standard” or “Just Right” on menus with two different-sized entrées increased the predicted probability of participants choosing reduced portion entrées by 8–12% in an online hypothetical restaurant setting. The impact of the portion-size descriptors did not differ by level of pricing. Non-linear pricing increased the predicted probability of selection of reduced portions in the fast-casual setting, though it did not impact selection of reduced portions in the full-service setting, keeping portion names constant.

These results suggest that it is possible to influence consumers to select smaller portion sizes by describing them as “Standard” or “Just Right.” Unexpectedly, referring to the larger size as “Hearty”—which was meant to connote “larger than average”—did not alter the odds of ordering a smaller size compared to the control condition. However, it is possible that in this study, “Hearty” did not properly convey how much bigger the larger size was, although all menus did have calorie counts that reflected that the larger size had twice as many calories as the smaller size. Simply naming the smaller size anything at all may have prompted participants to order that size, as the control condition and “no descriptor and ‘Hearty’” did not label the smaller size. Currently, restaurants like Chipotle and Blaze only describe the larger size (such as the control condition),27,28 so if merely adding a description for the smaller size can increase selection, then the strategy merits further exploration.

Despite the influence of price on restaurant choices,17 this study found that, holding portion-size descriptors constant, there were no differences in the odds of selecting reduced portions between linear and non-linear pricing in the full-service setting. Surprisingly, in the fast-casual setting, non-linearly priced menus compared to the linearly priced menus increased selection of reduced portions. This meant that participants were willing to pay more for a smaller portion. Even though reduced portions were more expensive in the non-linear price conditions, participants may have been using price instead of calories as a proxy for size. As a result, when they viewed prices that were only slightly lower than the larger size, they may have assumed the portion size would also be slightly smaller instead of half the size. It will be important for future research to examine whether this leads to consumer dissatisfaction when receiving their items. It is possible consumers were attracted to the price of the smaller size, but the price was consistent across experimental conditions, indicating it was the menu description that influenced their choice, or the combination of that description and price.

Consistent with the results of this study, research on the effect of linear and non-linear pricing on ordering decisions has been counterintuitive, with studies showing that behavior is not affected by the type of pricing used.19,20,23,31,32 Although this study did not find interactions between portion-size descriptors and pricing, it will be important for future research to examine interactions between pricing and item naming in real-world settings, and how such interventions impact restaurants’ bottom line.

Across all conditions, the most common reasons that participants chose entrées were that the item sounded delicious and they perceived the item to be a healthy choice. The least common reason selected across all conditions was that the calories seemed right for them. These results suggest that regardless of pricing structure and portion-size descriptor, the characteristics that drive consumer choice when dining out is the perceived deliciousness and health of an item. With those two characteristics in place, consumers can be prompted to select smaller portion sizes, which inherently do not change whether the dish is delicious or healthy.

In our exploratory analyses, compared to those with a high school degree, participants with higher educational attainment, which was used as a measure of socioeconomic status (SES), had increased odds of ordering reduced portions in the full-service setting when they viewed the “no descriptor and ‘Hearty’” menu compared to the control menu, even though this condition did not have an effect in the primary analyses. While it is possible that lower SES consumers may be more drawn to larger sizes, our results show that naming the smaller size “Standard” or “Just Right” can overcome the propensity to select the larger size. These results also highlight the importance of testing seemingly similar wording, especially when viewed through a health equity lens. Implementing “no descriptor and ‘Hearty’” portion naming could potentially widen health disparities among lower- and higher-educated populations. Future research with appropriate statistical power should further elucidate the potential role of education in determining the efficacy of portion-size descriptors.

Many fast-casual (e.g., Au Bon Pain, Panera) and full-service restaurants (e.g., Olive Garden, P.F. Chang’s, Cheesecake Factory) offer items in two sizes for at least a portion of their menus. Restaurants that offer more than two sizes tend to be fast food restaurants, which were not the focus of this paper. Renaming the reduced portion as “standard” or “just right” can be a practice adopted by restaurants in order to recalibrate norms around portion sizes, though consumers may be initially confused why the named “standard” size is smaller than what they are accustomed to. However, restaurants do not currently use the term “standard” on their menus, so consumers may not have preconceived notions as to what “standard” portions look like in the restaurant setting. This intervention would thus introduce a new term to restaurant menus rather than redefine an existing term. Though it is possible that the intervention may lose its effect over time, it may also lead some consumers going back to the same restaurant to form new ordering habits around selecting smaller portion sizes.

This study had several strengths. To the authors’ knowledge, this was the first study to examine the efficacy of changing the description of reduced portion entrées to influence what consumers ordered when dining out. It is also one of the few studies to assess the impact of pricing on ordering reduced sizes. It considered the importance of linear versus non-linear pricing when using these descriptors to address whether such naming practices would work in the context of non-linear pricing, which is often used in the restaurant industry. Finally, this study used a randomized, controlled design with a large sample size.

Limitations

This study also had several limitations. First, as this was an online hypothetical study testing a single exposure to mock menus, it may not reflect repeated ordering behavior from a real restaurant menu using real money. However, only participants who indicated that they either ate out or ordered online at least twice in the past four weeks were recruited, suggesting familiarity with both dining out and/or using an online platform to order, and the menus were modeled after real menus. In addition, the calories ordered in hypothetical scenarios are often in line with calories ordered in real world studies.33,34 Second, this study did not measure consumption. It is possible that participants who ordered the larger size in this study did not intend to finish the entire entrée as one meal, in which case, the portion size consumed may have been appropriate, though people typically overeat with larger portions.7 Third, over 80% of our participants identified as White and the median age was high (67 years) compared to the median age of the U.S. population, limiting the generalizability of our results. Although this is a limitation, to the authors’ knowledge, there is no existing literature to suggest that these menu descriptions might differentially influence people on the basis of race/ethnicity, though research has shown that older adults have decreased appetites.35 Fourth, this intervention did not account for dietary quality, though reducing caloric intake is another important aspect of chronic disease prevention.36 Future studies should assess the effects of repeated ordering from menus with modified portion-size descriptors and linear/non-linear pricing schemes in real-world restaurant settings. More research is also needed to better understand the mechanism of the observed effect. Finally, it will be important to test these interventions on populations with diverse sociodemographic characteristics.

Conclusions

Study results suggest restaurants can voluntarily alter their menus with this low-cost intervention that preserves customer choice while encouraging healthier choices. It may also reduce food waste because consumers can now order appropriately-sized portions. Finally, it may increase profits from an expanded customer market that is interested in more reasonable portion sizes.

Supplementary Material

1

Acknowledgements

Dr. Hua conceptualized and designed the study, designed the data collection instruments, collected and analyzed the data, drafted the initial manuscript, reviewed and revised the manuscript, and acquired funding. Dr. Kenney provided critical feedback on study design, data analytic plan, and interpretation of the data, and critically reviewed the manuscript for important intellectual content. Mr. Miller provided critical feedback on conceptualization of the study and critically reviewed the manuscript for important intellectual content. Dr. Musicus provided critical feedback on data collection instruments and critically reviewed the manuscript for important intellectual content. Drs. Roberto and Thorndike provided critical feedback on study design, data analytic plan, and interpretation of the data, and critically reviewed the manuscript for important intellectual content. Dr. Rimm provided critical feedback on study design, data analytic plan, and interpretation of the data, critically reviewed the manuscript for important intellectual content, and acquired funding. All authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work.

Funding Sources:

This study was partially funded by the Harvard Center for Population and Development Studies. Dr. Hua is supported by the NIH National Research Service Award (T32 DK 007703). Dr. Musicus is supported by NIH grants 2T32CA057711 and T32HL098048. The content of this article is solely the responsibility of the authors and does not necessarily represent the official views of the NIH.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Contributor Roles Taxonomy (CRediT)

Sophia V. Hua: Conceptualization; Methodology; Formal analysis; Resources; Writing - original draft; Writing - review & editing; Funding acquisition

Erica L. Kenney: Methodology; Writing - review & editing

Jeffrey M. Miller: Conceptualization; Writing - review & editing

Aviva A. Musicus: Methodology; Writing - review & editing

Christina A. Roberto: Methodology; Writing - review & editing

Anne N. Thorndike: Methodology; Writing - review & editing

Eric B. Rimm: Conceptualization; Methodology; Writing - review & editing; Supervision; Funding acquisition

Conflict of Interest: There are no conflicts of interests to report.

Financial Disclosure: Mr. Miller advises suppliers and commodity boards on how to create foodservice strategies. No other authors have financial disclosures to report.

References

  • 1.Elitzak H, Okrent A. New U.S. food expenditure estimates find food-away-from-home spending Is higher than previous estimates. 2018. https://www.ers.usda.gov/amber-waves/2018/november/new-us-food-expenditure-estimates-find-food-away-from-home-spending-is-higher-than-previous-estimates/
  • 2.Young LR, Nestle M. Portion Sizes of Ultra-Processed Foods in the United States, 2002 to 2021. Am J Public Health. Dec 2021;111(12):2223–2226. doi: 10.2105/AJPH.2021.306513 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Piernas C, Popkin BM. Food portion patterns and trends among U.S. children and the relationship to total eating occasion size, 1977–2006. J Nutr. Jun 2011;141(6):1159–64. doi: 10.3945/jn.111.138727 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Livingstone MB, Pourshahidi LK. Portion size and obesity. Adv Nutr. Nov 2014;5(6):829–34. doi: 10.3945/an.114.007104 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Bezerra IN, Curioni C, Sichieri R. Association between eating out of home and body weight. Nutr Rev. Feb 2012;70(2):65–79. doi: 10.1111/j.1753-4887.2011.00459.x [DOI] [PubMed] [Google Scholar]
  • 6.Auchincloss AH, Leonberg BL, Glanz K, Bellitz S, Ricchezza A, Jervis A. Nutritional value of meals at full-service restaurant chains. J Nutr Educ Behav. Jan 2014;46(1):75–81. doi: 10.1016/j.jneb.2013.10.008 [DOI] [PubMed] [Google Scholar]
  • 7.Hollands GJ, Shemilt I, Marteau TM, et al. Portion, package or tableware size for changing selection and consumption of food, alcohol and tobacco. Cochrane Database Syst Rev. Sep 14 2015;(9):CD011045. doi: 10.1002/14651858.CD011045.pub2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Geier AB, Rozin P, Doros G. Unit bias. A new heuristic that helps explain the effect of portion size on food intake. Psychol Sci. Jun 2006;17(6):521–5. doi: 10.1111/j.1467-9280.2006.01738.x [DOI] [PubMed] [Google Scholar]
  • 9.Marteau TM, Hollands GJ, Shemilt I, Jebb SA. Downsizing: policy options to reduce portion sizes to help tackle obesity. BMJ. Dec 2 2015;351:h5863. doi: 10.1136/bmj.h5863 [DOI] [PubMed] [Google Scholar]
  • 10.Zlatevska N, Dubelaar C, Holden SS. Sizing up the Effect of Portion Size on Consumption: A Meta-Analytic Review. Journal of Marketing. 2014/May/01 2014;78(3):140–154. doi: 10.1509/jm.12.0303 [DOI] [Google Scholar]
  • 11.Rolls BJ, Roe LS, Meengs JS. Larger portion sizes lead to a sustained increase in energy intake over 2 days. J Am Diet Assoc. Apr 2006;106(4):543–9. doi: 10.1016/j.jada.2006.01.014 [DOI] [PubMed] [Google Scholar]
  • 12.Levitsky D, Agaronnik N, Zhong W, Morace C, Barre L, Michael JJ. Reducing an entree portion size does not affect the amount of dessert consumed. Appetite. Aug 1 2020;151:104684. doi: 10.1016/j.appet.2020.104684 [DOI] [PubMed] [Google Scholar]
  • 13.Reinders MJ, Huitink M, Dijkstra SC, Maaskant AJ, Heijnen J. Menu-engineering in restaurants - adapting portion sizes on plates to enhance vegetable consumption: a real-life experiment. Int J Behav Nutr Phys Act. Dec 25 2017;14(1):41. doi: 10.1186/s12966-017-0496-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Vermote M, Versele V, Stok M, et al. The effect of a portion size intervention on French fries consumption, plate waste, satiety and compensatory caloric intake: an on-campus restaurant experiment. Nutrition Journal. 2018;17. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Sim AY, Cheon BK. Influence of impending healthy food consumption on snacking: Nudging vs. compensatory behaviour. Physiol Behav. Jan 1 2019;198:48–56. doi: 10.1016/j.physbeh.2018.10.010 [DOI] [PubMed] [Google Scholar]
  • 16.Schwartz J, Riis J, Elbel B, Ariely D. Inviting consumers to downsize fast-food portions significantly reduces calorie consumption. Health Aff (Millwood). Feb 2012;31(2):399–407. doi: 10.1377/hlthaff.2011.0224 [DOI] [PubMed] [Google Scholar]
  • 17.Hua SV, Sterner-Stein K, Barg FK, et al. A Qualitative Study of Parents With Children 6 to 12 Years Old: Use of Restaurant Calorie Labels to Inform the Development of a Messaging Campaign. J Acad Nutr Diet. Nov 2020;120(11):1884–1892 e4. doi: 10.1016/j.jand.2020.05.018 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Dobson PW, Gerstner E. For a Few Cents More: Why Supersize Unhealthy Food? Marketing Science. 2010/July/01 2010;29(4):770–778. doi: 10.1287/mksc.1100.0558 [DOI] [Google Scholar]
  • 19.Haws KL, Liu PJ. Half-size me? How calorie and price information influence ordering on restaurant menus with both half and full entree portion sizes. Appetite. Feb 1 2016;97:127–37. doi: 10.1016/j.appet.2015.11.031 [DOI] [PubMed] [Google Scholar]
  • 20.Harnack LJ, French SA, Oakes JM, Story MT, Jeffery RW, Rydell SA. Effects of calorie labeling and value size pricing on fast food meal choices: results from an experimental trial. Int J Behav Nutr Phys Act. Dec 5 2008;5:63. doi: 10.1186/1479-5868-5-63 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.John LK, Donnelly GE, Roberto CA. Psychologically Informed Implementations of Sugary-Drink Portion Limits. Psychol Sci. May 2017;28(5):620–629. doi: 10.1177/0956797617692041 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Zuraikat FM, Roe LS, Smethers AD, Reihart LW, Rolls BJ. Does the cost of a meal influence the portion size effect? Appetite. Aug 1 2018;127:341–348. doi: 10.1016/j.appet.2018.05.020 [DOI] [PubMed] [Google Scholar]
  • 23.John LK, Donnelly GE, Roberto CA. Using Behavioral Science To Inform Policies Limiting Sugary-Drink Portions: Reply to Wilson and Stolarz-Fantino (2018). Psychol Sci. Jul 2019;30(7):1103–1105. doi: 10.1177/0956797619851731 [DOI] [PubMed] [Google Scholar]
  • 24.Kansagra SM, Kennelly MO, Nonas CA, et al. Reducing sugary drink consumption: New York City’s approach. Am J Public Health. Apr 2015;105(4):e61–4. doi: 10.2105/AJPH.2014.302497 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Berkowitz AD. An Overview of the Social Norms Approach. Accessed May 6, 2022, http://alanberkowitz.com/articles/social%20norms%20approach-short.pdf
  • 26.Educational Attainment in the United States: 2020. United States Census Bureau. Accessed Dec. 10, 2021, https://www.census.gov/data/tables/2020/demo/educational-attainment/cps-detailed-tables.html
  • 27.Chipotle. Accessed Janurary 28, 2021, https://www.chipotle.com/
  • 28.Blaze Pizza. Accessed Janurary 28, 2021, https://www.blazepizza.com/
  • 29.Schwartz LM, Woloshin S, Black WC, Welch HG. The role of numeracy in understanding the benefit of screening mammography. Ann Intern Med. Dec 1 1997;127(11):966–72. doi: 10.7326/0003-4819-127-11-199712010-00003 [DOI] [PubMed] [Google Scholar]
  • 30.Hess R, Visschers VH, Siegrist M. The role of health-related, motivational and sociodemographic aspects in predicting food label use: a comprehensive study. Public Health Nutr. Mar 2012;15(3):407–14. doi: 10.1017/S136898001100156X [DOI] [PubMed] [Google Scholar]
  • 31.Haws KL, Liu PJ, Dallas SK, Cawley J, Roberto CA. Any Size for a Dollar: The Effect of Any-Size-Same-Price Versus Standard Pricing on Beverage Size Choices. Journal of Consumer Psychology. 2019;30(2):392–401. doi: 10.1002/jcpy.1129 [DOI] [Google Scholar]
  • 32.Ito K Do Consumers Respond to Marginal or Average Price? Evidence from Nonlinear Electricity Pricing. American Economic Review. 2014;104(2):537–63. doi: 10.1257/aer.104.2.537 [DOI] [Google Scholar]
  • 33.Musicus AA, Hua SV, Schwartz MB, et al. Messages Promoting Healthy Kids’ Meals: An Online RCT. Am J Prev Med. May 2021;60(5):674–683. doi: 10.1016/j.amepre.2020.11.012 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Tandon PS, Zhou C, Chan NL, et al. The impact of menu labeling on fast-food purchases for children and parents. Am J Prev Med. Oct 2011;41(4):434–8. doi: 10.1016/j.amepre.2011.06.033 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Pilgrim AL, Robinson SM, Sayer AA, Roberts HC. An overview of appetite decline in older people. Nurs Older People. Jun 2015;27(5):29–35. doi: 10.7748/nop.27.5.29.e697 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Guth E Counting Calories as an Approach to Achieve Weight Control. JAMA. Jan 16 2018;319(3):225–226. doi: 10.1001/jama.2017.21355 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS1868778-supplement-1.pdf (713.5KB, pdf)

RESOURCES