The influence of front-of-pack nutrition information on consumers' portion size perceptions

Which Front-of-Package Nutrition Label Is Better? The Influence of Front-of-Package Nutrition Label Type on Consumers' Healthy Food Purchase Behavior

To help consumers understand the healthfulness of food and make healthy food choices, manufacturers are increasingly providing front-of-package nutrition labels. However, not all types of front-of-package nutrition labels can promote consumers' healthy food purchase behavior. We explored the impact of front-of-package nutrition label type on the consumer purchase behavior of healthy food through three experiments. The results show that evaluative (vs. objective) front-of-package nutrition labels can improve consumer purchase intention and willingness to pay for healthy food. The spokesperson type moderates the influence of front-of-package nutrition labels on consumers' healthy food purchase behavior. Specifically, when the spokesperson type is a typical consumer, consumers are more willing to buy healthy foods with evaluative nutrition labels than those with objective nutrition labels. When the spokesperson type is a star, consumers are more willing to buy healthy food with objective nutrition labels (vs. evaluative nutrition labels). Finally, this study provides feasible suggestions for marketers to select appropriate front-of-package nutrition labels.

Systematic review of the portion size norm of discretionary foods

CONTEXT

Portion size norm is described as the perception of how much of a given food people choose to eat. Reducing the portion size norm of foods that are high in saturated fat, added sugar, and added salt toward smaller sizes might be a potential strategy to promote appropriate portion size selections. However, an overview of existing portion size norms for discretionary foods has yet to be established.

OBJECTIVE

The aim of this systematic review was to examine the portion size norm of discretionary foods and assess the methodologies used to investigate the norm.

DATA SOURCES

The literature search was conducted in 6 databases following the PRISMA guidelines (from inception to January 2022).

DATA EXTRACTION

Forty studies were eligible and grouped into 3 categories by portion size norm measures: normal (n = 26), appropriate (n = 8), and preferred portion sizes (n = 3). Study quality was assessed using the Joanna Briggs Institute Critical Appraisal Tools.

DATA ANALYSIS

A wide range of portion sizes were considered normal for each food type, with means/medians varying from 2- to 4-fold among studies. Studies differed considerably in design, with variables including the setting, food type, food presentation, the manner in which portion-size-related questions were formulated, and the range and number of displayed serving size options. The quality of reviewed studies was mixed (25 studies had low or moderate risk of bias, 15 had high risk of bias), and the method of assessing portion size was not validated in 15 of 33 quantitative studies.

CONCLUSION

The assessment of portion size in future studies should be conducted using tools that are validated for the population of interest so that more definitive conclusions can be drawn regarding portion size norms for discretionary foods.

SYSTEMATIC REVIEW REGISTRATION

PROSPERO registration number CRD42021249911.

Socioeconomic position, energy labelling and portion size selection: An online study comparing calorie and physical activity calorie equivalent (PACE) labelling in UK adults

Limited research has examined the impact of energy labelling on portion size selection. It is also unclear whether physical activity calorie equivalent (PACE) is more effective than standard kilocalorie (kcal) energy labelling in promoting healthier dietary behaviour and whether effectiveness varies based on socioeconomic position (SEP). In the present online study, 1667 UK adults of lower and higher SEP made virtual portion size selections for 18 common main meal foods under one of four conditions: kcal labelling only, PACE labelling only, kcal and PACE labelling, no labelling. Contrary to predictions, participants in the kcal labelling condition (+55 kcal, p < 0.001) chose larger portion sizes compared to the no labelling condition, whereas the PACE labelling (−17 kcal, p = 0.065) and no labelling condition did not significantly differ. The presence of PACE information on labels was associated with selection of significantly smaller portions when compared to labels that only included kcal information. Effects of labels on portion size selection were not moderated by participant SEP in primary analyses. The present study of virtual portion size selections suggests that kcal labelling resulted in larger portion size selections than no labelling, but this counter-intuitive effect was attenuated when kcal and PACE labelling were combined. Further research examining the impact of PACE labelling on real-world food selection in participants of lower and higher SEP is now warranted.

Mixed deep learning and natural language processing method for fake-food image recognition and standardization to help automated dietary assessment

OBJECTIVE

The present study tested the combination of an established and a validated food-choice research method (the 'fake food buffet') with a new food-matching technology to automate the data collection and analysis.

DESIGN

The methodology combines fake-food image recognition using deep learning and food matching and standardization based on natural language processing. The former is specific because it uses a single deep learning network to perform both the segmentation and the classification at the pixel level of the image. To assess its performance, measures based on the standard pixel accuracy and Intersection over Union were applied. Food matching firstly describes each of the recognized food items in the image and then matches the food items with their compositional data, considering both their food names and their descriptors.

RESULTS

The final accuracy of the deep learning model trained on fake-food images acquired by 124 study participants and providing fifty-five food classes was 92·18 %, while the food matching was performed with a classification accuracy of 93 %.

CONCLUSIONS

The present findings are a step towards automating dietary assessment and food-choice research. The methodology outperforms other approaches in pixel accuracy, and since it is the first automatic solution for recognizing the images of fake foods, the results could be used as a baseline for possible future studies. As the approach enables a semi-automatic description of recognized food items (e.g. with respect to FoodEx2), these can be linked to any food composition database that applies the same classification and description system.

Impact of Front-of-Pack Nutrition Labels on Portion Size Selection: An Experimental Study in a French Cohort

In the European Union (EU) three coloured graded Front-of-Pack labels (FoPLs), two endorsed by governments (Nutri-Score and Multiple Traffic Lights (MTL)) and one designed by industry (Evolved Nutrition Label (ENL)) are currently being discussed. This study aimed to investigate the impact of these FoPLs on portion size selection, specifically for less healthy products. In 2018, participants from the French NutriNet-Santé cohort study (N = 25,772) were exposed through a web-based self-administered questionnaire to products from three food categories (sweet biscuits, cheeses, and sweet spreads), with or without FoPLs, and were invited to select the portion they would consume (in size and number). Kruskall-Wallis tests, and mixed ordinal logistic regression models, were used to investigate the effects of FoPLs on portion size selection. Compared to no label, Nutri-Score consistently lowered portion sizes (OR = 0.76 (0.74⁻0.76)), followed by MTL (OR = 0.83 (0.82⁻0.84)). For ENL, the effects differed depending on the food group: It lowered portion size selection for cheeses (OR = 0.84 (0.83⁻0.87)), and increased it for spreads (OR = 1.19 (1.15⁻1.22)). Nutri-Score followed by MTL appear efficient tools to encourage consumers to decrease their portion size for less healthy products, while ENL appears to have inconsistent effects depending on the food category.

Can front-of-pack labels influence portion size judgements for unhealthy foods?

OBJECTIVE

By clearly conveying the healthiness of a food, front-of-pack (FOP) labels have the potential to influence the portion size considered appropriate for consumption. The present study examined the how the Daily Intake Guide (DIG), Multiple Traffic Lights (MTL) and Health Star Rating (HSR) FOP labels affect judgements of appropriate portion sizes of unhealthy foods compared with when no FOP label is present.

DESIGN

Respondents viewed mock packages of unhealthy variations of pizzas, cookies, yoghurts and cornflakes featuring the DIG, MTL, HSR or no FOP label, and indicated the portion size they believed should be eaten of each food on a single occasion.

SETTING

The survey was completed on the respondent's personal computer.

SUBJECTS

A total of 1505 Australian adults provided 4166 ratings across 192 mock packages relating to four product categories: pizza, yoghurt, cornflakes and cookies.

RESULTS

Compared with no FOP label, the HSR resulted in a small but significant reduction in the portion size selected as appropriate for consumption of pizzas and cornflakes (P<0·05). The MTL resulted in smaller portions of cornflakes being selected compared with no FOP label (P<0·05).

CONCLUSIONS

Respondents perceived smaller portion sizes as appropriate for some, but not all, of the foods tested when FOP labels with more interpretative formats (HSR, MTL) appeared on-pack compared with no FOP label. No effect was found for the less interpretive FOP label (the DIG). Interpretive FOP labels may have the potential to influence portion size judgements, albeit at modest levels.