Efeitos dos sistemas de rotulagem nutricional frontal mexicano e brasileiro no entendimento, percepção de saudabilidade e intenções de compra a partir do uso de aplicativo para smartphones – um estudo randomizado controlado.

Abstract

The difficulty in understanding the nutrition information of foods is common among consumers. Front-of-package nutrition labeling (FoPNL) is a strategy that aims to improve the interpretation of excessive contents of critical nutrients, such as added sugars, saturated fats and sodium, through objective and easily visible language on the front panel of labels. Studies on the performance of FoPNL considering real shopping situations are scarce. In this context, the objective of these studies was to evaluate, in two moments (pilot study and main study) the FoPNL models proposed in Brazil and Mexico, through the use of an application for smartphone in a real shopping situation, considering different outcomes to help consumers in understanding nutritional information, perceived healthiness and purchase intent. The presented studies had the previous development of an application for smartphone (RotulApp) whose database had data collected from 3,335 food products sold in a supermarket chain in Belo Horizonte, MG, from during March to May, 2021. The pilot study, previously carried out with a smaller sample (230 participants), a limited product data (a dairy products aisle of the supermarket chain where the data collection took place) and application of a different questionnaire, indicated that the MS FoPNL facilitated the decision of when to buy or not to buy a product and helped the consumer to quickly decide which products to buy, considering the samples selected for the present study, when compared to the control. Considering the correct answers on identifying excessive critical nutrients, the two FoPNL models, delivered by a smartphone app, performed well. We highlight the significant effect that the evaluated models of FoPNL had on the outcomes, especially when viewed by consumers, thus emphasizing the importance of implementing this strategy in helping consumers at the supermarket. The main study included 442 participants were randomized to one of the three study arms (Brazilian System (BS) (magnifying glass model, “high in”, nutritional profile from the IN no 75/2020), Mexican System (MS) (octagon model, “excess in”, nutritional profile NOM-051) or control (nutrition facts panel and list of ingredients). Using a smartphone at a supermarket, participants scanned 1,881 products they wanted to buy (793 products in the control arm, 493 in the BS arm and 595 in the MS arm) and received information about the excessive content of added/free sugars, sodium and/or saturated fats (in 57.8% of the BS scans and in 91.4% of the MS scans) or the control. Participants answered questions regarding the outcomes of interest (identification of excessive nutrients, perceived healthiness, quick purchase decision and decision of when to buy or not to buy a product). Both evaluated FoPNL models increased the chances of success in identifying excessive nutrients in products with at least one nutrient in excess, considering the Pan American Health Organization (PAHO)’s nutrient profile model (BS: odds ratio (OR): 2.04; 95% confidence interval (95%CI): 1.46-2.86 and MS: OR: 5.44; 95%CI: 4.06-7.29) and considering the nutrient profile model of each arm as reference (BS: OR: 5.94; 95%CI: 4.18-8.45 and MS: OR: 5.46; 95%CI: 4.05-7.35). The MS FoPNL model (2.40) statistically reduced the perceived healthiness when considering all the scans, compared to the BS (2.56) and compared to the control (2.66). Both models reduced this perceived healthiness when considering only the scans of products with excessive critical nutrients, compared to the control (BS: 2.22; MS: 2.35; control: 2.63). All FoPNL models also had a better effect on the quick purchase decision compared to the control (BS: 3.63; MS: 3.66; control: 3.25) and only the BS FoPNL model had a significant effect on the decision of when to buy or not to buy a product versus the control, considering only the scans of products “high in” some critical nutrient (BS: 4.00; control: 3.50).