Underconsumption of dietary fiber is prevalent worldwide and is associated with multiple adverse health conditions. Despite the importance of fiber, the labeling of fiber content on packaged foods and beverages is voluntary in most countries, making it challenging for consumers and policy makers to monitor fiber consumption. Here, we developed a machine learning approach for automated and systematic prediction of fiber content using nutrient information commonly available on packaged products. An Australian packaged food dataset with known fiber content information was divided into training (n = 8986) and test datasets (n = 2455). Utilization of a k-nearest neighbors machine learning algorithm explained a greater proportion of variance in fiber content than an existing manual fiber prediction approach (R² = 0.84 vs. R² = 0.68). Our findings highlight the opportunity to use machine learning to efficiently predict the fiber content of packaged products on a large scale.

The presentation of nutrition information on a serving size basis is a strategy that has been adopted by several countries to promote healthy eating. Variation in serving size, however, can alter the nutritional values reported on food labels and compromise the food choices made by the population. This narrative review aimed to discuss (1) current nutrition labelling legislation regarding serving size and (2) the implications of declared serving size for nutrition information available on packaged foods. Most countries with mandatory food labelling require that serving size be presented on food labels, but variation in this information is generally allowed. Studies have reported a lack of standardisation among serving sizes of similar products which may compromise the usability of nutrition information. Moreover, studies indicate that food companies may be varying serving sizes as a marketing strategy to stimulate sales by reporting lower values of certain nutrients or lower energy values on nutrition information labels. There is a need to define the best format for presenting serving size on food labels in order to provide clear and easily comprehensible nutrition information to the consumer.

Context: Reducing population intakes of sugar has become a focus of many national and international public health policies. Packaged foods and beverages are key contributors to sugar intakes, and food labels can be an effective tool to reduce sugar consumption. Objective: The aim of this systematic review was to examine the influence of sugar label formats on 2 outcomes: consumers’ understanding of sugar information, and the amount of sugar in consumers’ food choices. Data Sources: Scopus, Web of Science, PubMed, CAB Abstracts, SciELO, and the Cochrane Library databases were searched up until February 4, 2020. Study Selection: Randomized experiments or quasi-experiments were included if they investigated the influence of sugar label formats on consumers’ understanding of sugar information or on the amount of sugar in consumers’ food choices. Data Extraction: Data were extracted independently by 2 authors. Mean differences (MDs), standardized mean differences (SMDs), and odds ratios (ORs) plus 95%CIs were used to describe between-group differences for intervention label formats using random-effects models. Results: Twenty-three studies, which examined 39 comparisons, were included. Label formats using “high in sugar” interpretative texts (traffic light labels [MD 41.6; 95%CI 37.9–45.4] and warning signs [OR 1.33; 95%CI 1.0–1.78]) were most effective in increasing consumers’ understanding of the sugar content in packaged foods. Health warning messages (SMD −0.32; 95%CI −0.43 to −0.22), graphical depictions of sugar content in teaspoons (SMD −0.32; 95%CI −0.48 to −0.17), and warning signs (SMD −0.24; 95%CI −0.35 to −0.13) were most effective for influencing consumers to choose products with lower sugar content. Conclusions: Formats that provide an interpretation of sugar information, particularly those indicating if a product is high in sugar, were more helpful than only numerical information for improving consumer understanding and promoting food choices with less sugar. Systematic Review Registration: PROSPERO registration number CRD42018081222.

Retail food environments influence food purchasing and dietary patterns. A global analysis of the food retail landscape allowing comparisons across geographical regions is therefore needed to tackle diet-related non-communicable diseases. Here we examine trends in retail food environments from 2009 to 2023 across 97 countries, exploring associations with changes in obesity prevalence. Increases were observed in the density of chain outlets, grocery sales from chain retailers, unhealthy food sales per capita and digital grocery sales; non-chain outlet density and the ratio of non-chain to chain outlets declined over time. South Asia and low- and middle-income countries overall experienced the most rapid transformation. Changes in retail environments and the prevalence of obesity were found to be positively correlated. As retail environments become increasingly digital and dominated by large chains, important implications for diets and health should be expected, particularly in lower-income countries.

Online food delivery services (OFDS) are popular for purchasing meals prepared outside home, increasing access to energy-dense and nutrient-poor foods. This adversely impacts dietary choices and health outcomes. Our study examined trends in OFDS use in Australia, Canada, Mexico, the United Kingdom (UK), and the United States (US) from 2018 to 2021. Repeated annual cross-sectional data was sourced from the International Food Policy Study for five countries among adults over 18 years (N = 83,337). Weighted estimates for trends in i) the proportion of the respondent's purchasing meals per week using OFDS, and ii) average number (and standard deviation (SD)) of meals purchased per week using OFDS were assessed. Logistic regression models were fitted. OFDS use increased among adults between 2018-2021 (Australia: 17 % of respondents purchased at least one meal in the last 7 days using OFDS in 2018 to 25 % in 2021, Canada: 12 % to 19 %, Mexico: 28 % to 38 %, UK: 19 % to 28 %, and US: 17 % to 21 %). Average number of meals purchased per week outside home remained consistent for all countries over time (e.g., in Australia, 2.70 (SD 0.06) meals in 2018 and 2.63 (SD 0.06) in 2021). However, average number of meals purchased using OFDS nearly doubled between 2018 and 2021 (e.g., in Australia, 0.45 (SD 0.03) meals in 2018 to 0.81 (SD 0.04) in 2021). OFDS use is increasing and are substituting the conventional forms of purchasing meals outside home. Nutritional quality of foods sold, marketing practices and purchasing patterns on OFDS deserve further attention.

Objective: Studies have shown that consumption of low-calorie sweeteners (LCS) may be associated with harmful health effects. The current study investigated the presence and types of LCS added to packaged foods.Design: Cross-sectional study analysing the presence and types of LCS in the ingredients lists of packaged foods sold at a major Brazilian supermarket. To identify types of LCS allowed for use in foods in Brazil, current legislation was consulted. Data were organised and analysed through descriptive statistics, presenting simple and relative frequencies of LCS presence categorised by food group.Setting: Supermarket in Florianópolis, southern Brazil.Subjects: Packaged food products (n4539) from eight food groups.Results: One or more LCS were found in 602 (13·3 %) of the packaged foods analysed. There were 1329 citations of LCS among these foods, with a mean of 2·2 sweeteners per food. Groups with the highest frequency of foods containing LCS were: products with energy derived from carbohydrates and fats (25·0 %); milk and dairy products (11·7 %); bakery products, cereals, legumes, roots and tubers (11·2 %); and fruits, juices, nectars and fruit drinks (8·3 %).Conclusions: There was high prevalence of packaged foods with LCS, especially in food groups that form the basis of the Brazilian diet. The study was the first to extensively analyse the presence and types of LCS in packaged foods available for sale in a Brazilian supermarket and can be useful to monitor the use of LCS in these foods, as well as to support future changes in legislation to label sugars.