Related Experiment Video
Updated: Jun 4, 2025

Iterative Development of an Innovative Smartphone-Based Dietary Assessment Tool: Traqq
Published on: March 19, 2021
Method to define recommended portion sizes for consumer guidance
Alison L Eldridge1,2, Evangelia Kotzakioulafi3, Charlotte Debras3
1Nestlé Institute of Health Sciences, Nestlé Research, Vers-chez-les-Blanc, Lausanne, Switzerland. alison.eldridge@rdls.nestle.com.
Purpose:
Provision of nutrition information is mandated for packaged foods, but few countries regulate serving sizes. Our objective was to develop a methodology to establish globally consistent portion size recommendations for both nutrient-dense and discretionary foods.
Methods:
A stepwise systematic approach incorporated portion values from serving size regulations (n = 10), food-based dietary guidelines (FBDG; n = 90, aggregated into 6 regions), and reported food intakes from Europe and Australia. Global Portion Values (GPVs) for 50 food groups were derived as the median of all data inputs. Consistency was evaluated using 25th and 75th percentiles and dispersion of input values (percent of median absolute deviations (MAD)/median).
Results:
GPVs were calculated for 50 food groups organized into nine food categories: Milk/Dairy (n = 6 food groups), Protein Foods (n = 5), Mixed Dishes (n = 4), Grains (n = 7), Fruits/Vegetables (n = 5), Snacks/Sweets (n = 12), Sauces/Condiments (n = 5), Fats/Oils/Sugars (n = 4) and Beverages (n = 2). Data inputs for each portion value ranged from 6 to 18; only 4 foods had < 10 inputs; 21 had ≥ 15. Dispersion ranged from 0 to 33%; 88% of GPVs were considered "consistent" (dispersion < 25%) and 9 groups had 0% variation, indicating high consistency. Example GPVs include: 240mL for milk (16 inputs, 8% dispersion); 90 g for meat/poultry/fish main dishes (16 inputs, 13% dispersion); 50 g for bread/rolls (18 inputs, 20% dispersion); 130 g for canned vegetables (10 inputs, 2% dispersion); 30 g for chocolate (15 inputs, 17% dispersion); and 250mL for soft drinks (17 inputs, 20% dispersion).
Conclusions:
This standardized approach provides clear and consistent portion recommendations that, if adopted, could help consumers make informed choices about appropriate portions.

