This paper is only available as a PDF. To read, Please Download here.
Background
We developed the PortionSize™ app to provide users with real-time feedback on dietary intake including energy intake (kcal) and compliance to the United States Department of Agriculture (USDA) MyPlate recommendations. PortionSize relies on emerging technology (e.g., augmented reality) for portion size (gram weight) estimation. Currently, PortionSize contains a database of 1150 food items and each food item is linked to a food code in the USDA Food and Nutrition Database for Dietary Studies for food group estimates.
Objective
This pilot study assessed the validity of PortionSize for real-time estimates of portion size, energy intake, and MyPlate food group servings (fruits, vegetables, grains, dairy, and protein).
Study Design, Setting, Participants
Adults (4 male, 11 female), aged 18-65 years, were trained, and used PortionSize to quantify food intake from simulated meals, which were covertly weighed, in a laboratory setting in Baton Rouge, LA.
Measurable Outcome/Analysis
Equivalence tests (± 25% equivalence bounds) were performed to compare PortionSize and weigh-back estimates of portion size, energy intake, and food group servings.
Results
PortionSize and weigh-back were equivalent for mean (±SD) estimates of portion size (674±223 g vs. 717±207 g, respectively), total fruit servings (0.2±0.3 vs. 0.3±0.4), and total dairy servings (0.4±0.6 vs. 0.6±0.4) (all P values <0.05). PortionSize and weigh-back estimates were not equivalent for total energy intake (743±328 kcal vs. 659±191 kcal), total vegetable servings (0.9±1.0 vs. 0.6±0.4), total grain servings (1.7±1.7 vs. 1.2±1.1), and total protein servings (3.1±3.6 vs. 2.8±2.9) (all P values >0.05).
Conclusions
PortionSize shows promise for real-time estimation of portion size, and MyPlate servings of total fruits and total dairy. PortionSize requires further development and validity testing for real-time estimation of total vegetables, total grains, and total protein servings, which may assist with improving the validity of total energy estimates.
Funding
NIH.
Article info
Identification
Copyright
© 2022 Published by Elsevier Inc.
