Active Image-Assisted Food Records in Comparison to Regular Food Records: A Validation Study against Doubly Labeled Water in 12-Month-Old Infants

Abstract

Overreporting of dietary intake in infants is a problem when using food records (FR), distorting possible relationships between diet and health outcomes. Image-assisted dietary assessment may improve the accuracy, but to date, evaluation in the pediatric setting is limited. The aim of the study was to compare macronutrient and energy intake by using an active image-assisted five-day FR against a regular five-day FR, and to validate image-assistance with total energy expenditure (TEE), was measured using doubly labeled water. Participants in this validation study were 22 healthy infants randomly selected from the control group of a larger, randomized intervention trial. The parents reported the infants’ dietary intake, and supplied images of main course meals taken from standardized flat-surfaced plates before and after eating episodes. Energy and nutrient intakes were calculated separately using regular FR and image-assisted FRs. The mean (± standard deviations) energy intake (EI) was 3902 ± 476 kJ/day from the regular FR, and 3905 ± 476 kJ/day from the FR using active image-assistance. The mean EI from main-course meals when image-assistance was used did not differ (1.7 ± 55 kJ, p = 0.89) compared to regular FRs nor did the intake of macronutrients. Compared to TEE, image-assisted FR overestimated EI by 10%. Without validation, commercially available software to aid in the volume estimations, food item identification, and automation of the image processing, image-assisted methods remain a more costly and burdensome alternative to regular FRs in infants. The image-assisted method did, however, identify leftovers better than did regular FR, where such information is usually not readily available.