Background

Estimating food intake is a critical part of obesity prevention studies. Accurate and convenient methods to measure food intake among low income, minority preschool children is lacking. This study aimed to assess food intake among Head Start children in San Antonio, Texas using two plate waste methods.

Methods

Weighed and digital plate waste methods were used to estimate food intake during lunch for two consecutive days at two Head Start centers. Pre and post-consumption weight in grams and photos of each plate were collected. Pre- and post-consumption photos were used to estimate food consumption in 10% intervals by two independent researchers and percentages were multiplied by the weight of the food. NutriBase was used to analyze nutrient contents of both methods.Statistical analysis included spearman correlations and Bland-Altman regressions to assess differences between digital and weighed measurements.

Results

Data for 52 observation pairs (digital and weighed) were analyzed (69% females; age range=3-4 years old). Correlations were high among calories (.84, p<.01), protein (.84, p<.01), carbohydrates (.81, p<.01), and fats (.94, p<.01). Paired t-tests found all mean differences by method were not significant. Bland Altman regression revealed minimal mean bias with all means higher using digital method except for carbohydrates which were slightly higher among weighted (M=29.4, SD=12.8) compared to digital (M=28.94, SD=11.2; p>.05).

Conclusions

A digital method can be a convenient and accurate alternative to the weighed plate waste method used to estimate food intake in children at Head Start centers.