A simple method for identification of misreporting of energy intake from infancy to school age: Results from a...

Misreporting is a major source of reporting bias in nutritional surveys. It can affect the analysis of associations between diet and disease. Although various methods have been proposed to identify misreporting, their application to infants and young children is difficult. We identify misreporting of energy intake in infants and young children and propose a simplified approach.

1199 children were enrolled in the Childhood Obesity Programme (CHOP) based in 5 European countries (Belgium, Germany, Italy, Poland and Spain) with repeated measurements of 3-day weighed food protocol and anthropometric indices at 10 time points between ages 1-96 months. Individual cut-offs for the ratio of reported energy intake and estimated energy requirement were calculated to identify misreporters. Misreporting was studied according to age, gender, BMI z-scores and country.

We identified a higher proportion of over-reporters (18.9%) as compared to under-reporters (10.6%). The proportion of over-reporting was higher among infants while under-reporting was more prevalent in school-aged children. Under-reporting was higher in boys (12.0%) and in obese/over-weight children (36.3%). Mean values for upper and lower cut-offs for the ratio of reported energy intake and estimated energy requirement in children ≤12 months were 0.80 and 1.20, and 0.75 and 1.25 for children >12 months, respectively. Using these fixed (mean) values, 90.4% (kappa statistic: 0.78) of all misreporters could be identified.

Despite intensive measures to obtain habitual intake of children, an essential proportion of nutritional reports were found to be implausible. Both over- and under-reporting should be carefully analysed, even in studies on infants. Fixed cut-offs can be applied to identify misreporting if no individual variation in energy intake can be calculated.

This trial was registered at https://clinicaltrials.gov/show/NCT00338689.