Inter-individual differences in children's short-term energy compensation: a systematic review and meta-analysis

The ability to regulate energy intake is often assessed using a preloading paradigm to measure short-term energy compensation. In children, large variability exists with this paradigm both within- and across- studies and is poorly understood.

This systematic review and meta-analysis aimed to better understand factors contributing to variability in children's energy compensation. We tested 1) whether children demonstrated "good" energy compensation, defined as adjusting meal intake commensurate with preload intake and 2) differences in children's energy compensation by child age, sex, and weight status (assessed both continuously and categorically).

Standard guidelines for systematic review were followed to search PubMed, PsychInfo, and Web of Science. Data on study design (preload form, preload-to-meal interval, preload energy difference, study setting) and participant characteristics (sex, age, weight status) were extracted from 29 experiments meeting inclusion criteria, and 13 were included in meta-analyses. COMPx (energy compensation index), a linear transformation comparing food intake following a high- vs. low-energy preload, was the outcome. Hedge's g was calculated, and random intercept-only models tested associations between COMPx and sex, age, and weight status.

The systematic review revealed mixed results regarding children's energy compensation and the role of inter-individual differences. Meta-analytic models revealed that children undercompensated (overate) for preload energy (β = -0.38; P = 0.008). Sex (β = 0.11; P = 0.76), age (β = 0.03; P = 0.75), and weight (assessed continuously; β = -0.07, P = 0.37) were not related to compensation. Children with overweight/obesity (assessed categorically) undercompensated more than children with healthy weight (β = 0.18; P = 0.04).

The systematic review highlighted wide variability across studies, while the meta-analysis demonstrated differences in COMPx by child weight status but not by age or sex. Standardizing protocols across studies is recommended, along with designing adequately powered studies aiming to test inter-individual differences a priori. Alternative approaches to the use of COMPx are recommended to allow better characterization of children's energy compensation ability. This study was registered at PROSPERO as CRD42020197748.