Energy intake and meal portions: associations with BMI percentile in U.S. children.

OBJECTIVE We examined relationships of eating patterns and reported energy intake (rEI) with BMI percentile in U.S. children. RESEARCH METHODS AND PROCEDURES Two 24-hour dietary recalls from the Continuing Surveys of Food Intakes by Individuals 1994 to 1996 and 1998 (1005 boys, 990 girls) were averaged, and children were categorized into three age groups: 3 to 5 years (n = 1077), 6 to 11 years (n = 537), and 12 to 19 years (n = 381). Physiologically implausible reports due to reporting bias or abnormal intake (rEI outside +/-18% to 23% of predicted energy requirements; pER) were identified. RESULTS rEI averaged 109 +/- 34% and 100 +/- 10% of pER in the total and plausible samples, respectively. EI was overreported more in younger children and underreported more in overweight older children. Children with plausible rEI (45.3% of sample) averaged 4.7 eating occasions/d, 589 kcal/meal, 223 kcal/snack, and 2038 kcal/d. rEI was not associated with BMI percentile in the total sample. In the plausible sample, rEI, meal portion size, and meal energy were positively associated with BMI percentile in boys 6 to 11 years and in children 12 to 19 years. No relationships were found in children 3 to 5 years and girls 6 to 11 years. Relationships were more consistent and stronger in the plausible compared with the total sample. DISCUSSION Excluding implausible dietary reports may be necessary for discerning dietary associations with BMI percentile. EI and meal, but not snack, patterns may play a quantitatively greater role in weight regulation as children age.