Inconsistency between Self-Reported Energy Intake and Body Mass Index among Urban, African-American Children

BACKGROUND To prevent obesity, it is important to assess dietary habits through self-reported energy intake (EI) in children. We investigated how EI is associated with body mass index and which elements of dietary habits and status are associated with EI among African-American (AA) children. METHODS We assessed and included data from 218 10-14-year-old AA children in Baltimore, MD, USA. EI was calculated using a food frequency questionnaire. The basal metabolic rate (BMR) was used as the predicted minimal rate of energy expenditure of children. A fully adjusted multiple logistic regression was used to determine the prevalence of obesity (≥ 95th BMI-for-age percentile) among the quartiles of EI/BMR ratio using the third quartile for the reference. The differences in the age-adjusted mean EI/BMR among the categories of dietary habits, social support, and socio economic status were analyzed using a general linear model. RESULTS Children with the lowest EI/BMR had significantly higher adjusted odds ratio (aOR) of obesity as compared to those in the third quartile of EI/BMR (boys aOR 4.3; 95% confidence interval 1.08, 20 and girls aOR 4.1; 1.02, 21). In girls, the adjusted mean EI/BMR in the group that prepared food less than the means (3.8 times/week) was significantly lower than the group that prepared food over the means (P = 0.03). Further, the group that reported eating breakfast under 4 times/week indicated an adjusted mean EI/BMR lower than the group that ate breakfast over 5 times/week in both sexes. CONCLUSIONS When EI was under-reported with reference to BMR, we may observe high prevalence of obesity. Further, food preparation by children and frequent consumption of breakfast may instill food cognition with usual dietary habits. Therefore, holistic assessments including dietary habits are required to examine self-reported food intake especially among overweight/obese children.