Assessing energy requirements is a fundamental activity in clinical dietetic practice. A study was designed to determine if published linear regression equations were accurate for predicting resting energy expenditure (REE) in fasted, obese Hispanic children (7–15 y). REE was measured using indirect calorimetry; body composition was estimated with whole-body air displacement plethysmography. REE was predicted using four equations: Institute of Medicine for healthy-weight children (IOM-HW), IOM for overweight and obese children (IOM-OS), Harris-Benedict, and Schofield. Accuracy of the prediction was calculated as the absolute value of the difference between the measured and predicted REE divided by the measured REE, expressed as a percentage. Predicted values within 85%–115% of measured were defined as accurate. Participants (n=58; 53% male) were 11.8±2.1 y (mean±SD), had 43.5±5.1% body fat and a body mass index (BMI) of 31.5±5.8 kg/m2 (98.6±1.1 BMI percentile). Measured REE was 2339±680 kcal/d; predicted REE was 1815±401 kcal/d (IOM-HW), 1794±311 kcal/d (IOM-OS), 1151±300 kcal/d (Harris-Benedict), and, 1771±316 kcal/d (Schofield). Measured REE adjusted for body weight averaged 32.0±8.4 (95% CI: 29.8, 34.2) kcal/kg per day. Published equations predicted REE within 15% accuracy for only 36–40% of 58 participants, except for Harris-Benedict, which did not achieve accuracy for any participant. The most frequently accurate values were obtained using IOM-HW, which predicted REE within 15% accuracy for 55% (17/31) of boys. Published equations did not accurately predict REE for youth in the study sample. Further studies are warranted to formulate accurate energy prediction equations for this population.
Keywords: resting energy expenditure, Hispanic, obese, Institute of Medicine, Harris Benedict, Schofield, youth, lean body mass