Meal Size, Not Body Size, Explains Errors in Estimating the Calorie Content of Meals

BACKGROUND: Although most people underestimate the calories they consume during a meal or during the day, calorie underestimation is especially extreme among overweight persons. The reason for this systematic bias is unknown. OBJECTIVE: To investigate whether the association between calorie underestimation and body mass reflects a tendency for all persons to underestimate calories as the size of a meal increases. DESIGN: Overweight and normal-weight adults estimated the number of calories of a fast-food meal they had ordered and eaten (study 1) or of 15 fast-food meals that were chosen by the experimenter (study 2) in a randomized, controlled trial. Their estimations were compared with the actual number of calories of the meals. SETTING: Study 1 was a field study conducted in fast-food restaurants in 3 medium-sized midwestern U.S. cities. Study 2 was conducted in a laboratory at a major U.S. research university. PARTICIPANTS: Study 1 involved 105 lunchtime diners (average body mass index [BMI], 24.2 kg/m2 [range, 17.2 to 33.5 kg/m2]). Study 2 involved 40 undergraduate students (average BMI, 23.2 kg/m2 [range, 16.1 to 32.3 kg/m2]). MEASUREMENTS: Participants were asked to estimate the number of calories in a fast-food meal they had ordered and eaten (study 1) or in 15 sizes of the same fast-food meal (study 2). The actual number of calories in the meals in the field study was obtained by unobtrusively recording the food that was ordered (identified from the wrappings and containers). Weight and height were self-reported. RESULTS: Although participants strongly underestimated the number of calories in larger meals (by -38.0% in study 1 and by -22.6% in study 2), they almost perfectly estimated the number of calories in smaller meals (by -2.9% in study 1 and by 3.0% in study 2). After the authors controlled for body weight-related differences in meal size, the calorie estimations of normal-weight and overweight participants were identical in both studies. LIMITATIONS: These studies examined fast-food meals. Weight and height were self-reported. There were too few observations to distinguish between obese (BMI > or =30 kg/m2) and overweight (BMI >25 kg/m2 but <30 kg/m2) participants. CONCLUSIONS: Greater underestimation of calories by overweight persons is a consequence of their tendency to consume larger meals. Calorie underestimation is related to meal size, not body size.