Laboratory assessment of the food intake of children and adolescents with loss of control eating

Abstract

Background: Loss of control (LOC) eating in youth predicts excessive weight gain. However, few studies have measured the actual energy intake of children reporting LOC eating.

Objective: The objective was to characterize the energy intake and macronutrient composition of "normal" and "binge" laboratory meals in nonoverweight and overweight boys and girls with LOC eating.

Design: Children aged 8-17 y (n = 177) consumed 2 lunchtime meals ad libitum from a multi-item food array after being instructed to either binge eat (binge meal) or to eat normally (normal meal). Prior LOC eating was determined with a semistructured clinical interview.

Results: Participants consumed more energy at the binge meal than at the normal meal (P = 0.001). Compared with youth with no LOC episodes (n = 127), those reporting LOC (n = 50) did not consume more energy at either meal. However, at both meals, youth with LOC consumed a greater percentage of calories from carbohydrates and a smaller percentage from protein than did those without LOC (P < 0.05). Children with LOC ate more snack and dessert-type foods and less meats and dairy (P < 0.05). LOC participants also reported greater increases in postmeal negative affect at both meals than did those without LOC (P < or = 0.05). Secondary analyses restricted to overweight and obese girls found that those with LOC consumed more energy at the binge meal (P = 0.025).

Conclusions: When presented with an array of foods, youth with LOC consumed more high-calorie snack and dessert-type foods than did those without LOC. Further research is required to determine whether habitual consumption of such foods may promote overweight. This trial was registered at clinicaltrials.gov as NCT00320177.

Figures

FIGURE 1

Adjusted, back-transformed geometric mean (±SE) energy intakes for all participants at the 2 meals. **Energy intake was significantly greater at the binge meal than at the normal meal (linear mixed models with repeated measures, meal type). n = 177.

FIGURE 2

Adjusted, back-transformed mean (±SE) macronutrient contents of the meals. **Significantly different from participants with no loss of control, P ≤ 0.01 (linear mixed models with repeated measures). n = 176.

FIGURE 3

Mean (±SE) energy intakes from the different food types consumed at the 2 meals. *Significantly different from participants with no loss of control, P ≤ 0.05 (linear mixed models with repeated measures). Means were adjusted for sex, age, race, fat-free mass, and percentage fat mass. n = 172.

FIGURE 4

Mean (±SE) scores for postmeal state affect. *, **Significantly different from participants with no loss of control (linear mixed models): *P ≤ 0.05, **P ≤ 0.01. Means were adjusted for sex, age, race, fat-free mass, percentage fat mass, total energy intake, and respective pretest meal mood state. n = 176.