Children's Behavioral Responses to Portion Size
Open Access
- Author:
- Sharafi, Mastaneh
- Graduate Program:
- Nutrition
- Degree:
- Master of Science
- Document Type:
- Master Thesis
- Date of Defense:
- April 08, 2010
- Committee Members:
- Leann L Birch, Thesis Advisor/Co-Advisor
Leann L Birch, Thesis Advisor/Co-Advisor - Keywords:
- children
obesity
portion size
eating behavior - Abstract:
- Background: Childhood obesity has increased dramatically in recent decades; currently more than 30% of children are overweight or obese. The portion sizes of many commonly consumed foods have also increased substantially during this period and large portions may be contributing to the obesity epidemic. Previous research has shown that doubling the portion size of an entrée can significantly increase children’s intake at a meal. However little is known about how increasing entrée portion size, from small to very large affects children’s intake and eating behavior. Objectives: 1) To test a) how increasing portion size increased entrée intake, number of entrée bites, time spent eating the entrée or estimated bite size, b) whether these relations differed by child’s weight status and c) whether number of entrée bites, or time spent eating the entrée predicted children’s entrée intake; 2) to investigate changes in eating rate from the beginning to the end of a lunch session and whether changes in eating rate were related to children’s weight status; and 3) to test a) whether self-served entrée size was related to children’s weight status, and whether there was any difference between 400g condition and self-served condition with regards to entrée consumption and b) whether children’s estimated bite size or number of bites differed by weight status during this condition. Methods: Participants were 17 3 to 5-year-old children. A within-subject design was used to evaluate food consumption during a series of six lunches, in which 100, 160, 220, 280, 340 and 400 g entrées were served in one random order. Each child’s eating behavior was recorded by two cameras during the lunch. Two trained coders recorded the number of entrée bites taken by each child; inter-rater reliability was .99. Estimated bite size of an entrée was defined as total grams of entrée consumed divided by total number of bites taken. Time spent eating the entrée was calculated by subtracting the time at the last bite of the entrée from the time at the first bite. To calculate eating rate, the data from 340g portion lunch was used and changes in eating rate during the entire 20 minutes of lunch time were investigated across four five-minute intervals. Children’s food consumption and eating behavior was assessed during one lunch of self-served condition. Age and sex-specific BMI percentiles were calculated. Results: As portion size increased from 100 to 400 g, a significant interaction was found between BMI percentiles and portion size to predict entrée consumption. Children with higher BMI percentiles showed greater increases in entrée consumption with increasing portion size. Given the low variability in g intake of the entrée among participants within the two smallest portions, the analyses were also conducted excluding the 100g and 160g portions. The interaction of portion size and BMI percentiles was no longer significant. However, BMI percentiles and portion size were significant predictors of entrée intake. A significant interaction was found between BMI percentiles and portion size on number of entrée bites, such that children with higher BMI percentiles had greater increases in the number of entrée bites when portion size increased. No significant interactions were found between BMI percentiles and portion size to predict time spent eating the entrée or estimated bite size. There was a significant main effect of portion size on time spent eating the entrée and bite size. Number of entrée bites and time spent eating the entrée were significant predictors of entrée intake. Eating rate decreased from the beginning to the end of the meal and the changes in eating rate were moderated by BMI percentiles, with overweight children failing to show significant decreases in eating rate. In self-served condition, overweight children showed greater entrée intake and greater number of entrée bites than non-overweight children. The grams of the entrée that were self-served and the proportions of food eaten to food served did not differ between overweight and normal weight children. No significant differences in intake were found between self-served and 400g portion condition. Discussion: Increasing portion size across a wide range of portions increased children’s entrée intake by increasing number of entrée bites, and time spent eating the entrée. A moderating effect of weight status was noted for entrée intake and number of bites, suggesting that children with higher BMI percentiles are more responsive to the larger portions by eating more and taking greater number of entrée bites. Number of entrée bites, and time spent eating the entrée predicted children’s entrée intake, illustrating how these increased behavioral responses to the increased portion size may lead to greater entrée intake. Unlike normal weight children whose eating rate slowed as the meal progressed, overweight children’s eating rate did not slow significantly during the course of the meal, suggesting that deficient satiation may have been responsible for greater intake of overweight children. Findings from self-served condition were similar to those from portion size manipulation in which overweight children showed greater intake and number of entrée bites. These findings are not consistent with those of previous studies, which have shown that allowing children to self-serve portions mitigates effects of large portions. This research suggests that increasing portion size may increase children’s intake via changes in number of entrée bites and time spent eating the entrée and also suggests the possibility that children with higher BMI percentiles may be more responsive to large portions.