Effects of energy density and packaging format on portion size estimation
AuthorsDodd, Jessica Z.
MetadataShow full item record
AbstractConsumption of large portion sizes in particular of high energy dense foods has been associated with increased energy intakes and the potential to induce obesity. Estimating how much is adequate to consume at one occasion is however difficult, especially for highly energetic foods, foods sold in large packages and when meals contain a number of items (e.g. restaurant meals and on-the-go “meal deals”). This study aimed at exploring the role of food energy density (kcal/g) and presentation format (single unit vs. multi-unit) on food portion size estimation in young adults. We also explored the relationship between estimated fat/calorie content and actual fat/calorie content of the test foods; and compared two different methods of portion size estimation to find out which method was associated with less estimation error in multi-item foods. Thirty-two lean, non-restricted subjects (15 males, 17 females) rated the portion sizes of 33 foods under standardised appetite conditions in the laboratory. Foods varied according to energy density (ED); presentation format (single unit food vs. multi-item meals); texture and meal label (i.e. “to be consumed as a snack” or “as a meal”). Outcome measures included liking and familiarity ratings; estimated number of portions, from which %error in estimation versus reference amounts was calculated; estimated fat and calorie content; and comparison to usual portion. Portion size estimation was significantly different from the Food Standards Agency (FSA) portion size reference amounts for most food/drink items (p<0.05). Accuracy in estimation was not influenced by either energy density or presentation, as errors in estimation occurred regardless of energy density, food type and food unit. Energy density had a significant effect on the percentage error in estimation against FSA standards in both males (p=0.011) and females (0.016). Labelling food by their food type (meal, snack, beverage) had a significant effect on percentage error of portion size estimation (p=0.047), with a great affect in females (p=0.036) than in males (p=0.088). Displaying foods by food unit (single, multi-item) had no affect on percentage error. Subjects were able to accurately estimate calorie content when compared against energy density (r=0.442, p=0.010), total energy load (r=0.766, p=0.000) and energy density category (r=0.434, p=0.000). Fat estimation was also accurately estimated when compared against fat density (r=0.633, p=0.000), total fat content (r=0.866, p=0.000) and reference fat categories (r=0.454, p=0.000). Labelling foods dependent on their food type affects percentage error of portion size estimation, with labelling having a greater affect on females than males. Displaying foods as a single unit or multi-item food does not affect percentage error of estimation. Subjects were able to accurately estimate fat/calorie content against actual fat/calorie content in the foods displayed.
PublisherUniversity of Chester
TypeThesis or dissertation
SponsorsFaculty of Applied and Health Sciences, University of Chester
The following license files are associated with this item: