Estimated portion sizes of snacks and beverages differ from reference amounts and are affected by appetite status in non-obese men

Abstract

Objective: To explore the extent to which appetite status influences portion size estimation in men under laboratory conditions and to quantify how much participants' portion estimates differed from the recommended portion sizes defined by authoritative bodies (i.e. government and health professionals' reference amounts). Design: Repeated, randomized cross-over trial with each participant attending the laboratory on four separate occasions. At each session, participants rated the number of portions of eight foods and beverages displayed in front of them. Participants rated portions twice after consuming breakfast (full conditions) and twice after an overnight fast (hungry conditions). Portion estimates were compared with reference amounts from the British and American Dietetic Associations, from the UK Food Standards Agency and from the US Food and Drug Administration. Setting: Food skills laboratory, University of Chester, UK. Subjects: Twenty-seven non-obese men (mean age 24?9 (SD 6?5) years). Results: Portion size estimates for all items were significantly smaller under hungry than under full conditions (P , 0?01). Relative to reference instruments, estimates were significantly smaller for all foods except banana, irrespective of appetite status (P , 0?001). Conclusions: In this data set, appetite status altered the perception of food amounts. There were large discrepancies between participants' perception of a portion and recommendations from health professionals and government standards. Nutritional educational strategies should take into account the role of hunger along with a person's familiarity with existing portion size systems when advising on portion sizes.

Estimating food portions. Influence of unit number, meal type and energy density

“…In addition, some of these foods may be considered ‘unhealthy’ by some individuals which may lead them to overestimate appropriate portion sizes based on the perceived amount of energy (Carels et al, 2007; Kelly et al, 2008). In a previous study (Brogden & Almiron-Roig, 2011) energy density explained 13% of the variance in percent error of estimation and positively correlated with a high error in estimation vs. reference amounts across a sample of 8 foods. This association was dependent on the energy content more than the volume of the foods presented.…”

“…Participants were asked to refrain from drinking alcohol and to keep evening meals and activity levels similar on the day before each test. To enhance compliance, food intake and physical activity was monitored on arrival using a conditions check questionnaire (Brogden & Almiron-Roig, 2011). Subsequently participants completed baseline appetite ratings using visual analogue scales (VAS) (see ‘Appetite ratings’), and were allowed a few minutes to read the study questionnaire.…”

“…Methods based on habitual consumption contain a subjective component and may be influenced by learning or experience with that food, which may make them more user-friendly but less sensitive for estimating portions of foods composed of various units, than other methods (Kral, 2006). We therefore explored participants’ ability to estimate portions using two previously piloted methods, rating the number of portions displayed (Brogden & Almiron-Roig, 2011) and a scale-based rating of the portion amount against habitual amounts (Kral, 2006). Evidence suggests that men and women estimate portion sizes differently (Brunstrom et al, 2008; Burger et al, 2007), therefore our analyses were adjusted for gender differences.…”

Sensitivity to portion size of unhealthy foods

“…We focus in the current work on portion-size sensitivity for unhealthy foods, recognizing that their consumption is likely more linked to obesity, and that much prior work has focused primarily on them (e.g., Block et al, 2013;Brogden & Almiron-Roig, 2011;Bryant & Dundes, 2005;Geier, Rozin, & Doros, 2006;Hernández et al, 2006;Just & Wansink, 2013). Thus, we include numerous foods that are classified as unhealthy based on their level of processing (i.e., they contained significant added fat or added sugar).…”

“…Thus, we hypothesize that Healthy Food Consumption (5a) will be negatively associated with perceptions of food healthiness of the average-sized food, and (5b) will be associated with greater linear sensitivity to portion size. We also explore whether state hunger predicts portion-size sensitivity for unhealthy foods, given that state hunger is associated with perceptions of food (Brogden & Almiron-Roig, 2011;Brunstrom, Rogers, Pothos, Calitri, & Tapper, 2008;Shimizu, Payne, & Wansink, 2010). To our knowledge, this is the first study to investigate simultaneously both normative and person-specific sensitivity to food healthiness as portion size varies across three types of unhealthy food via the use of numerous standardized stimuli.…”