AJCN. First published ahead of print December 28, 2011 as doi: /ajcn

Transcription

1 AJCN. First published ahead of print December 28, 2011 as doi: /ajcn Serving smaller age-appropriate entrée portions to children aged 3 5 y increases fruit and vegetable intake and reduces energy density and energy intake at lunch 1 3 Jennifer S Savage, Jennifer O Fisher, Michele Marini, and Leann L Birch ABSTRACT Background: Previous portion size research in children has focused on the impact of large entrée portions on children s intake, but less attention has been given to how intake at a meal is affected across a broader range of entrée portions. Objective: The objective was to assess the effect of serving a range of entrée portions on children s ad libitum intake and energy density consumed at the meal. Design: A within-subject design was used to examine the effect of varying entrée portions (ie, 100, 160, 220, 280, 340, and 400 g) on children s ad libitum energy intake of macaroni and cheese and fixed portions of unsweetened applesauce, green beans, and whole-wheat roll served with the entree. Seventeen children (10 girls), aged 3 to 6 y, were served a series of 6 lunches, which varied only in entrée portionsize, once per week. Weight, height, and weighed food intake were measured. Results: Increasing portion size increased children s entrée intake (P, 0.01) and decreased intake of other foods served with the entrée, including fruit and vegetables (P, ). As a result, children consumed a more-energy-dense (kcal/g) lunch as portion size increased (P, ). Further examination showed that BMI percentile moderated the positive association between portion size and entrée intake (P, 0.01); overweight children showed greater increases in entree intake with increasing entrée portion. Conclusion: Serving smaller age-appropriate entrée portions may be one strategy to improve children s nutritional profile by decreasing intake of energy-dense foods and by promoting intake of fruit and vegetables served with the entree. Am J Clin Nutr doi: /ajcn INTRODUCTION The focus of previous portion size research in children has been on understanding the impact of relatively large entrée portions on children s intake and has shown that in comparison to a generous reference portion, doubling entrée portion size increased preschool-aged children s total energy intake at meals by 13 39% (1 5). This research has also shown that across a series of meals, increases in portion size have persistent effects, increasing children s total daily energy intake (2). Less attention, however, has been given to the influence of large main course portions on children s intake of other foods served with the meal, particularly fruit and vegetables. It is widely reported that most children eat less than the recommended amounts of fruit and vegetables (6 10). For example, one population survey estimated that as few as 50% and 22% of children aged 2 5 y meet recommendations for fruit and vegetable intake, respectively, and that intake was significantly lower among older children and adolescents (6). Studies have attempted to increase children s intake of fruit and vegetables by increasing the portion size of fruit and vegetables, either with the meal (11, 12) or by offering vegetables separately, without competing foods, as a first course (13, 14). In both cases, these approaches have produced significant but modest increases in children s fruit and vegetable intake. An alternative strategy for promoting fruit and vegetable intakes is to reduce portion sizes of the palatable, more energy-dense foods at the meal. Rolls et al (15) found that serving adults more vegetables by increasing the proportion of vegetables at a meal; however, it is unclear whether this strategy would be effective at increasing fruit and vegetable intakes among children. Because entrées are often both more energy dense and more palatable than other foods served at the meal (16 18), serving smaller age-appropriate portions of entrées while serving fixed but generous portions of fruit and vegetables may increase children s fruit and vegetable intakes. As a result, children would consume a less-energy-dense meal, which would improve the overall nutrition profile of the diet. The purpose of this research was to assess whether a linear increase in portion size (across a range of portions) influences preschool-aged children s intake of the entrée and of other foods served with the entrée, including fruit and vegetables, and of the energy density of the meal consumed. A within-subject design included 6 entrée portion sizes that varied across a 4-fold range (60-g linear increase from 100 to 400 g) that is representative of macaroni and cheese intakes among 2- to 5-y-old children participating in the Continuing Survey of Food Intakes of Individuals Survey (19). By using this data set, the smallest portion corre- 1 From the Center for Childhood Obesity Research (JSS, MM, and LLB) and the Departments of Nutritional Sciences (JSS and LLB) and Human Development and Family Studies (LLB), The Pennsylvania State University, University Park, PA, and the Department of Public Health and Temple University Center for Obesity Research and Education, Temple University, Philadelphia, PA (JOF). 2 Supported by US Department of Human and Health Services grant no , award ID no Address correspondence to JS Savage, Center for Childhood Obesity Research, 129 Noll Laboratory, University Park, PA ( jfs195@psu. edu), or LL Birch, Department of Human Development and Family Studies, 129 Noll Laboratory, University Park, PA ( llb15@psu.edu). Received April 12, Accepted for publication November 1, doi: /ajcn Am J Clin Nutr doi: /ajcn Printed in USA. Ó 2012 American Society for Nutrition 1 of 7 Copyright (C) 2011 by the American Society for Nutrition

2 2of7 SAVAGE ET AL sponded to between the 25th and 50th percentiles and the largest portion corresponded to the 95th percentile for children of this age. SUBJECTS AND METHODS Subjects Participants were 21 children aged 3 5 y (mean 6 SD: y) attending full-day childcare at the Child Development Laboratory located at The Pennsylvania State University, University Park, in the fall of Data are reported for 17 of the 21 children (10 girls); the data from 4 children were excluded from analyses because they failed to meet predetermined inclusion criteria. Exclusion criteria were the presence of food intolerance to study foods, chronic illness affecting food intake, consuming,22 g of the entrée (,10% of the 220-g entrée portion), dislike of the main entrée, uncooperative behavior during lunch, non- English speaking, or extended absences. Children were, on average, of normal weight (mean 6 SD BMI percentile: ). On the basis of CDC guidelines, 6 children (35%) had abmi.85th percentile and were identified as overweight. Parents tended to be highly educated and currently employed. The majority (70%) of parents reported being married, and most of the families (60%) reported combined family incomes of.$50,000. The Pennsylvania State University Institutional Review Board approved all study procedures, and parents provided written consent for their family s participation in the study before the initiation of data collection. Child assent was obtained for the preference assessment procedure at the beginning of the study. Design and procedures A within-subject design that included a range of portions was used to measure the effect of varying the portion size of an entrée on children s ad libitum energy intake of the entrée, of other foods served with the entrée, of milk, and of the total meal. In addition, the energy density of the meals consumed for the varying entrée portion sizes was also calculated. Children received a standard breakfast that was provided by the childcare setting. The menu at all lunches included the portion-manipulated macaroni and cheese entrée and fixed portions of 2% milk and other foods served with the entree (eg, green beans with butter, whole-wheat roll, and unsweetened applesauce). Serving sizes of other foods served with the entrée were generous to provide onethird of the daily caloric requirements at the meal for a moderately active child at the smallest entrée portion size (20, 21). Familiarity with and liking for the test meal, as well as height and weight, were assessed during the first week. Test meal intake data were not included in the analyses because that meal served to acquaint the children and preschool teachers with the food, study procedures, and the research staff. Children were then served a series of 6 lunches in a random order, once per week, which varied only in entrée portion size (entrée portion size order: 100, 160, 220, 280, 320, and 400 g). Lunch procedures Children were served lunch on the same day of the week at their regularly scheduled time in an eating laboratory dining room facility near their classroom. Because eating is typically a social activity in childcare settings, children ate lunch with classmates, seated at tables with 2 or 3 other children. Research staff was present as observers, but adults did not eat with the children or attempt to influence or control the children s intake. Children were served identically arranged trays containing the entrée and the other menu items, with uniform containers used at all meals. Instructions to children before the meal were to eat as much as they wanted, not to share food with others, to remain in their seats, and that they had 20 min to eat lunch. If children finished eating before the 20-min time limit, they were allowed to leave the eating laboratory dining room, a rule consistent with classroom procedures. Research staff recorded spillage and redirected mealtime food-related conversation to other topics to reduce peer influence on consumption. When children finished lunch, spilled or dropped foods were returned to the correct dish and lunch items were cleared. Children who did not participate in the study received a standard lunch in their classroom at their regularly scheduled time, which was provided by the childcare setting. Preference assessment Children s familiarity with and liking for the entrée wereas- sessed before data collection by using a procedure developed by Birch (22). Children were asked whether they had eaten macaroni and cheese previously, to taste the macaroni and cheese, and then categorized it as yummy, just okay, or yucky. Children who rated the entrée as yucky were excluded from the study. Experimental menu A macaroni and cheese entrée (Stouffer s) with an energy density of 1.52 kcal/g (6.36 kj/g) was served at each meal. Macaroni and cheese was selected for its familiarity and acceptability with this age group, and for comparison with prior studies examining entrée portion size in young children (3, 4). Fixed portion sizes of other foods served at the lunch were generous: green beans (85 g) with whipped butter (5 g), unsweetened applesauce (250 g), a whole-wheat roll (35 g), and a fixed serving of 2% milk (236 ml) or soy milk (244 g) for one child at the request of a parent. The nonentrée foods provided a total of 373 kcal. The portions of the other foods and milk were selected so that for the 100-g entrée, the meal would provide one-third of the kcal daily requirement for moderately active 3- to 6-y-olds (20, 21). All lunch items were consumed ad libitum. Total energy offered at study lunches ranged from 525 kcal (2197 kj) to 981 kcal (4105 kj). The energy density offered at each meal was 1.12, 1.16, 1.21, 1.24, 1.25, and 1.27 kcal/g for the 100-, 160-, 220-, 280-, 340-, and 400-g portions, respectively. Measures Background characteristics Parents completed a questionnaire to obtain information on family background characteristics including family income and mothers and fathers years of education. Anthropometric measures A trained staff member measured children s heights and weights following standard procedures described by Lohman et al (23). Height was measured in duplicate to the nearest 0.1 cm by using a portable stadiometer (Shorr Productions), and weight was

3 PORTION SIZE, CHILDREN S EATING BEHAVIOR, AND WEIGHT 3 of 7 measured in duplicate to the nearest 0.1 kg by using an electronic scale (Seca). Age- and sex-specific BMI percentile scores (kg/m 2 ) were computed for each child by using the SAS growth chart programs available online from the CDC ( growthcharts/). Overweight was defined as a BMI greater than or equal to the 85th percentile on the basis of standardized reference criteria (24). Weighed food and milk intakes Food and milk weights were recorded before and after consumption to the nearest 0.1 g by using digital scales (Mettler- Toledo PR5001 and Mettler-Toledo XS4001S; Mettler-Toledo Inc). The amount of each food item consumed (g) was determined by subtracting postmeal weights from premeal weights. Energy intake (kcal) was calculated by multiplying gram consumption by energy content according to the manufacturers nutrition information. Energy density (kcal/g) of the meal was calculated by dividing the energy content of all foods by the weight of all foods consumed, excluding milk. Milk was excluded from the energy density calculations on the basis of previous findings that indicated that the inclusion of beverages may diminish associations with outcome variables due to the increased day-to-day variance within individual respondents (25). Data for the child who consumed soy milk were excluded from energy intake from milk, total lunch energy intake, and energy-density calculations. Statistical analyses Linear mixed models were used to assess the effects of portion size on children s energy intake, across the range of portions, and all models were examined by using a within-subject design to account for the lack of independence in the measurements due to the repeated measures of consumption data (PROC MIXED; SAS Institute). Outcome variables examined were as follows: entrée energy intake, energy intake of fixed portions of other foods served with the entrée (ie, unsweetened applesauce, green beans, and whole-wheat roll), milk energy intake, total lunch energy intake, and energy density of the lunch. Child s sex, age, and BMI percentile and parents education and income were examined as covariates in models that included portion size; BMI percentile emerged as the only significant covariate, and therefore was included in all models presented below. In addition, BMI percentiles were dichotomized into 2 groups on the basis of the CDC guidelines (24), and the interaction of BMI percentile group (BMI,85th percentile and BMI 85th percentile) and portion size was analyzed to examine whether overweight children were more responsive to increasing entrée portion size than normal-weight children. Random factors in all models were child classroom, lunch table, day of week lunch was eaten, and session order. In addition, the percentage of children who consumed.95% of the entrée was examined across conditions. Data were analyzed by using SAS software (version 9.2, 2001; SAS Institute) (26). RESULTS Familiarity and liking of the foods Most children reported previously eating the entrée. About half reported the macaroni and cheese as tasting yummy, and the rest of the children rated it as just okay. Effect of varying portion size on children s entrée intake and meal energy density across g portions Entrée intake As shown in Table 1, the percentage of the entrée eaten decreased significantly (P, 0.001) as portion size increased, but despite these diminishing returns of entrée portion size on entrée intake, children consumed significantly more of the entrée (g) as the portion size increased (P, ) (see Figure 1). BMI percentile was a significant covariate (P, 0.01), and BMI was therefore included in the model as a dichotomous variable (,85th or 85th percentile) on the basis of the CDC guidelines to examine whether the effects of portion size were moderated by BMI percentile. Although not hypothesized, there was a significant interaction between BMI percentile group and portion size on entrée energy intake (P, 0.01) (see Figure 2A). In comparison with children who were not overweight, overweight children showed greater increases in intake as portion size increased; they also consumed proportionately more of the entrée as portion size increased. Intake of fixed portions of other foods served with the entree: fruit, vegetable, whole-wheat roll, and milk There was a significant effect of BMI percentile (P, 0.05) such that heavier children consumed more energy from other foods served with the entrée across all portions. As shown in Figure 1, after adjustment for BMI percentile, there was also a significant main effect of entrée portion size on the intake of the other foods served at the meal (P, 0.001); as entrée portion size increased, the amount of other foods served with the entrée thatchildrenate decreased. As shown in Table 1, when examined by each food, there was a significant linear decrease in children s intake of green beans (P, 0.05), unsweetened applesauce (P, ), and roll (P, 0.001) with increasing entrée portion size, after adjustment for BMI percentile. In contrast, milk intake was unaffected by variations in entrée portion size; across all conditions, children drank 40 50% of the served milk. As shown in Table 1, the percentage of the other foods consumed (unsweetened applesauce, green beans, and wheat roll) also decreased significantly (P, 0.001) as entrée portion size increased, after adjustment for BMI percentile. We also tested the relation between entrée and intake of other foods served with the entrée by using a withinsubject regression approach and found that as entrée intakein- creases, intake of other foods decreases (b = 20.13, P, 0.001). The intake data are presented in Table 1 and show that in the 100-g condition children, on average, consumed 95% of the entrée and 46% of the other foods (ie, unsweetened applesauce, green beans, and whole-wheat roll). As a result, even in the smallest portion size condition, on average, children consumed only 59% of the total energy available at the meal. Thus, despite being provided with ample calories, children, on average, did not approach consumption of one-third of the daily recommended intake; it was not until being provided with the 340-g portion of the energy-dense macaroni and cheese that children met this requirement. However, from the 100-g to the 340-g portion, plate waste of the other foods served with the entrée increased from 54% to 74%, respectively (ie, consumed and 97.2 g of the 375 g offered). Last, a subanalysis was conducted to remove the 2 smallest portions (100 and 160 g) because 81% and 76% of the children, respectively, consumed.95% of these portions, resulting in

4 4of7 SAVAGE ET AL TABLE 1 Mean energy intake (kcal), percentage consumed of each food, and energy density (kcal/g) by portion size, after adjustment for BMI percentile (n = 17) 1 Entrée portion size offered 100 g 160 g 220 g 280 g 340 g 400 g P Entree Macaroni and cheese kcal , Percentage consumed , Fixed portions of other foods Unsweetened applesauce kcal , Percentage consumed , Green beans kcal Percentage consumed Whole-wheat roll kcal ,0.001 Percentage consumed ,0.001 Milk (2%) kcal NS Percentage consumed NS Total meal kcal ,0.001 Percentage consumed ,0.001 Energy density consumed kcal/g , The beverage (milk) was not included when calculating energy density of foods (kcal/g). Probabilities tested the linear effect of portion size on each item, after adjustment for BMI percentile in each model. The total energy offered at the meal was 534, 623, 719, 809, 897, and 986 kcal for the 100-, 160-, 220-, 280-, 340-, and 400-g portions, respectively. Fixed portions of unsweetened applesauce, green beans, whole-wheat roll, and 2% milk provided 102, 61, 80, 130 kcal, respectively. Energy density offered at the meal was 1.12, 1.16, 1.21, 1.24, 1.25, and 1.27 kcal/g for the 100-, 160-, 220-, 280-, 340-, and 400-g portions, respectively. unequal variance across the entrée conditions. However, similar results emerged across the g conditions. As shown in Figure 1, serving larger portions of the entrée significantly increased entrée intake(p, ) and decreased intake of fixed portions of other foods served with the entrée (P, 0.01). FIGURE 1. Mean (6SE) main effects of portion size on entrée and other food intake (g) across the 100- to 400-g portions (n = 17). Fixed portions of Other Foods included unsweetened applesauce, green beans, and wheat roll. There was a significant positive linear effect of portion size on entrée intake across the 100- to 400-g portions (P, ) and across the 220- to 400-g portions (P, ). There was a significant negative linear effect of portion size on other food intake across the 100- to 400-g portions (P, ) and across the 220- to 400-g portions (P, 0.01). Data were analyzed by using a within-subject linear mixed model. Total energy intake As shown in Table 1, similar to the results observed for entrée intake, there was a significant main effect of portion size (P, ) on total lunch energy intake, after adjustment for BMI percentile; as entrée portion size increased, total lunch energy intake increased. For example, total energy intake increased 61% from the smallest to the largest condition. As shown in Figure 2B, there was also a significant interaction between BMI percentile group and portion size on total lunch energy intake (P, 0.01); overweight children had greater increases in energy intake with increasing portion size than did nonoverweight children. Last, as shown in Figure 3, as portion size increased, the percentage of energy consumed from the entrée (entrée kcal/ total meal kcal served) significantly increased (P, ) and the percentage of the other foods consumed significantly decreased (P, ) after adjustment for BMI percentile. For example, there was a 48% decrease in intake from other foods served with the entrée from the smallest to the largest condition. There was no change in the percentage of milk consumed across conditions; milk intake accounted for ~14% of children s meal energy intake. Energy density of meals consumed As shown in Table 1, the energy density of the lunches children consumed increased as portion size increased; there was a significant linear increase in the consumed energy density of the lunch with increasing portion size (P, ).

5 PORTION SIZE, CHILDREN S EATING BEHAVIOR, AND WEIGHT 5 of 7 FIGURE 2. Child weight status (BMI percentile group) moderated the effect of portion size on entrée (A) and total lunch intake (B) (P, 0.01; n = 17). Values are means 6 SEs. BMI percentile was dichotomized into 2 groups, overweight (BMI: 85th percentile; n = 6) or not overweight (n = 11), to examine differences in responsiveness to increasing portion on intake by child weight classification. Total lunch energy intake included manipulated entrée portions and fixed portions of unsweetened applesauce, green beans, wheat roll, and milk. Data were analyzed by using a within-subject linear mixed model. %tile, percentile. DISCUSSION The results of this study provide new evidence that serving children age-appropriate entrée portions results in greater fruit and vegetable intake at meals than serving larger portion sizes. Results showed that across a wide range of entrée portion sizes, serving larger entrée portions increased children s entrée intake, replicating previous findings. In addition, serving larger portions significantly reduced children s intake of fruit and vegetables that were served at the meal. As a result, total energy intake and energy density of the meal increased with increasing entrée portions. Although not hypothesized, our findings also suggest that overweight children may be particularly responsive to increased portion sizes. Additional research should be designed and powered to address this potentially important issue. Together, these findings provide evidence that across a range of entrée portions that reflect typical intakes of preschool-aged children, serving smaller age-appropriate portions of a palatable energydense entrée shows promise as one strategy to promote intake of FIGURE 3. Percentages of lunch energy consumed from the entrée, other foods, and milk across the 100- to 400-g portions (n = 17). Fixed portions of other foods included unsweetened applesauce, green beans, and wheat roll. The percentage of the entrée consumed in comparison with the percentage of the total meal consumed (entrée kcal/offered meal kcal) significantly increased (P, ), and the percentage of the other foods consumed significantly decreased (P, ), as entrée portions increased, after adjustment for BMI percentile. Data were analyzed by using a withinsubject linear mixed model. lower-energy-dense foods such as fruit and vegetables. Children consumed less of the lower-energy-dense foods as portions of the energy-dense entrée increased. In addition to increasing fruit and vegetable intake and reducing energy density and energy intake, an additional benefit of serving smaller entree portions was that the amount of plate waste was also less. In the present study, children consumed greater amounts of fruit and vegetables when served smaller age-appropriate portions of an energy-dense entrée than when served larger portions. For example, in the 100-g entrée condition, intake of the other foods (ie, green beans, unsweetened applesauce, and wheat roll) was 90% higher and intake of the entrée was 63% lower than in the 400-g portion condition. Thus, with the smaller age-appropriate portions of macaroni and cheese (0.5 cup or 120 g macaroni and cheese), when the proportion of fruit and vegetables on the plate was greater, children consumed a less-energy-dense meal, improving one index of overall diet quality at the meal. Further examination of this effect showed that intake of unsweetened applesauce decreased from 0.5 cup (123.9 g) to 0.33 cup (81.8 g) from the 160- to 400-g portions, respectively. Green bean consumption decreased from 11 to 4 g (<18% to 6% of a 0.5-cup serving, respectively) from the 100- to 400-g portions. Together, these findings suggest that children s intakes of lower-energydense foods such as fruit and vegetables, which tend to be less well liked, decrease as larger portions of palatable, energy-dense entrées are served. These findings add credence to the recommendation in the new 2010 Dietary Guidelines (27) that specifically instruct Americans to fill half their plates with fruit and vegetable. In the current study, increasing the proportion of other foods on the plate (ie, unsweetened applesauce, green beans, and whole-wheat roll) increased fruit and vegetable intake and reduced the energy density of the meal consumed. In the present study, serving larger portions of a palatable, energy-dense entrée pushed down energy intake of green beans and applesauce. In other words, children filled up and became satiated by eating the entrée and chose to forego the less-energydense, less-well-liked foods such as fruit and vegetables. In contrast to findings from the present study, previous portion size studies that doubled a reference portion, ranging from 125 to 200 g and 225 to 400 g for 2 4-y-olds and 4 6-y-olds, respectively, have not shown effects of entrée portion size on children s intake of other foods served with the meal (1, 3, 4). The reason that this

6 6of7 SAVAGE ET AL study was able to detect an effect of entrée portion size on intake of other foods is that it was designed to test the biological relation of a range of portion sizes on intake, which allows for broader inferences across the range of portions. It is more difficult to detect a single mean comparison difference than it is to detect a relative effect of an independent variable such as portion size across a range of sizes (28). Recent studies that have increased the portion size of fruit and vegetables to increase intake have had some success, but plate waste was high. For example, Spill et al (13) found that serving larger portions of low-energy-dense vegetables as a first course (30 compared with 60 g) nearly doubled total vegetable intake at a meal, without altering main course entrée or vegetable intake. However, tripling the portion size (90 g) did not lead to further increases in intake. It is also important to note that plate waste of the manipulated low-energy-dense vegetable, carrots, increased as portion increased; plate waste was 18%, 40%, and 58% for the 30-, 60-, and 90-g portions, respectively. Kral et al (11) altered the proportion of fruit and vegetables served by keeping entrée portion constant, but doubled the portion of both fruit and vegetable side dishes. This resulted in an increase in children s fruit consumption by 43%, but doubling the portion size of the vegetable did not increase vegetable intake. It is also important to note that fruit and vegetable plate waste also increased from ~47% to 76% when fruit and vegetable portions were doubled (11). In the present study, when children were served fixed portions of a fruit and vegetable and the availability of the entrée on the plate was reduced to 100 g, plate waste was still relatively high, but was less compared with the larger entrée portion conditions. Specifically, meal plate waste decreased from 72% to 54% when children were served the 400- compared with the 100-g entrée portions, respectively. Together with previous findings, research from the present study and others suggests that portion size effects may differ by the type of food children are served, and whether the portion-manipulated food is served alone or with other foods. Although not hypothesized, results also showed that in this small sample, overweight children were more responsive to portion size than children who were not overweight and showed greater increases in entrée intake with increasing portion size. This finding is consistent with the results of research comparing obese and normal-weight children s responses to food cues using other protocols, which have shown that obese children are more responsive to food cues and consume more energy than do their leaner counterparts (29 32). In the study by Kral et al (11), in which the portion sizes of fruit and vegetable side dishes were doubled, there was a trend for overweight children to show greater increases in intake of fruit and vegetables than that of normal-weight children. Although previous studies have not provided evidence that overweight children respond differently to increasing portion size than do normal-weight children (2 5, 33), none of the studies (including the present study) were designed to investigate weight status as a moderator of portion size effects on intake. To investigate this issue, research designs are needed that include a broader range of values of the independent variable, including portions smaller as well as larger than mean usual intakes and children differing in age, weight status, and ethnicity. Limitations of the present study include the use of a small sample of convenience, limited diversity of families, and parents tended to have a higher socioeconomic status because they were highly educated and earned a higher than average family income. Furthermore, our relatively small sample size limits the interpretability of the moderating effect of child weight on associations between portion size and intake. Thus, more research is warranted across a wide range of portion sizes, and with the inclusion of children differing in weight status and ethnicity, to determine whether heightened responsiveness to portion size is related to weight and is a component of a behavioral phenotype for obesity. In conclusion, although there is evidence that young children can adjust energy intake within meals (30, 34) and over the course of the day (35, 36), these findings confirm that children s energy intake can be readily affected across a range of entrée portions, including those that are typically consumed by preschool-aged children. Specifically, serving smaller age-appropriate portions promoted increased intake of low-energy-dense, nutrient-rich foods such as fruit and vegetables, while also decreasing plate waste, which is an important consideration in childcare settings. Thus, our findings suggest another way that large portions affect diet quality. Our findings also have public health implications because data from the NHANES show that 50% and 78% of children aged 2 to 5 y do not meet MyPyramid fruit and vegetable intake recommendations, respectively (6). Findings from the present study also raise the possibility that overweight children may be more susceptible to overeating in the presence of larger portions of palatable foods at meals. Thus, additional research that identifies strategies to alter children s energy intake while promoting fruit and vegetable intake is warranted. We thank the teachers and families at the Pennsylvania State University Child Development Laboratory for their participation in and support of this research. The authors responsibilities were as follows JSS: collected the data, analyzed the data, and wrote the manuscript; LLB and JOF: designed the experiment and participated in the interpretation of results and manuscript preparation; and MM: collected and analyzed the data and assisted with manuscript preparation. None of the authors had any financial or personal interests in organizations sponsoring this research. The authors declared that they had no other competing interests. REFERENCES 1. Fisher JO. Effects of age on children s intake of large and self-selected food portions. Obesity (Silver Spring) 2007;15: Fisher JO, Arreola A, Birch LL, Rolls BJ. Portion size effects on daily energy intake in low-income Hispanic and African American children and their mothers. Am J Clin Nutr 2007;86: Fisher JO, Liu Y, Birch LL, Rolls BJ. Effects of portion size and energy density on young children s intake at a meal. Am J Clin Nutr 2007;86: Orlet Fisher J, Rolls BJ, Birch LL. Children s bite size and intake of an entree are greater with large portions than with age-appropriate or self-selected portions. Am J Clin Nutr 2003;77: Rolls BJ, Engell D, Birch LL. Serving portion size influences 5-year-old but not 3-year-old children s food intakes. J Am Diet Assoc 2000;100: Lorson BA, Melgar-Quinonez HR, Taylor CA. Correlates of fruit and vegetable intakes in US children. J Am Diet Assoc 2009;109: Nielsen SJ, Popkin BM. Patterns and trends in food portion sizes, JAMA 2003;289: Fox MK, Condon E, Briefel RR, Reidy KC, Deming DM. Food consumption patterns of young preschoolers: are they starting off on the right path? J Am Diet Assoc 2010;110:S Boutelle KN, Birkeland RW, Hannan PJ, Story M, Neumark-Sztainer D. Associations between maternal concern for healthful eating and maternal eating behaviors, home food availability, and adolescent eating behaviors. J Nutr Educ Behav 2007;39:

7 PORTION SIZE, CHILDREN S EATING BEHAVIOR, AND WEIGHT 7 of Krebs-Smith SM, Guenther PM, Subar AF, Kirkpatrick SI, Dodd KW. Americans do not meet federal dietary recommendations. J Nutr 2010; 140: Kral TV, Kabay AC, Roe LS, Rolls BJ. Effects of doubling the portion size of fruit and vegetable side dishes on children s intake at a meal. Obesity (Silver Spring) 2010;18: Mathias KC, Rolls BJ, Birch LL, Kral TVE, Fisher JO. Does serving children larger portions of fruit affect vegetable intake? Obesity 2009; 17:S90 (abstr). 13. Spill MK, Birch LL, Roe LS, Rolls BJ. Eating vegetables first: the use of portion size to increase vegetable intake in preschool children. Am J Clin Nutr 2010;91: Spill MK, Birch L, Roe LS, Rolls BJ. Serving large portions of vegetable soup at the start of a meal affected children s energy and vegetable intake. Appetite 2011;57: Rolls BJ, Roe LS, Meengs JS. Portion size can be used strategically to increase vegetable consumption in adults. Am J Clin Nutr 2010;91: Skinner JD, Carruth BR, Wendy B, Ziegler PJ. Children s food preferences: a longitudinal analysis. J Am Diet Assoc 2002;102: Wardle J, Sanderson S, Leigh Gibson E, Rapoport L. Factor-analytic structure of food preferences in four-year-old children in the UK. Appetite 2001;37: Prentice AM. Manipulation of dietary fat and energy density and subsequent effects on substrate flux and food intake Am J Clin Nutr 1989;67:535S 41S. 19. Smiciklas-Wright H, Mitchell DC, Mickle SJ, Cook A, Goldman JD. Foods commonly eaten in the United States: quantities consumed per eating occasion and in a day: quantities consumed per eating occasion and in a day, US DANFS Report No. 96-5, Accessed January 18, Available from: place/ /pdf/portion.pdf. 20. USDA; US Department of Health and Human Services. Adequate nutrients within caloric needs. Dietary guidelines for Americans, Washington, DC: US Government Printing Office, USDA, Center for Nutrition Policy and Promotion. My Pyramid food intake pattern calorie levels Accessed January 18, Available from: pdf. 22. Birch L. Dimensions of preschool children s food preferences. J Nutr Educ 1979;11: Lohman T, Roche A, Martorell R. Anthropometric standardization reference manual. Champaign, IL: Human Kinetics Book, Ogden CL, Carroll MD, Curtin LR, Lamb MM, Flegal KM. Prevalence of high body mass index in US children and adolescents. JAMA 2010; 303: Ledikwe JH, Blanck HM, Khan LK, Serdula MK, Seymour JD, Tohill BC, Rolls BJ, et al. Dietary energy density determined by eight calculation methods in a nationally representative United States population. J Nutr 2005;135: SAS Institute Inc. Statistical analysis software. Cary, NC: SAS Institute Inc, USDA; US Department of Health and Human Services. Dietary guidelines for Americans, Washington, DC: US Government Printing Office, Lowry SR. Use and misuse of multiple comparisons in animal experiments. J Anim Sci 1992;70: Temple JL, Legierski CM, Giacomelli AM, Salvy SJ, Epstein LH. Overweight children find food more reinforcing and consume more energy than do nonoverweight children. Am J Clin Nutr 2008;87: Johnson SL, Birch LL. Parents and children s adiposity and eating style. Pediatrics 1994;94: Wallner-Liebmann S, Koschutnig K, Reishofer G, Sorantin E, Blaschitz B, Kruschitz R, Unterrainer HF, Gasser R, Freytag F, et al. Insulin and hippocampus activation in response to images of high-calorie food in normal weight and obese adolescents. Obesity (Silver Spring) 2010;18: Bruce AS, Holsen LM, Chambers RJ, et al. Obese children show hyperactivation to food pictures in brain networks linked to motivation, reward and cognitive control. Int J Obes (Lond) 2010;34: Leahy KE, Birch LL, Fisher JO, Rolls BJ. Reductions in entree energy density increase children s vegetable intake and reduce energy intake. Obesity (Silver Spring) 2008;16: Birch LL, Deysher M. Caloric compensation and sensory specific satiety: evidence for self regulation of food intake by young children. Appetite 1986;7: Birch LL, Johnson SL, Jones MB, Peters JC. Effects of a nonenergy fat substitute on children s energy and macronutrient intake. Am J Clin Nutr 1993;58: Birch LL, Johnson SL, Andresen G, Peters JC, Schulte MC. The variability of young children s energy intake. N Engl J Med 1991;324: