Abstract
Background: The overconsumption of energy-dense foods leads to excessive energy intakes. The substitution of low-energy-dense vegetables for foods higher in energy density can help decrease energy intakes but may be difficult to implement if individuals dislike the taste of vegetables.
Objective: We investigated whether incorporating puréed vegetables to decrease the energy density of entrées at multiple meals reduced daily energy intakes and increased daily vegetable intakes.
Design: In this crossover study, 20 men and 21 women ate ad libitum breakfast, lunch, and dinner in the laboratory once a week for 3 wk. Across conditions, entrées at meals varied in energy density from standard versions (100% condition) to reduced versions (85% and 75% conditions) by the covert incorporation of 3 or 4.5 times the amount of puréed vegetables. Entrées were accompanied by unmanipulated side dishes. Participants rated their hunger and fullness before and after meals.
Results: Subjects consumed a consistent weight of foods across conditions of energy density; thus, the daily energy intake significantly decreased by 202 ± 60 kcal in the 85% condition (P < 0.001) and by 357 ± 47 kcal in the 75% condition (P < 0.0001). Daily vegetable consumption significantly increased from 270 ± 17 g of vegetables in the 100% condition to 487 ± 25 g of vegetables in the 75% condition (P < 0.0001). Despite the decreased energy intake, ratings of hunger and fullness did not significantly differ across conditions. Entrées were rated as similar in palatability across conditions.
Conclusions: Large amounts of puréed vegetables can be incorporated into various foods to decrease the energy density. This strategy can lead to substantial reductions in energy intakes and increases in vegetable intakes. This trial was registered at clinicaltrials.gov as NCT01165086.
INTRODUCTION
Americans are exposed to an environment filled with easily accessible, energy-dense foods that promote the consumption of excess energy. To reduce energy intakes, government agencies recommend substituting low-energy-dense foods such as vegetables for foods higher in energy density (1, 2). Research has shown that this strategy has multiple benefits because it reduces energy intakes (3–5) and increases vegetable intakes (6); it may be difficult, however, for some adults to implement. One barrier that prevents individuals from meeting recommendations to increase vegetable intakes is a dislike for the taste of vegetables (7–9). Puréeing vegetables and covertly adding them to foods while maintaining palatability could be an effective strategy to help individuals overcome this barrier. The purpose of the current study was to determine whether covertly incorporating low-energy-dense puréed vegetables into foods increased vegetable intakes and decreased energy intakes in adults over 1 d.
Laboratory-based studies have shown that people tend to eat a consistent weight of food; as a result, if the energy density of the food is decreased, people consume less energy (3, 5, 10). Several methods can be used to decrease the energy density of foods, including reducing the fat and sugar content and increasing the proportion of water-rich fruit and vegetables. A few studies have focused on the strategy of decreasing energy density by increasing the vegetable content of foods. These studies in adults have shown that overtly substituting vegetables for more energy-dense ingredients in mixed dishes led to reduced energy intakes (3–5) and increased vegetable intakes (6). Participants in these studies were screened to ensure that they liked and would eat the vegetables used for the manipulations (3, 5, 6). Thus, the question of how to increase intakes of low-energy-dense vegetables in people who vary in their liking for vegetables remains unanswered. One study in preschool children, who are often picky about eating vegetables, indicated that covertly incorporating puréed vegetables in foods could increase vegetable intakes and reduce energy intakes (11). In adults, it is not known whether the incorporation of puréed vegetables into foods has similar effects on energy and vegetable intakes.
In the current study, the energy density of entrées at breakfast, lunch, and dinner was varied by covertly incorporating different amounts of puréed vegetables while maintaining a similar palatability. Entrées were accompanied by various unmanipulated side dishes, and all foods were consumed ad libitum. It was hypothesized that adding puréed vegetables to reduce the energy density of entrées would lead to a reduction in energy intakes and an increase in vegetable intakes at each meal. In addition, it was predicted that these effects would persist throughout the day and result in a reduction in daily energy intakes and an increase in daily vegetable intakes.
SUBJECTS AND METHODS
Study design
This experiment used a crossover design with repeated measures within subjects. One day a week for 3 wk, participants were provided with all of their foods and beverages for breakfast, lunch, dinner, and evening snack. Across test days, the entrées served at the 3 main meals were varied in energy density (100%, 85%, or 75%) by changing the vegetable content. The entrées were accompanied by various unmanipulated side dishes. Main meals were served in the laboratory, and unmanipulated evening snacks and bottled water were provided for consumption outside of the laboratory. All foods and beverages served in the study were consumed ad libitum. The order of experimental conditions was randomly assigned across participants.
Participants
Men and women aged 20–45 y were recruited for the study through advertisements in campus electronic newsletters. Potential subjects were interviewed by telephone to determine whether they met the initial study criteria, which included that subjects had a reported body mass index (BMI; in kg/m2) between 18 and 40, regularly ate 3 meals/d, did not smoke, did not have any food allergies or restrictions, were not athletes in training, were not dieting, were not taking medications that would affect appetite, and were willing to consume the foods served in the test meals.
Potential subjects who met the initial study criteria came to the laboratory to have their heights and weights measured (model 707; Seca Corp, Hanover, MD) and to rate the taste of food samples, including the manipulated study entrées at 85% energy density. The following questionnaires were also completed: a detailed demographic and weight-history questionnaire, the Zung self-rating scale (12), which was used to evaluate symptoms of depression, the eating attitudes test (13), which was used to assess indicators of disordered eating, and the eating inventory (14), which was used to measure dietary restraint, tendency toward hunger, and disinhibition. Exclusion criteria included a measured BMI <18 or > 40, a taste rating for any entrée sample ≤30 mm on a 100-mm scale, a score ≥40 on the Zung scale, or a score ≥20 on the eating attitudes test. Subjects were also excluded if they had not maintained their weight ≤4.5 kg during the 6 mo before the study. During the study, subjects were excluded if they did not meet a minimum intake of 50 kcal from the manipulated breakfast entrée and 100 kcal from the manipulated lunch and dinner entrées. Subjects were told that the purpose of the study was to determine perceptions of different tastes. Subjects provided signed consent forms and were financially compensated for their participation. All aspects of the study were approved by The Pennsylvania State University Office for Research Protections.
The sample size for the experiment was estimated by using data from previous 1-d studies in the laboratory. The minimum difference in daily energy intakes assumed to be clinically significant was 200 kcal. A power analysis was performed and estimated that a sample size of 37 subjects was needed to detect this difference in daily energy intakes with >80% power by using a one-sided test with a significance level of 0.05.
A total of 48 participants were enrolled in the study. Five participants were excluded from the study for failing to meet the minimum entrée intake, and one participant was excluded for not following the study protocol. The data of one additional participant was excluded for having undue influence on the outcomes according to the procedure of Littell et al (15); this individual had extremely low intakes on one test day. Thus, a total of 41 participants were included in the analyses; the characteristics of these subjects are shown in Table 1.
TABLE 1.
Characteristics of participants in a study in which the energy density of entrées was reduced by varying the amount of puréed vegetables1
| Characteristic | Women (n = 21) | Men (n = 20) |
| Age (y) | 23.9 ± 1.2 | 24.4 ± 1.0 |
| Height (m) | 1.66 ± 0.02 | 1.79 ± 0.022 |
| Weight (kg) | 64.4 ± 2.1 | 79.2 ± 2.52 |
| BMI (kg/m2) | 23.5 ± 0.8 | 24.7 ± 0.6 |
| Dietary restraint score3 | 8.4 ± 0.6 | 6.7 ± 0.9 |
| Disinhibition score3 | 6.0 ± 0.7 | 5.3 ± 0.7 |
| Hunger score3 | 4.0 ± 0.5 | 5.4 ± 0.7 |
All values are means ± SEMs.
Significantly different from women, P < 0.0001 (Student's t test).
Score from the eating inventory (14).
Test foods and meals
The entrées for each main meal were developed in 3 versions that differed in energy density as follows: 100% (standard), 85% of the standard, and 75% of the standard (Table 2). The manipulated entrées were carrot bread at breakfast, macaroni and cheese at lunch, and chicken and rice casserole at dinner. These foods were selected because they are commonly consumed, and the vegetable content could be manipulated while maintaining a similar taste, texture, and appearance across energy densities. The standard entrées were representative of the energy density and vegetable content of commonly used recipes. To reduce the energy density, the amounts of puréed vegetables (carrots, squash, and cauliflower) in the standard recipe were increased by 3 or 4.5 times as the other ingredients were decreased. Thus, low-energy-dense vegetables were substituted in the recipes for the other ingredients. Substantial portions of entrées were provided at each meal, but participants were allowed to request an additional entrée, if desired, in which case, they were served a second dish of the entrée that contained one-half the original amount. This happened infrequently during the study; 4 participants requested an additional breakfast entrée on one or more occasions and one participant requested an additional dinner entrée on one occasion. The entrées were accompanied by various unmanipulated side dishes (Table 3), and 1 L H2O in addition to the choice of coffee or tea at breakfast. Noncaloric sweeteners were provided for subjects who selected coffee or tea in addition to one creamer (20 kcal) with coffee.
TABLE 2.
Composition per 100 g of manipulated entrées served at breakfast, lunch, and dinner in a study in which the energy density was reduced by varying the amount of puréed vegetables
| 100% energy density | 85% energy density | 75% energy density | |
| Breakfast | |||
| Carrot bread | |||
| Energy (kcal) | 417 | 356 | 315 |
| Carbohydrate [g (%)] | 57.1 (55) | 49.2 (55) | 43.9 (56) |
| Protein [g (%)] | 2.3 (2) | 2.1 (2) | 2.0 (2.5) |
| Fat [g (%)] | 19.7 (43) | 16.7 (42) | 14.5 (41) |
| Vegetable (g) | 8.9 | 23.9 | 35.2 |
| Fiber (g) | 1.1 | 1.3 | 1.5 |
| Energy density (kcal/g) | 4.17 | 3.56 | 3.15 |
| Lunch | |||
| Macaroni and cheese | |||
| Energy (kcal) | 212 | 180 | 161 |
| Carbohydrate [g (%)] | 16.2 (31) | 14.1 (31) | 12.8 (32) |
| Protein [g (%)] | 9.9 (19) | 8.5 (19) | 7.6 (19) |
| Fat [g (%)] | 12.1 (51) | 10.2 (51) | 8.9 (50) |
| Vegetable (g) | 1.4 | 17.7 | 28.1 |
| Fiber (g) | 0.7 | 0.9 | 1.0 |
| Energy density (kcal/g) | 2.12 | 1.80 | 1.60 |
| Dinner | |||
| Chicken rice casserole | |||
| Energy (kcal) | 162 | 139 | 122 |
| Carbohydrate [g (%)] | 19.4 (48) | 16.8 (48) | 15.3 (50) |
| Protein [g (%)] | 7.3 (18) | 6.3 (18) | 5.7 (19) |
| Fat [g (%)] | 6.2 (34) | 5.2 (34) | 4.4 (32) |
| Vegetable (g) | 13.2 | 28.2 | 39.8 |
| Fiber (g) | 0.8 | 1.0 | 1.3 |
| Energy density (kcal/g) | 1.63 | 1.39 | 1.23 |
TABLE 3.
Foods served as side dishes at each meal and at the evening snack in a study in which the energy density was reduced by varying the amount of puréed vegetables
| Meal | Food | Amount |
| g | ||
| Breakfast | Strawberry yogurt1 | 280 |
| Sliced peaches2 | 160 | |
| Lunch | Buttered broccoli34 | 130 |
| Grapes | 200 | |
| Chocolate pudding2 | 200 | |
| Wheat roll5 | 43 | |
| Butter4 | 23 | |
| Dinner | Buttered green beans46 | 130 |
| Mandarin oranges2 | 160 | |
| Pound cake7 | 63 | |
| White roll8 | 43 | |
| Butter4 | 23 | |
| Evening snack | Fig cookies9 | 95 |
| Popcorn10 | 56 | |
| Baby carrots | 150 |
Yoplait USA Inc, Minneapolis, MN.
Sysco Corp, Houston, TX.
Birds Eye Foods Inc, Rochester, NY.
Land O'Lakes Inc, Arden Hills, MN.
Bakery de France, Rockville, MD.
Hanover Foods Corp, Hanover, PA.
Sara Lee Corp, Downers Grove, IL.
Flowers Bakeries Food Service, Tucker, GA.
Kraft Foods Global Inc, Northfield, IL.
Frito-Lay Inc, Plano, TX.
Unmanipulated evening snacks (Table 3) and bottled water were provided after the dinner meal to be consumed outside of the laboratory. Bottled water was also provided for consumption between meals. All foods and beverages were consumed ad libitum and were weighed before and after meals to determine the amount consumed to the nearest 0.1 g. Energy and macronutrient intakes were calculated by using information from food manufacturers and a standard nutrient database (16).
Procedures
The day before each test day, subjects were instructed to keep their food intakes and activity levels consistent, refrain from consuming alcohol, and refrain from eating after 2200 the evening before each test day. Subjects kept a record of their intakes and activities to encourage compliance with this protocol. On test days, subjects came to the laboratory at scheduled meal times, were seated in individual cubicles, and completed a brief questionnaire that asked whether they had felt ill, taken any medications, or consumed any foods or beverages not provided by the researchers since the last meal. Their meals were then served, and they were instructed to consume as much or as little of the foods and beverages as desired. Lunch was served ≥3 h after breakfast, and dinner was served ≥4 h after lunch.
Ratings of hunger, satiety, and food characteristics
Subjects used visual analog scales (17) to rate their hunger, fullness, thirst, prospective consumption, and nausea immediately before and after each meal and immediately before consuming the evening snack. The characteristics of the entrées were also assessed by using visual analog scales. Immediately before and after each meal, subjects were provided with a sample of the manipulated entrée and instructed to first rate the appearance of the sample and then eat the sample and answer the remaining questions about pleasantness of taste and pleasantness of texture. After the final meal, subjects completed a discharge questionnaire to report their ideas about the purpose of the study and any differences they noticed between test days. Subjects also completed a questionnaire about food preferences in which they were asked to rate their liking of a variety of foods on a scale of 1–7 with 1 representing “Dislike strongly” and 7 representing “Like strongly.” Of the various foods listed, the majority were foods that were served in the study.
Data analyses
Data were analyzed by using a mixed linear model with repeated measures (SAS 9.1; SAS Institute Inc, Cary, NC). The fixed effects in the model were the experimental condition (entrée energy density), study week, and subject sex. The primary outcomes for the study were food intake (g), vegetable intake (g), energy intake (kcal), and energy density (kcal/g) at each meal and over the entire day. Energy density was calculated based on foods only; beverages were excluded (18). Vegetable intakes were characterized by both weight and volume; the volume of one serving of vegetables was defined as 0.5 cups (118 mL) (19). Secondary outcomes were participant ratings of hunger, satiety, and food characteristics. Subject characteristics were investigated as covariates in the main statistical model. Analysis of covariance was also used to determine whether participant ratings of hunger and satiety influenced the relation between experimental condition and meal energy intake and whether participant ratings of entrée characteristics influenced the relation between the experimental condition and intakes of the manipulated entrées. Results are reported as means ± SEs and were considered significant at P < 0.05.
RESULTS
Vegetable intake
The incorporation of additional puréed vegetables to reduce the energy density of the entrées significantly increased the total vegetable intake at each meal (P < 0.0001; Figure 1) and over the day (P < 0.0001; Table 4). The amount of vegetables consumed daily from the entrées was 62 ± 3 g vegetables in the 100% condition, 198 ± 9 g vegetables in the 85% condition, and 288 ± 14 g vegetables in the 75% condition. This was equivalent to one additional vegetable serving per day in the 85% condition and 2 additional vegetable servings per day in the 75% condition. Although the vegetable intake from the entrées increased as their vegetable content was increased, the consumption of vegetable side dishes at the lunch and dinner meals did not change significantly. The intake of carrots at the evening snack was also consistent across conditions. Thus, the total vegetable intake over the entire day increased from 270 ± 17 g vegetables in the 100% condition to 401 ± 20 g vegetables in the 85% condition (a 50% increase) and to 487 ± 25 g vegetables in the 75% condition (an 80% increase) (P < 0.0001).
FIGURE 1.
Mean (±SEM) vegetable intakes by condition at meals and evening snack consumed by 41 women and men who were served manipulated entrées at each meal that were reduced in energy density by varying the amount of puréed vegetables. Within each meal and snack, values with different letters were significantly different (P < 0.0001) as assessed by a mixed linear model with repeated measures and with a Tukey-Kramer adjustment for multiple comparisons.
TABLE 4.
Total food and energy intakes over 1 d for 41 participants in a study in which the energy density of entrées was reduced by varying the amount of puréed vegetables1
| 100% energy density | 85% energy density | 75% energy density | |
| Energy (kcal) | 3117 ± 132a | 2915 ± 118b | 2760 ± 110c |
| Weight (g) | 1775.8 ± 73.5 | 1812.6 ± 74.1 | 1806.4 ± 76.0 |
| Carbohydrate (g) | 390.6 ± 17.6a | 376.2 ± 16.2a,b | 362.7 ± 15.3b |
| Protein (g) | 89.0 ± 3.7a | 81.6 ± 3.4b | 75.4 ± 3.2c |
| Fat (g) | 133.2 ± 5.6a | 121.1 ± 4.9b | 113.0 ± 4.5c |
| Vegetable (g) | 269.6 ± 17.4a | 401.0 ± 20.3b | 487.4 ± 24.6c |
| Fiber (g) | 19.8 ± 0.8a | 22.0 ± 0.9b | 23.7 ± 1.0c |
| Energy density (kcal/g) | 1.76 ± 0.03a | 1.62 ± 0.03b | 1.54 ± 0.03c |
All values are means ± SEMs. Values in the same row with different superscript letters were significantly different (P < 0.0002) as assessed by a mixed linear model with repeated measures and with a Tukey-Kramer adjustment for multiple comparisons.
Food intake
The total weight of food consumed at each meal and snack (Figure 2A) and over the entire day (Table 4) did not differ significantly across conditions of entrée energy density. Intakes of manipulated entrées at lunch and dinner were not significantly different across conditions, but the intake of the breakfast entrée was higher in the 75% condition than in the 100% condition (P = 0.011). Participants consumed, on average, 113 ± 9 g of carrot bread in the 100% condition, 121 ± 9 g of carrot bread in the 85% condition, and 128 ± 10 g of carrot bread in the 75% condition. Participants consumed consistent weights of all unmanipulated side dishes and evening snacks across conditions (P > 0.27 for all items).
FIGURE 2.
Mean (±SEM) food (A) and energy (B) intakes by condition at meals and evening snack consumed by 41 women and men who were served manipulated entrées at each meal that were reduced in energy density by varying the amount of puréed vegetables. Within each meal and snack, values with different letters were significantly different (P < 0.02) as assessed by a mixed linear model with repeated measures and with a Tukey-Kramer adjustment for multiple comparisons.
Energy intake and energy density
The energy intake over the day significantly decreased as the energy density of entrées was reduced (P < 0.0001; Figure 2B). Compared with the 100% condition, participants consumed 202 ± 60 kcal less in the 85% condition and 357 ± 47 kcal less in the 75% condition. These differences were equivalent to mean decreases in the daily energy intake of 6% and 11%, respectively. The energy intake from unmanipulated side dishes and evening snacks did not differ significantly across conditions. Men consumed a mean of 73 ± 1.8% of the energy provided from unmanipulated side dishes and evening snacks, and women consumed a mean of 51% ± 1.5% of the energy provided from unmanipulated side dishes and evening snacks. None of the participants consumed the entire amount of side dishes and evening snacks provided over 1 d. Thus, the decrease in daily energy intake was a result of the reduction in energy intake from the manipulated entrées.
At lunch and dinner meals, the energy intake from the entrées significantly decreased as the energy density was reduced (P < 0.0001). The reduction in energy intake from the entrées paralleled the reduction in energy density. The energy intake from the breakfast entrée was only significantly different between the 100% and 75% conditions (P < 0.01). Although participants consumed more carrot bread in the 75% condition, the reduction of energy density by 25%, reduced the energy intake from the carrot bread by 14% compared with that in the 100% condition.
The dietary energy density over the day decreased significantly as the energy density of the 3 entrées was decreased (P < 0.0001). The mean dietary energy density was 1.76 ± 0.03 kcal/g entrée in the 100% condition, 1.62 ± 0.03 kcal/g entrée in the 85% condition, and 1.54 ± 0.03 kcal/g entrée in the 75% condition. Thus, the reduction of the energy density of the entrées by 15% and 25% decreased the dietary energy density by means of 8% and 13%, respectively.
Ratings of hunger, satiety, and food characteristics
Before meals, participant ratings of hunger, fullness, prospective consumption, nausea, and thirst did not vary significantly across conditions (data not shown). After meals, ratings of hunger and satiety were not significantly different across conditions with one exception. After breakfast, ratings of fullness were higher in the 75% condition (82 ± 2 mm) than in the 100% condition (76 ± 3 mm; P = 0.014), which was consistent with the differences in intakes. This higher rating did not influence ratings of fullness before the lunch meal. Ratings of fullness before consumption of the evening snack were not significantly different across conditions.
Ratings of pleasantness of appearance, taste, and texture for each of the manipulated entrées are shown in Table 5. All entrées were well liked by participants. Across conditions, there were no significant differences in ratings of pleasantness of appearance, taste, or texture for the chicken rice casserole, pleasantness of appearance or texture for the macaroni and cheese, or pleasantness of appearance for the carrot bread. The pleasantness of taste of the macaroni and cheese in the 75% condition was rated significantly lower than in the 100% condition (P = 0.0167). For the carrot bread, ratings of pleasantness of taste and pleasantness of texture were significantly higher in the 85% (P < 0.001) and 75% (P < 0.0001) conditions than in the 100% condition. Analysis of covariance showed that these differences in ratings for the macaroni and cheese and carrot bread did not significantly influence the relation between the experimental condition and intake of entrées.
TABLE 5.
Ratings of food characteristics for each of the manipulated entrées by 41 participants in a study in which the energy density of entrées was reduced by varying the amount of puréed vegetables1
| 100% energy density | 85% energy density | 75% energy density | |
| Carrot bread | |||
| Appearance | 63.8 ± 2.9 | 66.5 ± 2.7 | 68.9 ± 2.8 |
| Taste | 62.6 ± 3.7a | 75.7 ± 2.3b | 76.6 ± 2.1b |
| Texture | 56.1 ± 3.7a | 74.5 ± 2.6b | 77.0 ± 2.3b |
| Macaroni and cheese | |||
| Appearance | 65.4 ± 3.1 | 63.2 ± 3.7 | 59.0 ± 3.7 |
| Taste | 72.9 ± 2.9a | 68.0 ± 3.1a,b | 66.2 ± 2.9b |
| Texture | 64.2 ± 3.7 | 61.4 ± 3.7 | 63.2 ± 3.3 |
| Chicken rice casserole | |||
| Appearance | 63.2 ± 3.3 | 63.4 ± 2.9 | 60.2 ± 3.4 |
| Taste | 64.0 ± 3.0 | 64.8 ± 3.0 | 62.7 ± 2.8 |
| Texture | 67.1 ± 3.0 | 66.1 ± 2.9 | 64.7 ± 2.4 |
All values are mean ± SEM ratings from 100-mm visual analog scales. Values in the same row with different superscript letters were significantly different (P < 0.02) as assessed by a mixed linear model with repeated measures and with a Tukey-Kramer adjustment for multiple comparisons.
Influence of subject characteristics
Analysis of covariance showed that the relation between the entrée energy density and outcomes of total food, vegetable, and energy intakes was not significantly affected by participant age, sex, height, weight, BMI, or scores for dietary restraint, disinhibition, or hunger.
Ratings of the participant liking of puréed vegetables used to manipulate entrée energy density (carrots, yellow squash, and cauliflower) and the vegetable side dishes (broccoli and green beans) obtained from the food preference questionnaire are shown in Table 6. Analysis of covariance revealed that the participant liking of vegetables used to manipulate the entrées did not significantly influence the relation between the entrée energy density and intake of entrées (data not shown). For example, the 75% energy density version of the macaroni and cheese had more puréed cauliflower than the 100% version, but the participant liking for cauliflower did not influence how much macaroni and cheese was consumed across conditions.
TABLE 6.
Ratings of participant liking of vegetables used in a study in which the energy density of entrées was manipulated by varying the amount of puréed vegetables1
| Dislike strongly, dislike, or dislike somewhat |
Neither like nor dislike |
Like strongly, like, or like somewhat |
||||
| Vegetable | n | % | n | % | n | % |
| Carrots | 4 | 10 | 7 | 17 | 30 | 73 |
| Yellow squash | 14 | 36 | 12 | 31 | 13 | 33 |
| Cauliflower | 11 | 28 | 9 | 23 | 19 | 49 |
| Broccoli | 5 | 12 | 2 | 5 | 34 | 83 |
| Green beans | 11 | 27 | 5 | 12 | 25 | 61 |
Ratings were collected at the end of the study by using a 7-point scale. The puréed vegetables used to manipulate entrée energy density were carrots, yellow squash, and cauliflower; and the unmanipulated vegetable side dishes were broccoli and green beans.
Comments from the discharge questionnaire showed that 18 participants (44%) noticed differences in the appearance, taste, or texture of the different versions of the entrées, and 2 participants (5%) noticed differences in the vegetable content; in particular, participants commented on the difference in moistness of the carrot bread. The effect of the experimental manipulation on the main outcomes of food, vegetable, and energy intakes did not differ significantly between participants who did and did not notice differences between the entrées.
DISCUSSION
The findings from this study show that incorporating puréed vegetables into meals as a method of decreasing energy density can be an effective strategy to reduce energy intakes and increase vegetable intakes over 1 d. Participants consumed a similar weight of food across conditions, and therefore, when the energy density of the breakfast, lunch, and dinner entrées was reduced by 15% and 25%, daily energy intakes was reduced by 6% and 11%, respectively. In addition, vegetable intakes over the day increased by ≈50% in the 85% condition and by 80% in the 75% condition. Despite the reduction in daily energy intakes across conditions, participant ratings of hunger and fullness did not differ significantly. Adding puréed vegetables to foods is a simple strategy that can lead to large effects on daily energy and vegetable intakes.
The energy density of foods can be manipulated in a variety of ways such as decreasing the fat and sugar content and increasing the amount of water-rich fruit and vegetables. Many studies have used a combination of these methods (20–24), but only a few studies have incorporated vegetables as the primary method of reducing energy density (3–5). In these studies, vegetables were overtly substituted for higher-energy-dense ingredients in a mixed dish. The results showed that participants consumed a similar weight of food across energy densities and, thus, consumed less energy as the energy density was decreased. Despite these differences in energy intakes, ratings of hunger and satiety did not differ across conditions (3–5). The current study extended these findings by covertly incorporating puréed vegetables into foods to reduce the energy density and similarly showed reductions in energy intakes with no differences in hunger and satiety. In contrast to previous studies, the manipulated entrées in the current experiment were served with palatable side dishes that were not varied in energy density and could be consumed ad libitum. The finding that intakes of these side dishes, as well as the unmanipulated evening snacks, were consistent across conditions showed that the effects of energy density on intake can persist over 1 d, even when participants are given opportunities to compensate for reductions in energy intakes. It is possible that individuals could compensate by consuming more energy on subsequent days, and this should be explored in future studies. The results of this study added to the evidence that decreasing the energy density of foods by increasing vegetable content is an effective strategy to reduce energy intakes. Compared with the use of chopped or whole vegetables, puréed vegetables can be covertly incorporated into a wide variety of sweet and savory foods and therefore provide more opportunities for influencing energy intakes.
Substituting low-energy-dense vegetables for foods higher in energy density was recommended by several government agencies to help reduce energy intakes (1, 2). This strategy also has the potential to help increase vegetable intakes. One study investigated this approach by increasing the amount of whole vegetables served at a meal while decreasing the amount of grain and meat. The results showed that, when the portion of vegetables was increased by 90 g, the vegetable intake at the meal increased by about one-half of a serving; when the portion was increased by 180 g, the vegetable intake increased by three-quarters of a serving (6). A similar substitution strategy was used in the current study, except that puréed vegetables were used rather than whole vegetables. Comparable with the effects showed in the previous study, increasing the amount of puréed vegetables in each of the lunch and dinner entrées by 90 or 160 g increased the vegetable intake at each meal by one-half of a serving or three-quarters of a serving, respectively. As a result, the substitution increased vegetable consumption over the day by one additional serving of vegetables in the 85% condition and 2 additional vegetable servings in the 75% condition. Furthermore, because intakes of vegetable side dishes did not change, puréed vegetables in the entrées added to the overall vegetable intake. The strategy of substituting either whole or puréed vegetables for foods higher in energy density can be effective in increasing vegetable intakes and reducing energy intakes, but using puréed vegetables may be especially beneficial in individuals who dislike the taste of vegetables.
A dislike for the taste or texture of vegetables is a barrier to achieving recommended vegetable intakes for many Americans (7–9). The incorporation of puréed vegetables into foods may be one way to help individuals overcome this barrier. In the current study, the entrées were formulated so that the taste, texture, and appearance of each entrée remained similar as the amount of puréed vegetables was increased. Subject ratings of the entrée characteristics and comments from the discharge questionnaire revealed that some individuals noticed differences in the taste and texture of some entrées. For example, as the amount of vegetables increased in the carrot bread, ratings for pleasantness of taste and texture increased. On the basis of participant comments, the higher ratings were likely because of the increased moistness; this could also explain why participants consumed more carrot bread in the 75% condition. Despite consuming a greater amount, participants still consumed less energy from the carrot bread and the entire breakfast meal in the 75% condition than the 100% condition. At the lunch meal, taste ratings of the macaroni and cheese were significantly lower in the 75% condition than in the 100% condition, but this did not affect intakes of the entrée across conditions. The participant liking of the vegetables used to manipulate energy density (carrots, squash, and cauliflower) did not influence intakes of the entrées. These results suggested that large amounts of puréed vegetables can be incorporated into foods with only slight differences in palatability and that such differences are unlikely to affect intakes. Although covertly incorporating vegetables into entrées should not be advised as the only method of increasing vegetable consumption, this strategy provides an additional opportunity for meeting recommended intakes, particularly in individuals with a low liking for vegetables.
In conclusion, this study showed that substantial amounts of puréed vegetables could be effectively incorporated into a variety of sweet and savory foods to increase vegetable intakes and reduce energy intakes. In addition, these effects can persist over 1 d even when subjects are given the opportunity to compensate for decreases in energy intakes by consuming other palatable foods. Hiding vegetables in foods has been shown to be effective in children to reduce energy intakes and increase vegetable intakes (11), and the current study showed it was also effective in adults, including in adults who disliked the taste of vegetables. This simple strategy provides an opportunity that could be implemented in many settings; for example, it can be used by individuals at home, by restaurant chefs, or by the food industry to influence vegetable and energy intakes. The effect of this strategy needs to be investigated over time to determine whether it can have persistent effects on energy intakes that could affect the rising rates of obesity.
Acknowledgments
We acknowledge the contributions of the staff and students in the Laboratory for the Study of Human Ingestive Behavior at The Pennsylvania State University.
The authors’ responsibilities were as follows—ADB: design of the experiment, collection and analysis of data, and writing of the manuscript; LSR: analysis of data and writing of the manuscript; and BJR: design of the experiment and writing of the manuscript. BJR is the author of The Volumetrics Weight Control Plan (HarperTorch, New York, NY, 2003) and The Volumetrics Eating Plan (HarperCollins, New York, NY, 2007). None of the authors had a conflict of interest.
REFERENCES
- 1.US Department of Health and Human Services, Centers for Disease Control and Prevention Nutrition for everyone: healthy weight: how to use fruits and vegetables to help manage your weight. 2005 [Google Scholar]
- 2.Centers for Disease Control and Prevention, Division of Nutrition, Physical Activity and Obesity Can eating fruits and vegetables help people to manage their weight? Research to Practice Series, No. 1. Centers for Disease Control and Prevention, Atlanta, GA, 2005 [Google Scholar]
- 3.Kral TVE, Roe LS, Rolls BJ. Does nutrition information about the energy density of meals affect food intake in normal-weight women? Appetite 2002;39:137–45 [DOI] [PubMed] [Google Scholar]
- 4.Chang UJ, Hong YH, Suh HJ, Jung EY. Lowering the energy density of parboiled rice by adding water-rich vegetables can decrease total energy intake in a parboiled rice-based diet without reducing satiety on healthy women. Appetite 2010;55:338–42 [DOI] [PubMed] [Google Scholar]
- 5.Bell EA, Castellanos VH, Pelkman CL, Thorwart ML, Rolls BJ. Energy density of foods affects energy intake in normal-weight women. Am J Clin Nutr 1998;67:412–20 [DOI] [PubMed] [Google Scholar]
- 6.Rolls BJ, Roe LS, Meengs JS. Portion size can be used strategically to increase vegetable consumption in adults. Am J Clin Nutr 2010;91:913–22 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Glasson C, Chapman K, James E. Fruit and vegetables should be targeted separately in health promotion programmes: differences in consumption levels, barriers, knowledge and stages of readiness for change. Public Health Nutr (Epub ahead of print 25 June 2010) [DOI] [PubMed] [Google Scholar]
- 8.Appleton KM, McGill R, Neville C, Woodside JV. Barriers to increasing fruit and vegetable intakes in the older population of Northern Ireland: low levels of liking and low awareness of current recommendations. Public Health Nutr 2010;13:514–21 [DOI] [PubMed] [Google Scholar]
- 9.Lucan SC, Barg FK, Long JA. Promoters and barriers to fruit, vegetable, and fast-food consumption among urban, low-income African Americans–a qualitative approach. Am J Public Health 2010;100:631–5 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Stubbs RJ, Johnstone AM, O'Reilly LM, Barton K, Reid C. The effect of covertly manipulating the energy density of mixed diets on ad libitum food intake in 'pseudo free-living’ humans. Int J Obes Relat Metab Disord 1998;22:980–7 [DOI] [PubMed] [Google Scholar]
- 11.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;16:1559–65 [DOI] [PubMed] [Google Scholar]
- 12.Zung WWK. Zung self-rating depression scale and depression status inventory Sartorius N, Ban TA. Assessment of depression. Berlin, Germany: Springer-Verlag, 1986:221–31 [Google Scholar]
- 13.Garner DM, Olsted MP, Bohr Y, Garfinkel PE. The Eating Attitudes Test: psychometric features and clinical correlates. Psychol Med 1982;12:871–8 [DOI] [PubMed] [Google Scholar]
- 14.Stunkard AJ, Messick S. The three-factor eating questionnaire to measure dietary restraint, disinhibition and hunger. J Psychosom Res 1985;29:71–83 [DOI] [PubMed] [Google Scholar]
- 15.Littell RC, Milliken GA, Stroup WW, Wolfinger RD, Schabenberger O. SAS for mixed models. 2nd ed Cary, NC: SAS Institute Inc; 2006 [Google Scholar]
- 16.US Department of Agriculture, Agricultural Research Service USDA nutrient database for standard reference, release 21. 2008. Nutrient Data Laboratory Home Page. Available from: http://www.ars.usda.gov/ba/bhnrc/ndl (cited October 2009)
- 17.Flint A, Raben A, Blundell JE, Astrup A. Reproducibility, power and validity of visual analogue scales in assessment of appetite sensations in single test meal studies. Int J Obes Relat Metab Disord 2000;24:38–48 [DOI] [PubMed] [Google Scholar]
- 18.Ledikwe JH, Blanck HM, Kettel-Khan L, et al. Dietary energy density determined by eight calculation methods in a nationally representative United States population. J Nutr 2005;135:273–8 [DOI] [PubMed] [Google Scholar]
- 19.US Department of Health and Human Services, US Department of Agriculture Dietary guidelines for Americans 2005. 6th ed. Washington, DC: The National Academies Press, 2005 [Google Scholar]
- 20.Rolls BJ, Roe LS, Meengs JS. Salad and satiety: energy density and portion size of a first course salad affect energy intake at lunch. J Am Diet Assoc 2004;104:1570–6 [DOI] [PubMed] [Google Scholar]
- 21.Rolls BJ, Bell EA, Castellanos VH, Chow M, Pelkman CL, Thorwart ML. Energy density but not fat content of foods affected energy intake in lean and obese women. Am J Clin Nutr 1999;69:863–71 [DOI] [PubMed] [Google Scholar]
- 22.Leahy KE, Birch LL, Rolls BJ. Reducing the energy density of an entree decreases children's energy intake at lunch. J Am Diet Assoc 2008;108:41–8 [DOI] [PubMed] [Google Scholar]
- 23.Rolls BJ, Roe LS, Meengs JS. Reductions in portion size and energy density of foods are additive and lead to sustained decreases in energy intake. Am J Clin Nutr 2006;83:11–7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Bell EA, Rolls BJ. Energy density of foods affects energy intake across multiple levels of fat content in lean and obese women. Am J Clin Nutr 2001;73:1010–8 [DOI] [PubMed] [Google Scholar]