Obesity is the most common nutritional disorder, especially in industrialized countries (1). Childhood obesity has tripled over the past 3 decades in the United States and is progressively increasing in Europe (2,3). Fat storage is the result of a gene–environment interaction that causes a prolonged energy imbalance in which energy intake is greater than energy expenditure. Recent environmental and nutritional changes such as an increase in food portion sizes, changes in diet composition and reduction of physical activity promote fat gain (4–6). Focusing on nutritional risk factors, it has been shown that fat intake, energy density, and patterns of food intake play a role in the development and maintenance of obesity (4,7,8). Snacking can enhance the potential risk factor. Snacking on certain foods between meals may favor appetite control and a more adequate food intake as concerns daily requirements (9). Nevertheless, snack composition and frequency of intake are both crucial to fat gain.
Evidence exists that high-energy-dense snacks are less satiating, potentially contributing to an increase in energy intake (10,11). A higher consumption of energy-dense snacks may favor the increase in body size, although some studies in the United States do not support this hypothesis. Bandini et al showed that obese children did not eat more high-energy, low-nutrient-dense foods than nonobese children (12). The same authors confirmed these results in a longitudinal study on 196 8- to 12-year-old girls who were studied for 4 years (13). In particular, they did not find any relation between energy dense snacks and obesity, but they did find an association with the consumption of soft drinks (13).
Snack consumption seems to have increased in recent years. In the United States, between 1977 and 1996, snacking increased in all age groups (14). The average size of a snack and the energy per snack has stayed relatively constant, but the number of occasions for snacking has increased significantly (14). In Italy, a country with food preferences and diet composition different from the United States, but with the same high prevalence of childhood obesity, there are few data available on the relation between snack intake and overweight in children (15). D'Amicis et al studied the diet composition of 1036 children 6 to 14 years of age and found that snack intake accounted for ∼20% of total energy intake (15). In the Italian population, as in the United States, energy intake with snacks was not associated with overweight and obesity. However, the relation between the type of snacks consumed and obesity, adjusting for potential risk cofactors such as parents' body mass index (BMI) or television viewing, was never explored.
Therefore, the purposes of our study were to assess the kind and number of snacks consumed weekly in a sample of 8- to 10-year-old children, representative of the Italian population, and to assess the relationship between body size and the number and type of snacks consumed, adjusted for parents' BMI, television viewing (a gross index of sedentary behavior) and sports activity.
PATIENTS AND METHODS
We recruited 1837 8- to 10-year-old healthy children (male/female = 1) from primary schools in 3 different cities in Italy: Verona (north), Pisa (central), and Naples (south). To get a representative sample of children with different socioeconomic backgrounds and environmental conditions, we selected a number of school districts within each area, from town centers, modern suburbs and rural zones. The study was run in the spring of 2003.
Selection of Participants
From school registers in the selected areas, a sample of 8- to 10-year-old children was selected at random. The total number of children originally selected was 2400. A total of 169 children was excluded because of chronic disease with relevant disabilities or because they were absent from school the day of selection. The parents of 363 children refused to participate in the study. The final sample was made up of 1837 children (924 males, 913 females). Mean age was 9.3 (±0.9) years. Six hundred eight children were recruited in Verona (316 males, 292 females), 635 (307 males, 328 females) in Pisa, and 594 (301 males, 293 females) in Naples. The protocol was in accordance with the 1975 Declaration of Helsinki as revised in 1983. Informed consent was obtained from all of the parents.
Two questionnaires were filled out by a pediatrician with the mother or father of each child in each recruiting area (Verona, Pisa, Naples). On the first questionnaire, the children's height and weight measured at recruitment were recorded. Self-reported parents' height and weight were also recorded. The parents' self-reported height and weight corresponded closely to the measured height and weight (16,17). In addition, parents' reported how much time their children spent in front of a viewing screen (television/video games/computer) and in sports or organized physical activity every day. The validity of parents' report of time spent in front of a video screen was previously assessed by Anderson et al compared with videotaped observation (18). Parental reporting on outdoor playtime was validated by Burdette et al. (19).
A second questionnaire on the frequency of a child's snacking (food and drink) weekly was also filled out by a pediatrician with the mother or father of each child. The pediatrician asked how often children consumed each of the 22 snacks listed. The snacks were milk, tea, fruit juice, soft drinks, chocolate-flavored beverages, cookies (biscuits), crackers, french fries, chocolate candy bars, muffins or cupcakes, shortbread, bread and cheese, cold cuts, white pizza, bread, dry toast, toast, popcorn, cake, pizza, sandwiches, and ice cream. Serving sizes were of natural units or typical serving sizes.
Solid snacks were then divided depending on whether they were sweet or savory. Sweet foods were cookies, chocolate candy bars, muffins and cupcakes, shortbread, bread with chocolate spread, cake, and ice cream; savory snacks were french fries, crackers, bread and cheese, white pizza, bread, dry toast, toast, popcorn, pizza, and sandwiches.
Energy and nutrient intakes were calculated by multiplying the frequency of weekly consumption by the nutrient composition of the portion size for each specific snack listed. Energy and nutrient intake was calculated using the food composition database of the National Institute of Research for Food and Nutrition (20). Energy and nutrients of all of the foods were summed to obtain the total weekly intake for each child. Servings of specific snacks were converted into daily servings and total daily snack servings were calculated by summing across all of the snacks. Food quotient was calculated using the following equation: O2 consumption (l/day) = (0.966 × protein intake) + (2.019 × fat intake) + (0.829 × carbohydrate intake), and CO2 production (l/day) = (0.744 × protein intake) + (1.427 × fat intake) + (0.829 × carbohydrate intake), where the intake of protein, fat and carbohydrate is expressed in grams per day; FQ = VCO2/VO2(21).
Finally, the validity of the frequency of partial food intake (all food ingested except the 3 main meals of breakfast, lunch, and dinner) was assessed by comparison with average snack intake derived from a 3-day food record (the reference method) for 449 (225 girls, 224 boys) 8- to 10-year-old children. The Pearson correlation coefficient for energy intake was 0.37 for girls and 0.40 for boys (P < 0.001). No significant difference was found in the proportion of protein, carbohydrates, and fat between the 2 methods in both sexes.
Definition of Obesity and Overweight
Body mass index was calculated by dividing weight (kilograms) by height squared (meters) for each child. Overweight and obesity were defined using the BMI reference cutoffs proposed by the International Obesity Task Force (22). In particular, children with a BMI higher than the percentile corresponding to a BMI of 25 at age 18 were defined as overweight, whereas children with a BMI higher than the percentile corresponding to a BMI of 30 at age 18 were defined as obese. BMI z scores were also calculated according to the LMS method, using national BMI reference tables (23).
Data are shown as mean (SEM). Student unpaired t test was used to compare physical characteristics, time spent watching television and playing video games, and physical activity of the children divided into males and females. The Kruskal-Wallis test was used to compare these variables in males and females divided on the basis of their BMI z scores in 3 categories—normal weight, overweight, and obese children.
As a preliminary analysis (not shown), we compared, using the Student unpaired t test, the total number of snack servings per day, the savory and sweet groups of snack servings per day, as well as the macronutrient composition and energy density of snacks consumed by males and females. Because no statistical differences were found in any of the variables analyzed, we performed an additional analysis, putting males and females together as a single population. In the total sample, classified as normal weight, overweight, or obese, we compared the total number of servings per day, number of savory and sweet snack servings per day, and the distribution of servings during the day, using the Kruskal-Wallis test. Post hoc analysis was performed by Tukey test. The association between variables was assessed using the Pearson regression coefficient.
Finally, we performed a logistic regression analysis in 2 steps using body size as the dependent variable: overweight and obese children (on the basis of the International Obesity Task Force definition) versus normal-weight children (reference group). In the first step, sex, energy density of snacks, kind of snack (sweet or savory), intake of soft drinks with snack, television viewing per day, and hours spent at sports activity per week were used as independent variables. Television viewing and time devoted to the computer or playing video games were correlated (r = 0.16; P < 0.001). Similarly, hours of sports activity per week and hours per week devoted to moderate to vigorous nonprogrammed physical activity were correlated (r = 0.08; P < 0.01). Therefore, we used television viewing as an index of sedentary behavior and time devoted to sports per week as an index of physical activity. In the second step, to adjust the odds ratios (OR) for the effect of parental obesity, we included the mother's and father's BMI among the independent variables. All of the statistical analyses were carried out using SPSS version 13.0 software for Windows (SPSS Inc, Chicago, IL) package for personal computers. The probability level of P < 0.05 was used to indicate statistical significance in all of the analyses.
Physical Characteristics, Physical Activity and Television Viewing
The children's physical characteristics, amount of video exposure and hours per week of physical activity, divided into males and females, are reported in Table 1. The prevalence of overweight and obesity were 21% and 5.7%, respectively, in males, and 21.2% and 4.3%, respectively, in females: no statistical difference was found between sexes. The prevalence of overweight and obesity in the 3 areas studied was 14% and 3%, respectively, in Verona; 23% and 4%, respectively, in Pisa, and 27% and 13%, respectively, in Naples.
In the total sample, we did not find significant differences between boys and girls concerning physical characteristics and parents' BMI. Children spent an average of 2 hours per day in front of the television (males and females, P = 0.68). Boys spent significantly more time playing video games than girls (0.8 vs 0.4 hours/day, respectively; P < 0.001), but they were also more physically active than girls (4.7 vs 3.9 hours/week, respectively, of moderate/intense physical activity; P < 0.001 and 2.5 hours/week vs 2, respectively of sports; P < 0.001).
The physical characteristics, time spent in front of television and video games, and physical activity of the boys divided into normal weight, overweight, or obese are shown in Table 2. As expected, overweight and obese boys were taller, weighed more, and had higher BMI, BMI z scores, and parents' BMI than normal-weight boys. Overweight and obese boys spent less time in sports activities than normal-weight boys (2.4 vs 1.7 vs 2.6 hours/week, respectively; P = 0.03); no differences were found in time spent at sedentary activities or at moderate/intense physical activity among body size categories.
The physical characteristics, time spent in front of the television and video games, and physical activity of girls divided into normal weight, overweight, or obese groups are shown in Table 3. Overweight and obese girls were taller, weighed more, and had higher BMI, BMI z scores, and parents' BMI than normal-weight girls. As for boys, overweight and obese girls spent less time at sports than normal-weight girls (1.6 vs 1.1 vs 2.2 hours/week, respectively; P < 0.001). Moreover, overweight and obese girls spent more hours per week watching television than normal-weight girls (2 vs 2.5 vs 1.9 hours/week, respectively; P < 0.001). We did not find any other difference among body size categories.
Children consumed on average 4 snacks per day (boys: 3.9 [0.07]; girls: 3.8 [0.07], P = 0.27). In both sexes, the favorite snacks were, in order of preference: fruit juice, fruit, bread with cold cuts, milk, tea, soft drinks, brioche, crackers, yogurt, bread, and cookies. Children preferred salty snacks to sweet snacks; they consumed 8.4 (0.16) servings per week of savory snacks vs 7.2 (0.13) servings per week of sweet snacks (P < 0.001). We did not find any significant difference between sexes. In particular, savory snack intake was 8.6 (0.24) servings per week for boys and 8.3 (0.23) for girls (P = 0.5); sweet snack intake was 7.3 (0.19) servings per week for boys and 7 (0.19) for girls (P = 0.2). Consumption of soft drinks was not statistically different in the 2 sexes (males vs females: 1.09 [0.06] vs 0.9 [0.06]; P = 0.09); nor was the energy density (6.4 [0.08] vs 6.5 [0.08] kJ/g; P = 0.3). Boys consumed one serving per day of low-energy-dense foods, 0.5 servings of medium-energy-dense foods, and 1.2 servings of high-energy-dense foods. Girls consumed 1.1 daily servings of low-, 0.4 servings of medium-, and 1.2 servings of high-energy-dense foods.
The number of snack servings per day, the energy content and the macronutrient composition of snacks were not different in boys and girls (Table 4). Therefore, in further analysis, we combined boys and girls in an overall comparison among children of different weight categories: normal weight, overweight, or obese (Table 5).
The obese children consumed more servings of snacks per day compared to overweight and normal weight children, but this difference did not assume any statistical significance (4.4 vs 3.9 vs 3.9, respectively; P = 0.06) (Fig. 1). Similarly, energy intake with snacks per day as well as food quotient and macronutrient composition of the snacks were not statistically different among the 3 groups of children. The mean energy density of the foods consumed was significantly different among the weight categories: 6.3 (0.08) kJ/g in normal weight, 6.8 (0.16) kJ/g in overweight, and 6.8 (0.3) kJ/g in obese children (P < 0.05). Obese children consumed significantly (P < 0.001) more servings of soft drinks per day (2.09 [0.29]) than overweight (1.08 [0.1]) or normal weight children (0.96 [0.05]).
Savory snack intake was significantly different in the 3 weight categories (χ2 = 11.8; P = 0.003) as opposed to that of sweet snacks (χ2 = 5.9; P = 0.05). In particular, obese children consumed 10.8 (0.8) servings per week of savory snacks versus 8.4 (0.7) of sweet snacks; overweight children consumed 9.1 (0.42) servings per week of savory snacks versus 6.9 (0.32) of sweet snacks; normal weight children consumed 8.3 (0.18) servings per week of savory snacks versus 7.2 (0.15) of sweet snacks.
Correlation and Regression Analysis
Correlation analysis showed that BMI z scores were positively associated with parents' BMI (r = 0.25, P < 0.001 for both mother's and father's BMI), television viewing (r = 0.10, P < 0.001), soft drink consumption (r = 0.07, P < 0.01), and savory snack (r = 0.06, P < 0.05) intake; BMI z scores were negatively associated with programmed physical activity (r = −0.10, P < 0.01) (Table 6). Mother's BMI correlated with BMI z scores, television viewing (r = 0.15, P < 0.001), soft drink intake (r = 0.07, P < 0.01), and savory snack intake (r = 0.05, P < 0.05), but correlated negatively with programmed physical activity (−0.06, P < 0.05). Father's BMI correlated with BMI z scores, television viewing (r = 0.08, P < 0.001), soft drink (r = 0.07, P < 0.01) and savory snack intake (r = 0.06, P < 0.05), food quotient (r = 0.06, P < 0.05), and high-energy-dense snacks (r = 0.08, P < 0.05). Mother's and father's BMI were positively correlated (r = 0.23, P < 0.001). The daily hours of television were correlated with playing video games (r = 0.16, P < 0.001), energy intake with snack (r = 0.09, P < 0.001), soft drink intake (r = 0.13, P < 0.001), savory snack (r = 0.07, p < 0.01) and sweet snack (r = 0.09, P < 0.01) intake, and with medium- and high-energy-dense snack intake (r = 0.11, P < 0.001 and r = 0.07, P < 0.01, respectively). Moreover, television viewing was negatively associated with programmed physical activity (r = −0.08, P < 0.01). Programmed physical activity was correlated, other than with BMI z scores and television viewing, with food quotient (r = 0.06, P < 0.05) and energy density of snacks (r = −0.07, P < 0.01). Food quotient was negatively correlated with the energy density of snacks (r = 0.34, P < 0.001). The energy density of snacks was significantly (P < 0.001) correlated with salty-tasting snacks (r = 0.18) as well as with sweet-tasting snacks (r = 0.30).
Logistic regression analysis showed that the energy density of snacks, intake of soft drinks with snacks, intake of savory snacks, television viewing, and hours of sports activity per week independently contributed to predict obesity in children (Table 7). However, when parents' BMI was included among the independent variables of the regression, television viewing, energy density of snacks, and intake of soft drinks with snacks were no longer significant (Table 8). The same analysis revealed that for each serving of savory snacks per week, the children increased their likelihood to become obese by 2% (OR 1.02, 95% CI 1.002–1.04) and that for each hour of programmed physical activity per week, the children had a 10% lower risk of becoming obese (OR 0.90, 95% CI 0.84–0.96).
The main results of this study were obese children did not consume more snacks than nonobese children; obese children preferred energy-dense snacks, especially those with a salty taste, compared to nonobese children; each serving per week of salty snacks increase by 2% the likelihood of children becoming obese, independently from age, sex, parents' BMI, lifestyle (sport or sedentary behavior), and energy density of snacks.
The increase in eating frequency—healthy snacking between the 2 main meals—may be helpful in reducing appetite and thereby favor a more appropriate daily food intake as regards daily requirements. A study of 4370 German children ages 5 to 6 years reported a protective effect of meal frequency on childhood obesity, independent of confounders such as parental education, parental obesity, television viewing, breast-feeding, physical activity, smoking during pregnancy, and regular snacking while watching television (24). Similar results were reported for adults (25–27).
Few data on the relationship between snacking and adiposity are available on children. A prospective study of 8203 girls and 6774 boys, 9 to 14 years of age, found that snack foods were not associated with subsequent changes in BMI z scores, after controlling for age, pubertal development, height change, physical activity, inactivity, dieting status, and mother's overweight (28). Moreover, a 10-year longitudinal study by Phillips et al on 196 nonobese prepubertal girls showed that energy-dense snack intake was not associated with BMI z score or body fat percentage over the adolescent period (13). In particular, the average intake of energy-dense snacks in these girls was 2.3 servings per day (16% of total energy) (13). Interestingly, they observed a significant relation between the consumption of soft drinks and BMI z scores over the 10-year study period (13). In Italy, D'Amicis et al studied the diet composition of 1036 children ages 6 to 14 years and found that snack intake accounted for ∼20% of total energy intake (15). In this population, as in the United States, energy intake with snacks was not associated with overweight and obesity (15).
In our study, we did not find differences in the number of daily servings consumed between obese and nonobese children. However, there were differences in the energy content of foods consumed. Obese and overweight children tended to consume more food with high- or medium-energy density than normal-weight children. The difference in the energy density of snacks between obese/overweight and normal-weight children was apparently small (∼5 kJ/g). However, taking the total weight of food consumed at snacks (∼400 g/day in both obese/overweight and nonobese children) into consideration, the difference of energy intake with snacks, due to the different density, is ∼200 kJ/day (ie, 400 g × 0.5 kJ/g). Interestingly, this rate of energy is ∼2% of the estimated total energy requirement of a 8- to 10-year-old child (29). Studies of energy balance clearly demonstrated that a subtle chronic extra-energy intake is responsible for significant fat gain in the long term (30). The energy density of the snacks was correlated with other risk factors for obesity such as father's BMI, television viewing, and little physical activity. In particular, an increase in father's BMI, increases in the intake of high-energy-dense snacks, and the hours per day of television viewing were positively associated with the energy density of foods. The hours per week of programmed physical activity (sports) were inversely related to the energy density of snacks. Moreover, the hours of sports activity were positively correlated with food quotient, suggesting that physical activity may affect food preferences, favoring the consumption of snacks high in carbohydrates and low in fat.
The taste of food is another potential contributing factor to obesity. Obese children consumed more salty- than sweet-tasting foods. The energy density of snacks was more correlated with a savory taste of snacks than with a sweet taste. In addition, parents' BMI was associated with savory-tasting snack intake but not with sweet snacks. However, logistic regression analysis showed that even when adjusting for parents' BMI, the likelihood of becoming obese increased by 2% with each serving per week of salty snacks, regardless of energy density. Our results are in agreement with Cox et al who studied lean and obese healthy adults and found that the obese consumed a diet with a significantly higher energy density and significantly more salty food intake than lean adults (31). Moreover, in the obese group, the percentage of salty foods eaten correlated strongly with energy density.
Our results confirmed that parents' BMI is associated with childhood obesity. Television viewing, which accounted for 15% of the likelihood to be obese in the first analysis, was no longer significant after adjusting for parents' BMI. Mother's and father's BMI contributed to explain childhood obesity more than the energy density of snacks, taste of snacks, or sedentary lifestyle (television viewing). This result is in agreement with the hypothesis of the obesigenic environment in the family, in which parents are more overweight and who have less healthy eating habits and an inactive lifestyle (32). Children from obesigenic families consumed more fat in their diet and reported higher levels of television viewing. The effect of this behavior, when it starts early (at the age of 5 years for instance) can have long-term effects on children's BMI and their dietary and activity habits (33,34). Similar results were reported by Nguyen et al after studying the fat intake and fat mass in children 4 to 7 years of age with lean or obese parents. They found that maternal obesity may have a greater influence on fat intake in children and that fat intake was related to fat mass in boys, regardless of the amount of physical activity (33).
Family contribution to obesity is also determined by genes. Data obtained in twins suggest that genes affect both food intake (composition, size and frequency of meals) and energy expenditure (about 40% of the variance in resting metabolic rate, thermic effect of food, and energy cost of low- to moderate-intensity exercise is explained by genes) (35,36). Several studies have tried to identify genes that may underlie susceptibility/resistance to obesity in the general population, but at present the results are not conclusive (37,38).
The hours per week of programmed physical activity had a protective effect on obesity, similar to that observed in other studies, in which the level of physical activity was negatively associated with overweight (39–41). Moreover, a recent study showed that there is a negative association between the clustering of risk factors related to the metabolic syndrome (ie, insulin resistance, lipid profile, blood pressure) and physical activity in 9- and 15-year-old schoolchildren, regardless of the degree of adiposity (42). The authors concluded that children could prevent insulin resistance if they did at least 90 minutes of activity per day (42).
In conclusion, obese children consumed more energy-dense food—especially salty food—when snacking than nonobese children. A sedentary lifestyle was associated with this behavior. The reasons for this relation are unknown. However, the influence of parents' food habits and lifestyle on those of their children is emphasized by the association between parents' obesity and their children's energy-dense food intake at snacktime, the savory taste of snacks, and sedentary behavior. Nevertheless, adjusting for parents' BMI, the savory taste of snacks and time (hours per week) devoted to sports activity were independently associated with childhood obesity. The potential causal relationship between the preference for salty snacks and the development of obesity in children needs to be investigated further.
1. Flegal KM, Carrol MD, Ogden CL, et al
. Prevalence and trends in obesity
among US adults, 1999–2000. JAMA 2002; 288:1723–1727.
2. Ogden CL, Flegal KM, Carrol MD, et al
. Prevalence and trends in overweight among US children
and adolescents, 1999–2000. JAMA 2002; 288:1728–1732.
3. Lobstein T, Frelut ML. Prevalence of overweight among children
in Europe. Obes Rev 2003; 4:195–200.
4. Maffeis C. Aetiology of overweight and obesity
and adolescents. Eur J Pediatr 2000; 159(Suppl 1):S35–S44.
5. Young LR, Nestle M. The contribution of expanding portion sizes to the US obesity
epidemic. Am J Public Health 2002; 92:246–249.
6. Prentice AM, Jebb SA. Obesity
in Britain: gluttony or sloth? BJT 1995; 311:437–439.
7. Maffeis C, Pinelli L, Schutz Y. Fat intake and adiposity in 8 to 11-year-old obese children
. Int J Obes 1996; 20:170–174.
8. Maffeis C, Provera S, Filippi L, et al
. Distribution of food intake as a risk factor for childhood obesity
. Int J Obes 2000; 24:75–80.
9. Jackson SJ, Leahy FE, Jebb SA, et al
. Frequent feeding delays the gastric emptying of a subsequent meal. Appetite 2007; 48:199–205. Epub November 1, 2006.
10. Marmonier C, Chapelot D, Louis-Sylvestre J. Metabolic and behavioural consequences of a snack consumed in a satiety state. Am J Clin Nutr 1999; 70:854–866.
11. 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(1):11–17.
12. Bandini LG, Vu D, Must A, et al
. Comparison of high-calorie, low-nutrient-dense food consumption among obese and non-obese adolescents. Obes Res 1999; 7(5):438–443.
13. Phillips SM, Bandini LG, Naumova EN, et al
. Energy dense snack food intake in adolescence: longitudinal relationship to weight and fatness. Obes Res 2004; 357:505–508.
14. Jahns L, Siega-Riz AM, Popkin BM. The increasing prevalence of snacking among US children
from 1977 to 1996. J Pediatr 2001; 138:493–498.
15. D'Amicis A, Intorre F, Maccati F, et al
. Studi sui consumi alimentari e ripartizione dei pasti degli scolari dell'obbligo in Italia (SCARPS). Riv Sci Aliment 2002; 31:235–248.
16. Stunkard AJ, Albaum JM. The accuracy of self-reported weights. Am J Clin Nutr 1981; 34:1593–1599.
17. Schlichting P, Høilund-Carlsen PF, Quaade F. Comparison of self-reported and measured height and weight with controlled height and weight in women and men. Int J Obes 1981; 5:67–76.
18. Anderson DR, Field DE, Collins PA, et al
. Estimates of young children
's time with television: a methodological comparison of parent reports with time-lapse video home observation. Child Dev 1985; 56:1345–1357.
19. Burdette HL, Whitaker RC, Daniels SR. Parental report of outdoor playtime as a measure of physical activity in preschool-aged children
. Arch Pediatr Adolesc Med 2004; 158:353–357.
20. Carnovale E, Marletta L, Miuccio F. Tabelle di composizione degli alimenti. Istituto Nazionale della Nutrizione, Roma, 1995.
21. Jequier E, Acheson K, Schutz Y. Assessment of energy expenditure and fuel utilization in man. Ann Rev Nutr 1987; 7:187–208.
22. Cole TJ, Bellizzi MC, Flegal KM, et al
. Establishing a standard definition for child overweight and obesity
worldwide: international survey. BMJ 2000; 320:1240–1243.
23. Luciano A, Bressan F, Zoppi G. Body mass index reference curves for children
aged 3–19 years from Verona, Italy. Eur J Clin Nutr 1997; 51(1):6–10.
24. Toschke AM, Kuchenhoff H, Koletzko B, et al
. Meal frequency and childhood obesity
. Obes Res 2005; 13:1932–1938.
25. Fabry P, Hejl Z, Fodor J, et al
. The frequency of meals. Its relation to overweight, hypercholesterolaemia, and decreased glucose-tolerance. Lancet 1964; 18:614–615.
26. Ruidavets JB, Bongard V, Bataille V, et al
. Eating frequency and body fatness in middle-aged men. Int J Obes Relat Metab Disord 2002; 26(11):1476–1483.
27. Berteus Forslund H, Torgerson JS, Sjostrom L, et al
. Snacking frequency in relation to energy intake and food choices in obese men and women compared to a reference population. Int J Obes (Lond) 2005; 29(6):711–719.
28. Field AE, Austin SB, Gillman MW, et al
. Snack food intake does not predict weight change among children
and adolescents. Int J Obes 2004; 28:1210–1216.
29. SINU. Livelli di assunzione raccomandati di energia e nutrienti per la popolazione italiana (LARN). Milano: EDRA; 1996.
30. Swinburn BA, Jolley D, Kremer PJ, et al
. Estimating the effects of energy imbalance on changes in body weight in children
. Am J Clin Nutr 2006; 83(4):859–863.
31. Cox DN, Perry L, Moore PB, et al
. Sensory and hedonic associations with macronutrient and energy intakes of lean and obese consumers. Int J Obes Relat Metab Disord 1999; 23:403–410.
32. Davison KK, Francis LA, Birch LL. Re-examining obesigenic families: parents'obesity
-related behaviours predict girls' change in BMI. Obes Res 2005; 13:1980–1990.
33. Nguyen VT, Larson DE, Johnson RK, et al
. Fat intake and adiposity in children
of lean and obese parents. Am J Clin Nutr 1996; 63:507–513.
34. Krahnstoever Davison K, Francis LA, Birch LL. Reexamining obesigenic families: parents' obesity
-related behaviors predict girls' change in BMI. Obes Res 2005; 13:1980–1990.
35. Wade J, Milner J, Krondl M. Evidence for a physiological regulation of food selection and nutrient intake in twins. Am J Clin Nutr 1981; 34:143–147.
36. Bouchard C, Tremblay A. Genetic effects in human energy expenditure components. Int J Obes 1990; 14:49–59.
37. Barsh GS, Farooqi IS, O'Rahilly S. Genetics of body weight regulation. Nature 2000; 404:644–651.
38. Rankinen T, Zuberi A, Chagnon YC, et al
. The human obesity
gene map: the 2005 update. Obesity
(Silver Spring) 2006; 14:529–644.
39. Rennie KL, Livingstone MB, Wells JC, et al
. Association of physical activity with body-composition indexes in children
aged 6-8 y at varied risk of obesity
. Am J Clin Nutr 2005; 82:13–20.
40. Patrick K, Norman GJ, Calfas KJ, et al
. Diet, physical activity, and sedentary behaviors as risk factors for overweight in adolescence. Arch Pediatr Adolesc Med 2004; 158(4):385–390.
41. Epstein LH, Paluch RA, Gordy CC, et al
. Decreasing sedentary behaviors in treating pediatric obesity
. Arch Pediatr Adolesc Med 2000; 154:220–226.
42. Andersen LB, Harro M, Sardinha LB, et al
. Physical activity and clustered cardiovascular risk in children
: a cross-sectional study (The European Youth Heart Study). Lancet 2006; 368:299–330.