Nutrition
Volume 26, Issue 1 , Pages 61-67, January 2010

Physical activity mediates the protective effect of the Mediterranean diet on children's obesity status: The CYKIDS study

Department of Nutrition and Dietetics, Harokopio University, Athens, Greece

Received 3 July 2008; accepted 18 May 2009. published online 27 July 2009.

Article Outline

Abstract 

Objective

There is some evidence regarding the association between adherence to the Mediterranean diet (MD) and obesity among adults; to our knowledge, however, no relevant data exist for children. We investigated the association between adherence to the MD and obesity status in children.

Methods

A national cross-sectional study among 1140 children (mean age 10.7±0.98 y) was carried out in Cyprus using stratified multistage sampling design. Body mass index was calculated according to International Obesity Task Force criteria, from parental reference. Adherence to the MD was assessed by the Mediterranean Diet Quality Index for children and adolescents (KIDMED diet score). To test the research hypothesis, a logistic regression analysis was applied with two dependent variable categories of obesity status, normal weight (NW) versus overweight/obese (OW/OB), and the three categories of the KIDMED score independently, after controlling for several potential confounders.

Results

Compared with low MD adherers, children with a high KIDMED score were 80% less likely to be OW/OB (95% confidence interval 0.041–0.976), adjusted for age, gender, parental obesity status, parental educational level, and dietary beliefs and behaviors (model 2). When physical activity was taken into account, however, the aforementioned relation was not significant (model 3; odds ratio 0.20, 95% confidence interval 0.021–1.86). Furthermore, male gender, maternal obesity, and dietary beliefs and behaviors emerged as more significant in predicting obesity in children compared with their KIDMED score.

Conclusion

Adherence to the MD is inversely associated with obesity in this sample of 9- to 13-y-old children; however, other behaviors, and in particular physical activity, maternal obesity, dietary beliefs and behaviors, seem to be more significant.

Keywords: Mediterranean diet, Obesity, Nutrition, Diet quality, Physical activity, Cyprus

 

Back to Article Outline

Introduction 

Regarded as a prototype model for healthy eating, the Mediterranean diet (MD) has generally been associated with longevity [1] and a lower burden of several morbid conditions including cardiovascular disease, diabetes mellitus, arthritis, and cancer [1], [2], [3], [4], [5], [6], [7].

Obesity is regarded as a common risk factor for the aforementioned morbid conditions and its prevalence is not quite as high in Mediterranean countries as it is in the United States and the rest of the developed world [8]. Given that the MD has been regarded as a healthy eating pattern, the investigation of the relation of adherence to the MD in the prevention and management of obesity has been an area of special interest [9]. Such studies, however, are rather limited; they present conflicting results and are relevant to only adult populations [9]. Therefore, to reach a definitive conclusion on the existence, direction, and strength of any such relation, there is a need for more well-designed studies involving more diverse populations. Moreover, it would be of additional interest to investigate the association of adherence to the MD with obesity status in children. Apart from the research interest that emerges from the aforementioned studies involving adults, there is emerging evidence from studies in children that suggests that certain MD characteristics, such as an increased consumption of plant foods, high-fiber foods, and a decreased intake of sweetened beverages and other foods high in simple carbohydrates [10], may be influential in the prevention and management of childhood obesity; in addition, the MD may be a health-promoting factor in children. Specifically, adherence to the MD has been linked in pediatric populations with nutrient adequacy and quality of diet [11], and it has been shown that the MD may be a protective factor for diseases, such as asthma and hypertension, for which obesity is known to be a risk factor in development and in deterioration [12], [13], [14], [15].

Consequently, we investigated the association between the degree of adherence to the MD and obesity status in a nationwide, free-living sample of children living in Cyprus, a Mediterranean country and a member of the European Union.

Back to Article Outline

Materials and methods 

Study sample 

The Cyprus Kids Study (CYKIDS) is a nationwide, cross-sectional survey that was conducted during school years 2004–2005 and covered all freely accessed districts of the Republic of Cyprus. Sampling was multistage (four stages) and stratified by the number of students in each of the five provinces, as provided by the Ministry of Education and by place of residence (school location was used as a proxy), urban or rural, as provided by the Cyprus Statistical Service [16].

A total of 1589 children in the fourth, fifth, and sixth grades (9–13 y of age) in 24 primary schools were identified for potential enrollment; 1140 agreed to participate (72% participation rate), representing 3.7% of the total population.

The study design was approved by the ethics committee of the Harokopio University and by the Cyprus Ministry of Education and Culture (Department of Primary Education) as required by law in Cyprus for studies carried out at schools during formal school hours. An informed consent was signed by the parent or guardian of each participant.

Assessment of adherence to the MD 

We assessed the children's adherence to the MD by applying the Mediterranean Diet Quality Index for children and adolescents (KIDMED). The index [17] is derived from 16 components that summarize the principles of the MD prototype and provide an arithmetic score that ranges from 0 to 12; a score of 0–3 reflects a poor diet relative to the MD principles, whereas values between 4–7 and 8–12, respectively, signify average and good adherence to the principles of the MD. More specifically, the KIDMED index presumes a daily consumption of at least one serving of fruit and vegetables, and consumption of at least two servings of each group is preferable. The recommended level of consumption of dairy products is at least three servings daily: one dairy product for breakfast and at least two servings of yogurt and/or cheese during the rest of the day. Consumption of grains and cereals is recommended daily for breakfast, and pasta or rice should be consumed at least five times per week. A weekly consumption of at least two to three servings of nuts and fish and two servings of pulses is advisable. Olive oil is recommended for culinary use, but no frequency is suggested. Dietary behaviors that are viewed as detrimental and alien to the principles of the MD include the frequent intake of sweets and candies (we defined this as more than twice daily), the use of commercially baked goods and pastries for breakfast, consumption of fast-foods, and non-consumption of breakfast.

Assessment of other covariables 

Dietary beliefs and behaviors 

A Short Eating Habits Behaviors and Beliefs Questionnaire, aimed at evaluating children's dietary beliefs (such as body image and quality of diet/self-perception) and dietary behaviors (such as dieting) was used. Response categories for these questions were “none,” “some,” “much,” and “very much.” This tool is presented in detail elsewhere [18]. Test–retest repeatability was tested by giving the same questionnaire in 100 children of the sample, within 1 mo apart, and was found to be good (median Spearman's ρ correlation coefficient 0.484).

Sociodemographic variables 

Data on sociodemographic characteristics such as age, gender, place of residence, and family size were provided by the children themselves. Information on the parents' educational level, income, and parents' occupation was collected by means of a short questionnaire that was completed by the parents. Socioeconomic status was assigned based on the parents' educational level and occupation as has been proposed by others [19].

Physical activity variables 

Children responded to a 32-item, semiquantitative questionnaire, which assessed organized and free-time physical activity. Principal component analysis (varimax rotation) was employed to extract the main factors from 21 variables from this questionnaire assessing children's frequency and duration of physical activity.

Eight factors emerged as significant, explaining approximately 63% of the total variance in children's physical activity and sedentary patterns. These are “physical activity and sports after school”; “home chores and outside home chores, aerobics, gymnastics, sports”; “sports for all, after-school activities (except sports)”; “video, electronic games and computers”; “watching TV, video and DVD”; “homework and private lessons”; “theater, cinema, use of mobile phone”; and “afternoon sleep, fewer private lessons” [20].

Anthropometry and obesity index 

The children's height and weight were reported by parents. Obesity and overweight among children were calculated using the International Obesity Task Force age- and sex-specific body mass index (BMI) cutoff criteria [21]. Parents' obesity was also calculated from self-reported values of body weight and height. Parents' overweight and obesity were assigned for BMI values from 25 to 29.9 and ≥30kg/m2, respectively.

Statistical analysis 

Continuous variables are presented as mean ± standard deviation, and categorical variables are presented as relative frequencies. Normality of variable distributions was tested by the Kolmogorov-Smirnov test. Associations between normally distributed variables were tested by t test, and the Mann-Whitney test was used for non-normally distributed continuous variables. Associations between categorical variables were tested by contingency tables and chi-square test. We further applied multiple logistic regression analysis evaluating the association between the level of obesity status and the level of adherence to the MD (as assessed by the KIDMED score) adjusted for potential confounders (i.e., age, gender, parental education level, the three factors of dietary beliefs and behaviors that emerged from the principal components analysis [18]: “guilty about eating,” “concerned about own body weight,” and “eating all my food,” parental obesity status, and physical activity levels as assessed by principal component analysis [20]). All reported P values are based on two-sided tests and 0.05 was set as the level of statistical significance. Statistical analyses were performed using SPSS 13.0 (SPSS Inc., Chicago, IL, USA).

Back to Article Outline

Results 

Characteristics of sample 

Body mass index and obesity status were estimated for 823 children (73% of participants), representing 2.7% of the reference population in Cypriot children of corresponding ages 9 to 13 y. Overall prevalence rates were 18.6% for overweight and 4.3% for obesity (for detailed data, see Lazarou et al. [22]).

Table 1 presents the characteristics of the study population by level of obesity status (normal weight versus overweight/obese).

Table 1. Selected demographic and lifestyle characteristics of the study sample by obesity status∗
Normal weightOverweight (including obesity)P
(n=635)(n=188)
Gender 0.198
Male75.124.9
Female78.921.1
Age (y)10.73±0.9510.56±0.970.037
Place of residence 0.322
Urban75.924.1
Rural78.821.2
KIDMED index categories 0.032
Poor-quality diet (0–3)40.737.8
Average-quality diet (4–7)52.460.3
Good-quality diet (8–12)7.01.9
Maternal education status 0.940
Primary school7.46.8
Secondary12.014.2
Higher secondary38.936.9
College38.238.1
Postgraduate3.64.0
Paternal education status 0.467
Primary school8.312.5
Secondary15.917.6
Higher secondary38.435.2
College28.425.6
Postgraduate
Father's obesity status 0.005
Normal weight42.630.6
Overweight/obese57.469.4
Mother's obesity status <0.0001
Normal weight76.061.5
Overweight/obese24.038.5
Factors of children's dietary beliefs and behaviors
Factor 1: guilty about eating−0.05±0.970.23±1.09<0.0001
Factor 2: concerned about body weight−0.18±0.890.42±1.23<0.0001
Factor 3: eating all my food0.03±1.01−0.29±0.87<0.0001
Physical activity factors (from PCA)
Component 1: physical activity and sports after school−0.05±0.94−0.05±1.060.599
Component 2: video, electronic games and computers−0.02±0.930.02±1.130.973
Component 3: watching TV, video and DVD−0.01±0.970.07±1.000.477
Component 4: homework and private lessons−0.02±0.960.03±0.940.516
Component 5: home chores and outside home chores, aerobics, gymnastics, sports0.02±1.00−0.07±0.990.341
Component 6: theater cinema, use of mobile phone0.09±1.02−0.06±1.020.144
Component 7:afternoon sleep, less private lessons0.05±0.990.01±1.100.665
Component 8: sports for all, after-school activities (except sports)0.06±1.00−0.06±0.960.230

KIDMED, Mediterranean Diet Quality Index for children and adolescents; PCA, principal component analysis.

∗Continuous variables are presented as mean±SD, and categorical variables as percentages.

Significant differences were observed between the two groups of obesity status in relation to maternal and paternal obesity statuses, the children's quality of diet (as assessed by KIDMED score), and the three factors describing certain children's beliefs and behaviors. More specifically, with regard to diet quality (KIDMED score), it seems that a significantly larger percentage of children who were high MD adherers (i.e., KIDMED score 8–12) were more likely to be of normal weight compared with low (i.e., KIDMED score 0–3) or medium (i.e., KIDMED score 4–7) level MD adherers (Table 1). Further details about the association of children's obesity status and other demographic and lifestyle factors are presented in Table 1.

Multivariable analyses 

We applied logistic regression analysis to control for various potential confounders (i.e., children's age, gender, place of residence, parental educational level, factors of children's dietary beliefs and behaviors, parental BMI class, and physical activity) and estimate the magnitude of the effect of any observed differences.

In the first model, when we controlled for all but the two last confounders (i.e., parental BMI class and physical activity), it appeared that children who had a high KIDMED score were 83% less likely to be overweight or obese. In the second model, we added the confounders of maternal and paternal weights and, although this association became weaker and confidence intervals widened more, it remained significant and the mean odds ratio was about the same (odds ratio 0.20, 95% confidence interval 0.04–0.98).

In the third model, we added the factors of physical activity and inactivity and the inverse relation that was observed in previous models was subsequently shown to be insignificant (Table 2). Instead, other factors such as age, gender, maternal obesity status, and factors of type 2 and 3 of children's behavior (i.e., “concerned about body weight,” “eating all my food”) were significantly associated with children's obesity status.

Table 2. Odds ratios (95% confidence intervals) derived from logistic regression analysis, showing the association between children's obesity status (normal versus overweight/obese) and adherence to Mediterranean diet
Model 1∗Model 2†Model 3‡
Age (per 6 mo)0.60 (0.47–0.76)0.60 (0.47–0.77)0.57 (0.41–0.78)
Boys111
Girls0.57 (0.36–0.88)0.55 (0.35–0.88)0.47 (0.25–0.88)
KIDMED score
Poor-quality diet (0–3)111
Average-quality diet (4–7)1.48 (0.94–2.33)1.50 (0.93–2.44)1.80 (0.97–3.34)
Good-quality diet (8–12)0.17 (0.03–0.82)0.20 (0.04–0.98)0.20 (0.02–1.86)
Maternal education status
Primary school111
Secondary school1.26 (0.44–3.65)1.26 (0.41–3.87)1.67 (0.43–6.53)
Higher secondary school1.62 (0.60–4.41)1.43 (0.48–4.22)2.20 (0.56–8.60)
College–university2.40 (0.85–6.79)2.47 (0.80–7.65)2.92 (0.70–12.13)
Postgraduate3.70 (0.77–17.63)3.67 (0.71–18.97)3.29 (0.42–25.81)
Paternal education status
Primary school111
Secondary school0.82 (0.34–1.99)0.84 (0.33–2.13)0.83 (0.27–2.55)
Higher secondary school0.53 (0.23–1.24)0.55 (0.22–1.39)0.47 (0.15–1.48)
College–university0.34 (0.13–0.86)0.33 (0.12–0.89)0.35 (0.10–1.24)
Postgraduate0.38 (0.12–1.23)0.34 (0.10–1.17)0.41 (0.09–1.94)
Factors of children's dietary beliefs and behaviors
Component 1: guilty about eating1.44 (1.17–1.77)1.40 (1.12–1.75)1.15 (0.86–1.55)
Component 2: concerned about body weight1.92 (1.56–2.36)1.90 (1.53–2.37)2.06 (1.55–2.73)
Component 3: eating all my food0.74 (0.59–0.94)0.73 (0.57–0.93)0.63 (0.46–0.87)
Father's obesity status
Normal weight—11
Overweight/obese—1.78 (1.09–2.91)1.55 (0.86–2.81)
Mother's obesity status
Normal weight—11
Overweight/obese—1.83 (1.12–2.98)2.17 (1.19–3.99)
Physical activity factors (from PCA)
Component 1: physical activity and sports after school——0.92 (0.71–1.30)
Component 2: video, electronic games, and computers——0.97 (0.73–1.28)
Component 3: watching TV, video, and DVD——0.90 (0.66–1.23)
Component 4: homework and private lessons——0.93 (0.69–1.26)
Component 5: home chores and cores outside home, aerobics, gymnastics, sports——0.90 (0.67–1.21)
Component 6: theater, cinema, use of mobile phone——0.88 (0.66–1.17)
Component 7: afternoon sleep, less private lessons——1.19 (0.89–1.59)
Component 8: sports for all, after-school activities (except sports)——0.87 (0.63–1.18)

KIDMED, Mediterranean Diet Quality Index for children and adolescents; PCA, principal component analysis.

∗Adjusted for age, gender, parental education level, and the three factors of dietary beliefs and behaviors.

†Adjusted for age, gender, parental education level, the three factors of dietary beliefs and behaviors, and parental obesity status.

‡Adjusted for age, gender, parental education level, the three factors of dietary beliefs and behaviors, parental obesity status, and physical activity levels (as assessed by PCA).

Back to Article Outline

Discussion 

We studied a nationwide population-based sample of Cypriot children with the aim of ascertaining whether there are associations between obesity levels and adherence to MD, as evaluated by the KIDMED score. After controlling for several confounders, it appeared that the initially observed significant inverse association between high adherence to the MD and obesity/overweight was mediated by the effect of physical activity.

Our results also showed that other factors were significantly associated with children's obesity status. Specifically, with respect to demographic factors, male gender, younger age, and maternal obesity status appeared to be positively associated in all three logistic models with children's obesity (Table 2). We previously reported this association of demographic factors to children's obesity status in a recent publication [22]. Furthermore, specific dietary beliefs and practices, such as those represented by principal components of types 2 (“concerned about body weight”) and 3 (“eating all my food”) seemed to be more important determinants of children's obesity status than the actual content or quality of diet (Table 2).

Our findings regarding the mediation of the initially inverse association observed between children's obesity status and KIDMED score by physical activity has also been reported by other researchers [23], [24], [25] As has been suggested, the strength and relative importance of physical activity to obesity status in children may be age dependent [23]. In addition, it might be that children who follow a healthy diet are those who might also maintain high levels of physical activity and this may at least partly explain our observation regarding the attenuation of the initially significant inverse association of KIDMED score to obesity status when physical activity was considered in the model. It should also be acknowledged that the small number of cases in the “optimal” KIDMED score (Table 1) may have resulted in the large confidence intervals listed in Table 2 and, consequently, in the insignificant results in model 3 (which included physical activity).

Our findings in relation to those features of diet that are associated with obesity are also confirmed by the relevant literature. Specifically, the conclusions of several reviews regarding the contribution of several dietary constituents to the development of childhood obesity indicate that there is conflicting evidence as to which dietary constituents are obesogenic [10], whereas findings regarding the contribution of dietary patterns and behaviors to children's obesity status are more consistent [10], [18]. This is in accordance with our present findings, which show that dietary behaviors may be more significant predictors of children's obesity status than any dietary constituents.

Another possible reason for the previous observation regarding the attenuation of the initially inverse association observed between children's obesity status and KIDMED score may be related to the construction of the KIDMED. It is possible that the specific constituents of the KIDMED might not be associated with the obesity status or the dichotomous, low-information response categories of the KIDMED questionnaire may further limit its ability to capture some existing associations.

Similar studies in adults have shown conflicting evidence, as reported in a recently published review regarding the association of obesity to the degree of adherence to the MD in adults [9]. These observations, however, may be due to the nature of the studies, because all the studies that did not show any association were cross-sectional or prospective, whereas all intervention studies revealed a significant inverse association.

Implications 

Health professionals' awareness of these findings is important for helping children and parents to prevent and manage obesity. Traditionally, the focus on prevention and management of childhood obesity, regarding its dietary aspect, has been on the actual content of diet [26], [27], [28], [29]. Recommendations and policies [26], [27] to combat childhood obesity have focused more on the composition of diet and less on dietary patterns and the psychological aspect of the diet. The most successful interventions, however, proved to be those that, besides the composition of the diet, also address the psychological–behavioral aspect of diet and physical activity [26], [28], [29], [30]. In a relatively recent review, it was stressed that, for the successful prevention and management of childhood obesity, the focus should be on establishing certain dietary behaviors [28] and maintaining certain levels of physical activity. Therefore, public health programs aiming to prevent and manage childhood obesity should, in addition to dietary composition, equally target dietary behaviors and beliefs and physical activity.

Strengths and limitations 

One of the strengths of this study is that this is a nationwide study and, to our best knowledge, the first internationally that has investigated the association of children's obesity status and the degree of adherence to the MD.

We are, however, aware of some limitations in our study. First, its cross-sectional methodology can only be suggestive of the observed relations.

A second limitation concerns the fact that the various Mediterranean cultures exhibit variations in their traditional diets and there is an inherent difficulty in describing one uniform MD prototype. We used the criteria adopted specifically for children by a group of Spanish experts who, we believe, provide a good description of the basic characteristics of all MDs, but despite the use of this children-specific index, certain other components that are not included in the index and the scoring method may have an impact on the association under investigation.

Third, the assessment of the children's and parents' obesity statuses was based on parental reported heights and weights, but this practice has been previously used with satisfactory results [31], [32], [33].

Back to Article Outline

Conclusions 

Adherence to the MD is inversely associated with obesity in children; however, other behaviors, in particular physical activity status, maternal obesity, and children's dietary beliefs and behaviors, seem to be more significant. Health professionals should assess children's dietary beliefs and behaviors, explore how these may influence children's diets, and subsequently include certain of their features/characteristics into any obesity intervention program.

Back to Article Outline

Acknowledgments 

The authors give warm thanks to the participant children and their parents, to the Cyprus Ministry of Education and Culture (Primary Education Department), and to all the teachers who readily consented to carry out the study during school hours.

Back to Article Outline

References 

  1. Trichopoulou A, Bamia C, Trichopoulos D. Mediterranean diet and survival among patients with coronary heart disease in Greece. Arch Intern Med. 2005;165:929–935
  2. Esposito K, Ciotola M, Giugliano D. Mediterranean diet, endothelial function and vascular inflammatory markers. Public Health Nutr. 2006;9(Suppl 8A):1073–1076
  3. Mantzoros CS, Williams CJ, Manson JE, Meigs JB, Hu FB. Adherence to the Mediterranean dietary pattern is positively associated with plasma adiponectin concentrations in diabetic women. Am J Clin Nutr. 2006;84:328
  4. Trichopoulou A, Costacou T, Bamia C, Trichopoulos D. Adherence to a Mediterranean diet and survival in a Greek population. N Engl J Med. 2003;348:2599–2608
  5. de Lorgeril M, Salen P. Modified Cretan Mediterranean diet in the prevention of coronary heart disease and cancer. World Rev Nutr Diet. 2000;87:1–23
  6. Panagiotakos DB, Pitsavos C, Arvaniti F, Stefanadis C. Adherence to the Mediterranean food pattern predicts the prevalence of hypertension, hypercholesterolemia, diabetes and obesity, among healthy adults; the accuracy of the MedDietScore. Prev Med. 2007;44:335–340
  7. In:  Matalas A-L,  Zampelas A,  Stavrinos V,  Wolinsky I editor. The Mediterranean diet. Constituents and health promotion. Boca Raton: CRC Press; 2001;
  8. International Association for the Study of Obesity. International Obesity Task Force prevalence data. Available at: http://www.iotf.org/database/index.aspAccessed June 30, 2008
  9. Buckland G, Bach A, Serra-Majem L. Obesity and the Mediterranean diet: a systematic review of observational and intervention studies. Obes Rev. 2008;9:582–593
  10. In:  Woodward-Lopez G,  Ritchie LD,  Gerstein DE,  Crawford PB editor. Obesity: dietary and developmental influences. Boca Raton: CRC Press; 2006;p. 31–255
  11. Serra-Majem L, Ribas L, Garcia A, Perez-Rodrigo C, Aranceta J. Nutrient adequacy and Mediterranean diet in Spanish school children and adolescents. Eur J Clin Nutr. 2003;57(Suppl 1):S35–S39
  12. Chatzi L, Apostolaki G, Bibakis I, Skypala I, Bibaki-Liakou V, Tzanakis N, et al. Protective effect of fruits, vegetables and the Mediterranean diet on asthma and allergies among children in Crete. Thorax. 2007;62:677–683
  13. Garcia-Marcos L, Canflanca IM, Garrido JB, Varela AL, Garcia-Hernandez G, Grima FG, et al. Relationship of asthma and rhinoconjunctivitis with obesity, exercise and Mediterranean diet in Spanish schoolchildren. Thorax. 2007;62:503–508
  14. Avellone G, Di Garbo V, Panno AV, Cordova R, Abruzzese G, Rotolo G, et al. Cardiovascular risk factors and dietary habits in secondary school children in southern Italy. Int Angiol. 1994;13:148–153
  15. Ferro-Luzzi A, Mobarhan S, Maiani G, Scaccini C, Virgili F, Knuiman JT. Vitamin E status in Italian children subsisting on a Mediterranean diet. Hum Nutr Clin Nutr. 1984;38:195–201
  16. Department of Statistics and Research, Ministry of Finance. Statistical codes of districts, municipalities, communities and quarters of Cyprus (in Greek). Series I. Report no 3. Cyprus: Printing Office of Republic of Cyprus; 2000.
  17. Serra-Majem L, Ribas L, Ngo J, Ortega RM, Garcia A, Perez-Rodrigo C, et al. Food, youth and the Mediterranean diet in Spain. Development of KIDMED, Mediterranean Diet Quality Index in children and adolescents. Public Health Nutr. 2004;7:931–935
  18. Lazarou C, Kalavana T, Matalas A-L. The influence of parents' dietary beliefs and behaviours, on children's dietary beliefs and behaviours: The CYKIDS study. Appetite. 2008;51:690–696
  19. te Velde SJ, De Bourdeaudhuij I, Thorsdottir I, Rasmussen M, Hagstromer M, Klepp KI, et al. Patterns in sedentary and exercise behaviors and associations with overweight in 9–14-year-old boys and girls—a cross-sectional study. BMC Public Health. 2007;7:16
  20. Lazarou C, Soteriades ESS. Physical activity patterns among preadolescent children in Cyprus: the CYKIDS study. J Phys Act Health 2009;6:185–94.
  21. Cole TJ, Bellizzi MC, Flegal KM, Dietz WH. Establishing a standard definition for child overweight and obesity worldwide: international survey. BMJ. 2000;320:1240–1243
  22. Lazarou C, Panagiotakos DB, Panayiotou G, Matalas A-L. Overweight and obesity in preadolescent children and their parents, in Cyprus: prevalence and associated socio-demographic factors—the CYKIDS study. Obes Rev. 2008;9:185–193
  23. Parsons TJ, Power C, Manor O. Physical activity, television viewing and body mass index: a cross-sectional analysis from childhood to adulthood in the 1958 British cohort. Int J Obesity (Lond). 2005;29:1212–1221
  24. Berkey CS, Rockett HR, Field AE, Gillman MW, Frazier AL, Camargo CA, et al. Activity, dietary intake, and weight changes in a longitudinal study of preadolescent and adolescent boys and girls. Pediatrics. 2000;105:E56
  25. Jago R, Baranowski T, Baranowski JC, Thompson D, Greaves KA. BMI from 3–6 y of age is predicted by TV viewing and physical activity, not diet. Int J Obesity (Lond). 2005;29:557–564
  26. Hearn LA, Miller MR, Campbell-Pope R. Review of evidence to guide primary health care policy and practice to prevent childhood obesity. Med J Aust. 2008;188(Suppl):S87–S91
  27. Davis MM, Gance-Cleveland B, Hassink S, Johnson R, Paradis G, Resnicow K. Recommendations for prevention of childhood obesity. Pediatrics. 2007;120(suppl 4):S229–S253
  28. Lean M, Lara J, Hill JO. ABC of obesity. Strategies for preventing obesity. BMJ. 2006;333:959–962
  29. Summerbell CD, Waters E, Edmunds LD, Kelly S, Brown T, Campbell KJ. Interventions for preventing obesity in children. Cochrane Database Syst Rev. 2005;3:CD001871
  30. Savoye M, Shaw M, Dziura J, Tamborlane WV, Rose P, Guandalini C, et al. Effects of a weight management program on body composition and metabolic parameters in overweight children: a randomized controlled trial. JAMA. 2007;297:2697–2704
  31. Andersen LF, Lillegaard IT, Overby N, Lytle L, Klepp KI, Johansson L. Overweight and obesity among Norwegian schoolchildren: changes from to 2000. Scand J Public Health. 1993;33:99–106
  32. Goodman E, Hinden BR, Khandelwal S. Accuracy of teen and parental reports of obesity and body mass index. Pediatrics. 2000;106:52–58
  33. Sekine M, Yamagami T, Hamanishi S, Kagamimori S. Accuracy of the estimated prevalence of childhood obesity from height and weight values reported by parents: results of the Toyama Birth Cohort study. J Epidemiol. 2002;12:9–13

 The study was partly supported by Charalambides Dairies and the Cyprus Dietetic Association.

PII: S0899-9007(09)00237-8

doi:10.1016/j.nut.2009.05.014

Nutrition
Volume 26, Issue 1 , Pages 61-67, January 2010

Article Tools