PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
 
Int J Public Health. Author manuscript; available in PMC 2010 September 1.
Published in final edited form as:
PMCID: PMC2735089
NIHMSID: NIHMS109022

Overweight in school-aged children and its relationship with demographic and lifestyle factors: Results from the WHO-Collaborative Health Behaviour in School-aged Children (HBSC) Study

Abstract

Objectives

To examine overweight prevalence and its association with demographic and lifestyle factors in 11–15 year olds in the HBSC 2005–2006 survey.

Methods

Self-reports of height, weight, eating patterns, physical activity and sedentary behaviours were obtained from nationally representative samples in 41 countries (n=205,939).

Results

Overweight prevalence was highest in USA (28.8%) and lowest in Latvia (7.6%). In most countries, overweight was more prevalent in boys than girls. Overweight was consistently negatively associated with breakfast consumption and moderate to vigorous physical activity; OR range: 0.48–0.79 and 0.50–0.78, respectively.

Conclusion

Overweight prevalence in youth remained high across the countries examined. The primary factors linked to overweight were breakfast consumption and physical activity. These data should contribute to formulating preventive programs and policies.

INTRODUCTION

Overweight negatively impacts the present and future psycho-social and physical aspects of health of youth. Data on overweight in young persons are, however, still lacking in many countries including those undergoing transition. To better understand the global prevalence and circumstances associated with overweight in youth, there is a need to provide comparable updated information on the burden of overweight in young persons across several nations by using nationally representative samples and standardized international definitions for overweight1.

Considering the complex aetiology of overweight and the lack of agreement on its determinants,2 it is important to concomitantly examine the relations of overweight with several demographic and potentially modifiable lifestyle factors such as eating habits, physical activity and sedentary behaviours of youth. There is growing literature on the association between eating patterns and overweight, particularly on breakfast habits. The findings generally support that skipping breakfast is associated with increased probability of being overweight3, 4. For the association of fruit and vegetable intake and soft drink consumption with overweight, however, the evidence is less consistent58. In contrast, most studies support that childhood physical activity is negatively related to overweight9 and that this association endures into adulthood10. Studies examining sedentary behaviour also show an independent and causal effect on weight status2, 8.

If consistent associations between overweight and potentially modifiable lifestyle factors are observed across countries, such information could provide substantive evidence to support preventive policies and programs to reduce overweight and associated health risks in young persons in the national- and international-context.

Thus, the objectives of the current paper were to describe the prevalence of overweight (pre-obesity and obesity) in 11-, 13-, and 15-year-olds from 41 countries participating in the 2005–2006 WHO Collaborative HBSC survey, and to examine the associations between overweight and certain lifestyle factors including dietary habits, physical activity and sedentary behaviours.

METHODS

Data for present analyses were collected in 41 countries participating in the 2005–2006 WHO collaborative HBSC study, an international collaboration between research teams across Europe and North America with the aim of gaining insights into adolescents’ health and health behaviours. The standardized international research protocol was followed within each country to ensure consistency in survey instruments, data collection and processing procedures11. Participation was voluntary, and anonymity and confidentiality were ensured. Questionnaires were administered in school classrooms by trained personnel, teachers, or school nurses. The time frame for filling out the questionnaires was one school period. Each country respected ethical and legal requirements in their countries for this type of survey.

The population selected for sampling was 11, 13 and 15 year olds attending school with the desired mean age for the three age groups being 11.5, 13.5 and 15.5 years. Participating countries were required to include a minimum of 95 percent of the eligible target population within their sample frame. In the majority of countries, national representative samples were drawn and samples were stratified to ensure representation by, for example, geography, ethnic group and school type. Participants were selected using cluster sampling, with school or class as the sampling unit. The recommended sample size for each of the three age groups was approximately 1,500 students, assuming a 95% confidence interval of +/− 3 percent around a proportion of 50 per cent and allowing for the clustered nature of the samples. More detailed information about the study is provided elsewhere.1215

Of 205,939 school-aged children participating in the survey those not reporting their age, weight or height were excluded, leaving 171,809 pupils (83%) in the analyses.

Body Mass Index (BMI) (kg/m2) was calculated using self-reported weight and height. Overweight included both pre-obesity and obesity, that were based on age- and gender-specific cutoffs corresponding to adult reference levels of 25–30 and ≥30 kg/m2, respectively as recommended by the International Obesity Task Force1.

Family affluence was determined with the Family Affluence Scale as a score of four items: Does your family own a car, van or truck? (0–2 points). Do you have your own bedroom for yourself? (0–1 points). During the past twelve months, how many times did you travel away on holiday (vacation) with your family? (0–2 points); and how many computers does your family own? (0–2 points).

Lifestyle variables

Usual eating habits were assessed by asking participants how many times a week they eat fruit, vegetables and soft drinks. The possible responses were: “never”, “less than once a week”, “about once a week”, “two to four days a week”, “five to six days a week”, “once a day, every day”, “every day, more than once”. For each of these variables, responses were dichotomised: less than daily and daily16.

To assess regular breakfast consumption, students were asked to estimate how many weekdays they had breakfast (i.e. having more than a glass of milk or fruit juice). Possible response categories were “never”, and 1, 2, 3, 4, or 5 days. Responses were recoded: “less than daily” versus “daily”.

Moderate to vigorous physical activity (MVPA) was assessed by asking: “On how many days in the past week were you physically active for 60 minutes or more”. Physical activity was defined as “any activity that increases your heart rate and makes you get out of breath some of the time” with examples of such activities. Response categories were: “0 days”, “1”, “2”, etc up to “7 days”, recoded as < or ≥ 5 times/week17. Vigorous physical activity (VPA) was assessed by asking: “Outside school hours, how many hours a week do you usually exercise in your free time so much that you get out of breath or sweat?” Response categories were none, about 30 minutes, and 1, 2–3, 4–6, ≥7 hours; recoded into < or ≥ 2 hour/week18.

Three items assessed sedentary screen-based activities: 1) “About how many hours a day do you usually watch television (including DVDs and videos) in your free time?“ 2) “About how many hours a day do you usually play games on a computer or games console (Playstation, Xbox, GameCube etc.) in your free time?” and 3) “About how many hours a day do you usually use a computer for chatting on-line, internet, emailing, homework etc. in your free time? All three items had nine possible responses: “none at all”, about 30 min, 1 hour, 2 hour, up to ≥7 hour/day. Responses for weekdays were recoded into ≤ versus > 2 hour/day19.

Statistical Analyses

Statistical analyses were performed using SPSS version 15 and STATA 9.2. Association of overweight with gender and age were examined with chi-square or spearman rho correlation analysis, as indicated. To examine the possibility of a selection bias for countries where response rate on BMI-related variables (height, weight, and age) was 80%, the differences in eating patterns, physical activity and sedentary behaviours of students with versus those without data on BMI were assessed using chi-square test. Multilevel logistic regression analyses were conducted using the svy, vec (linearized) command in STATA, with school as the level-2 sampling unit variable and age group as strata (a two-level random intercept model). All analyses were conducted separately for each country and gender. Initially, interaction with age was tested in the main effect models for each lifestyle variable; significant interactions were generally not observed (< 2%). The multilevel regressions on the association of overweight with each independent lifestyle factor (dummy variable) were controlled for age and family affluence because of their potential association with overweight. P-values <0.05 were considered significant. Results are presented by geographic region as defined by the United Nations20 to examine regional trends.

RESULTS

The response rate of study participants on height, weight or age, necessary to estimate BMI, varied between 32% in Ireland and 99% in Czech Republic (Table 1). It increased with age consistently across most countries (data not shown). Ten countries, most in Northern-Europe, had response rates on BMI-related variables of ≤ 80%. For these countries we compared youth providing data concerning BMI to those without BMI values, and found that those without BMI data generally had less healthy lifestyle (eating patterns, and physical and sedentary activities) (Tables 2 and and3).3). Thus, in all tables, countries with ≤ 80% response rate on BMI-related variables are presented, but not included in cumulative cross-national comparative findings. The results described below are, therefore, generally based on 31 countries.

Table 1
Number of students surveyed (N), availability of BMI data (%), and prevalence (%) of overweight by region, country, age, and gender
Table 2
Eating patterns (%) by region, country and gender
Table 3
Physical activity and screen based sedentary behaviors (%) by region, country and gender

The highest mean prevalence of overweight was seen for the non-European countries (24.2%) followed by countries in Southern-European region (15.8%). Little difference was seen between Central (10.5%), Eastern (11.3%) and Northern-European (11.7%) regions. Across countries, the highest prevalence of overweight (pre-obesity and obesity) was observed in USA (28.8%) followed by Italy (18.3%), and that of obesity was seen in USA (8.9%) followed by Canada (5.2%). The lowest prevalence of overweight was observed in Latvia (7.6%) followed by Ukraine (8.0%), and that of obesity in Slovakia and Ukraine (0.8%) (Table 1).

In virtually all (29 of 31) countries, the prevalence of overweight was significantly higher among boys than among girls. Among boys, prevalence of overweight increased with age in eight countries and decreased with age in five countries. The positive correlation between overweight and age for boys was predominately observed in the Central- and Northern-European regions while a negative association was noted in Eastern-European countries. Among girls, negative correlations between overweight and age were seen in 18 countries, predominantly countries from the Southern (6 countries) and Eastern-European (7 countries) regions.

Table 2 provides descriptive data on patterns of healthy eating among participants. Less than 50% of young people report eating fruit or vegetables daily. Specifically, the proportion of young people who eat fruit daily during the week ranged from 19 to 47%, with girls reporting significantly greater fruit consumption than boys in nearly all (29 of 31) countries and regions, with considerable geographic variation. For vegetable consumption, similar gender effect was noted in virtually all countries and regions (30 of 31); girls consuming more vegetables daily than boys, with considerable geographic variation (range for both genders: 14–65%). However for breakfast consumption (range: 39–84%), this gender effect was inversed with more boys consuming breakfast daily than girls in nearly all (30 of 31) countries.

Not having soft drinks daily was reported by a majority of young people across all countries and regions. In most countries girls were less likely to consume soft drinks than boys. Soft drink consumption was highest in Bulgaria (~50%) and lowest in Sweden, Iceland, Finland, and Estonia – countries in Northern-Europe – where only 10% or less had soft drinks daily.

Table 3 shows the physical activity and sedentary behaviours of young people. In most countries across all regions, about a third of young people met the guideline of 60 minutes of MVPA on five or more days a week, with the exception of Russia and Portugal. Boys met this recommendation more often than girls in all countries examined; the range being 35 to 67% for boys and 18 to 51% for girls, with considerable regional variation. One-third of young people also reported engaging in VPA at least 2 hours/week; with boys reporting this behaviour more often than girls.

With respect to television-viewing, electronic games and computer use (each ≤ 2 hour/day) (Table 3), girls were more likely to report engaging in these sedentary behaviours than boys in almost all countries, with the exception of Bulgaria, Greece and Romania for television; and Canada and USA for computer use. The reported rate of television-viewing varied across countries a lot more than computer use (ranges were 34–83, 76–97, and 65–96% for television-viewing, electronic games and computer use, respectively).

Tables 4 and and55 present the age- and SES-adjusted odds ratios for being overweight in relation to lifestyle factors. With regards to eating patterns (Table 4), only daily breakfast consumption was consistently negatively associated with overweight (significant OR ranged between 0.48 and 0.79); this association was stronger for boys than girls (noted in 26 and 18 of 31 countries, respectively) across all regions. Daily fruit, vegetable or soft drink consumption were generally not associated with overweight.

Table 4
Age- and SES-adjusted odds ratios and 95% confidence intervals for being overweight by eating patterns, stratified by region, country, and gender
Table 5
Age- and SES-adjusted odds ratios and 95% confidence intervals for being overweight by physical activity and screen based sedentary behaviours, stratified by region, country, and gender

Among the physical activity and sedentary behaviours examined, the most important and consistent associations were observed for physical activity. Engaging in MVPA for 1 hour on at least 5 days a week showed a consistent negative correlation with being overweight (OR range: 0.50–0.78) in 26 of 31 countries for boys and 14 of 31 countries for girls. No regional trends in MVPA were seen for boys, but for girls this association was noted in Central and non-European regions. VPA was also negatively associated with being overweight (OR range: 0.50–0.79). However this association was not as consistent as that for MVPA across countries. Only 14 and 7 of 30 countries showed this negative association between VPA and overweight for boys and girls, respectively.

Among sedentary activities examined, television-viewing ≤ 2 hour/day was associated with reduced likelihood of being overweight (OR range: 0.51–0.78) in 10 of 30 countries for boys and 13 of 30 countries for girls. Playing games on any electronic media ≤ 2 hour/day was also associated with reduced likelihood of being overweight; but this was found in only 8 of 30 countries (OR range: 0.39–0.71). Computer use for other activities was generally not associated with overweight. For most physical activity and sedentary behaviours, regional patterns were observed with stronger associations in Central-European countries. In addition, for most countries stronger associations were noted for boys than for girls concerning overweight and physical activity and sedentary behaviours with the exception of television-viewing.

DISCUSSION

The results from this large scale international survey among school-aged youth, utilising standardized methods for data collection, and international cut-offs to define overweight1 showed that overweight prevalence was >10% in most nations (range: 7.6% in Latvia to 28.8% in USA). Within Europe, the highest prevalence was seen in Southern-European countries. Similar geographical patterns have been observed elsewhere5, 21. Thus the prevalence of overweight continues to be an important public health challenge in most countries participating in the HBSC study. In particular, two thirds of the countries that participated in both the 2001–2002 and 2005–2006 study showed a tendency of increasing overweight.

A clear pattern of boys (16.2%) being more likely to be overweight than girls (10%) was consistently noted across countries. Such obvious cross-national gender differences have not been observed elsewhere22. The emerging gender patterns could indicate that the obesogenic environmental influences may have become more detrimental and/or that preventive initiatives may be inadequate and/or less effective for boys.

The pattern of overweight prevalence according to age was less consistent and varied regionally. For example, in most Eastern-European countries the prevalence of overweight decreased with age in both genders, whereas in Southern-European countries this pattern was observed only in girls. In most Central-European countries a positive correlation between overweight and age was found among boys. Most of the previous literature suggests a positive correlation between age and prevalence of overweight, although varying patterns are often observed23.

It is important to consider that the results are based on self-reported data that could be subject to socially desirable reporting bias. However, students responses were anonymous; therefore, participants had no reason to dissemble or misreport their height or weight. BMI based on self-reported data can produce lower prevalence estimates of overweight (pre-obesity and obesity) than those based on actual height and weight measurements24 while others have reported high accuracy for classification of youth as obese or non-obese based on self-reported data25. Furthermore, BMI based on self-reports has been found to be fairly reliable25 and suitable for identifying valid relationships in epidemiological studies25, 26. Associations between weight status and lifestyle factors (e.g. physical activity, television viewing, breakfast habits) did not differ when based on self-reported versus measured height and weight data25. In the current study 17% of the sample had missing values on BMI; a high proportion of missing data on height and weight has been reported in this age group5. In order to maintain validity the ten countries with the highest risk of low generalizability due to a large proportion of missing values, were omitted from current analyses. The current findings are based therefore on analyses restricted to 31 countries where data on variables of interest were available on >90% of the original representative sample in the country.

With respect to eating patterns, the presented results show that compared to boys, girls’ daily consumption of fruits and vegetables was higher, and of soft drinks and breakfast was lower. A more healthy eating pattern in girls has been previously reported16. It is interesting to note, however, that breakfast consumption, which is usually considered a positive practice for several health outcomes3, was lower in girls than boys, as reported previously4. In this study, no consistent relations between eating patterns and overweight were noted except for an inverse association with breakfast consumption. Importantly our finding of a negative association between regular breakfast consumption with overweight fits well with the literature3, 4.

The presented results showing lack of an association between fruit and vegetables intake and overweight are consistent with findings from the HBSC 2001–2002 survey5 and in contrast to other reports6, 7. Similarly the finding of no clear association between soft drink consumption and overweight, although reported previously5, 7, is generally inconsistent with the negative, albeit weak association reported elsewhere27. The current results were based solely on the frequency of consumption of food items without any details on quantity consumed; it is likely that the absence of information concerning portion size, particularly for items where the portions could vary considerably (fruit, vegetables, and beverages), could have masked some associations. However, the finding of a strong and significant association between breakfast consumption and lower probability of overweight despite the limitations associated with food frequency questions suggests the strength of this association across nations.

Consistent with the existing literature on inverse association between childhood physical activity and obesity9, 28, the present study shows a clear pattern across gender, countries, and regions that MVPA, and to a lesser extent VPA, being negatively related to adolescent overweight. Data were obtained by self-reports which, compared to objective measures, tends to underreport MVPA and over-report VPA10; however, this is less likely to affect measures of associations. For boys, MVPA for 1 hour on at least 5 days a week was strongly and negatively associated with overweight. The relation of MVPA on weight status may be different across genders and cultures, and influenced by several factors. In all countries, girls reported less physical activity than boys and fewer girls may have reached the threshold necessary to demonstrate the relationships between MVPA and weight status. The weaker association of VPA with overweight suggests that the guidelines for VPA may need to be increased16. Because MVPA is more accessible to all and showed a consistent negative association with overweight, the findings suggest that it should be integrated into public health messages and programs for young persons.

Studies examining both sedentary behaviour and physical activity report weak or no relations between them and yet there is evidence that each has an independent effect on weight status2. In the current study, overweight status had a negative relationship with television-viewing and electronic game playing in many countries; this supports former findings8 and current guidelines but demonstrates variations across countries.

In conclusion, the results of this large multinational survey indicate that overweight in youth continues to be a public health concern. Furthermore, the strong and consistent negative association of overweight with certain lifestyle factors including breakfast consumption and MVPA suggest the importance of formulating and strengthening preventive public health policies concerning these practices.

Acknowledgments

The writing of this paper has been facilitated by the EU-funded HOPE project: “Health-promotion through Obesity Prevention across Europe” (the Commission of the European Communities, SP5A-CT-2006-044128). The study does not necessarily reflect the Commission’s views and in no way anticipates the Commission’s future policy in this area. We are grateful to Bjorn Holstein and Veronika Ottova for their support for realization of this work.

References

1. Cole TJ, Bellizzi MC, Flegal KM, Dietz WH. Establishing a standard definition for child overweight and obesity worldwide: international survey. BMJ. 2000;320:1240. [PMC free article] [PubMed]
2. Ekelund U, Brage S, Froberg K, et al. TV Viewing and Physical Activity Are Independently Associated with Metabolic Risk in Children: The European Youth Heart Study. PLoS Medicine. 2006;3:e488. [PMC free article] [PubMed]
3. Rampersaud CG, Pereira AM, Girard LB, Adams J, Metzl DJ. Breakfast Habits, Nutritional Status, Body Weight, and Academic Performance in Children and Adolescents. Journal of the American Dietetic Association. 2005;105:743–60. [PubMed]
4. Keski-Rahkonen A, Kaprio J, Rissanen A, Virkkunen M, Rose RJ. Breakfast skipping and health-compromising behaviors in adolescents and adults. Eur J Clin Nutr. 2003;57:842–53. [PubMed]
5. Janssen I, Katzmarzyk PT, Boyce WF, et al. Comparison of overweight and obesity prevalence in school-aged youth from 34 countries and their relationships with physical activity and dietary patterns. Obesity Reviews. 2005;6:123–32. [PubMed]
6. Delva J, Johnston DL, O’Malley MP. The Epidemiology of Overweight and Related Lifestyle Behaviors: Racial/Ethnic and Socioeconomic Status Differences Among American Youth. American Journal of Preventive Medicine. 2007;33:S178–S86. [PubMed]
7. Roseman GM, Yeung WK, Nickelsen J. Examination of Weight Status and Dietary Behaviors of Middle School Students in Kentucky. Journal of the American Dietetic Association. 2007;107:1139–45. [PubMed]
8. DeMattia L, Lemont L, Meurer L. Do interventions to limit sedentary behaviours change behaviour and reduce childhood obesity? A critical review of the literature. Obes Rev. 2007;8:69–81. [PubMed]
9. Strong WB, Malina RM, Blimkie CJ, et al. Evidence based physical activity for school-age youth. J Pediatr. 2005;146:732–7. [PubMed]
10. Riddoch CJ, Bo Andersen L, Wedderkopp N, et al. Physical activity levels and patterns of 9- and 15-year-old European children. Med Sci Sports Exerc. 2004;36:86–92. [PubMed]
11. Currie C, Samdal O, Boyce W. Edinburgh, Child and Adolescent Health Research Unit. University of Edinburgh; 2001. Health Behaviour in School-aged Children: a World Health Organization cross-national study (HBSC). Research protocol for the 2001/2002 survey.
12. Roberts C, Freeman J, Samdal O, et al. The Health Behaviour in School-aged Children (HBSC) study: methodological developments and current tensions. International Journal of Public Health. 2009;xx:xx–xx. [PMC free article] [PubMed]
13. Roberts C, Currie C, Samdal O, Currie D, Smith R, Maes L. Measuring the health and health behaviours of adolescents through cross-national survey research: recent developments in the Health Behaviour in School-aged Children (HBSC) study. J Public Health. 2007;15:179–86.
14. Currie C, Gabhainn SN, Godeau E, et al. Inequalities in young people’s health: HBSC international report from the 2005/2006 Survey. Copenhagen: WHO Regional Office for Europe; 2008.
15. Currie C. Overview paper on development of HBSC. International Journal of Public Health. 2009;xx:xx–xx.
16. Vereecken CA, Inchley J, Subramanian SV, Hublet A, Maes L. The relative influence of individual and contextual socio-economic status on consumption of fruit and soft drinks among adolescents in Europe. Eur J Public Health. 2005;15:224–32. [PubMed]
17. Prochaska JJ, Sallis JF, Long B. A physical activity screening measure for use with adolescents in primary care. Arch Pediatr Adolesc Med. 2001;155:554–9. [PubMed]
18. Booth ML, Okely AD, Chey T, Bauman A. The reliability and validity of the physical activity questions in the WHO health behaviour in schoolchildren (HBSC) survey: a population study. Br J Sports Med. 2001;35:263. [PMC free article] [PubMed]
19. American Academy of Pediatrics, Committee on Public Education. Children, Adolescents, and Television. Pediatrics. 2001;107:423–6. [PubMed]
20. Composition of macro geographical (continental) regions, geographical subregions, and selected economic and other groupings. United Nation Statistic Division, 31 jan 2008. (Accessed 27/07/2008, at http://unstats.un.org/unsd/methods/m49/m49regin.htm.)
21. Lobstein T, Frelut ML. Prevalence of overweight among children in Europe. Obesity Reviews. 2003;4:195–200. [PubMed]
22. Sweeting H. Gendered dimensions of obesity in childhood and adolescence. Nutrition Journal. 2008;7:1. [PMC free article] [PubMed]
23. Livingstone B. Epidemiology of childhood obesity in Europe. European Journal of Pediatrics. 2000;159:S14–S34. [PubMed]
24. Himes HJ, Hannan P, Wall M, Neumark-Sztainer D. Factors Associated with Errors in Self-reports of Stature, Weight, and Body Mass Index in Minnesota Adolescents. Annals of epidemiology. 2005;15:272–8. [PubMed]
25. Strauss RS. Comparison of measured and self-reported weight and height in a cross-sectional sample of young adolescents. Int J Obes Relat Metab Disord. 1999;23:904–8. [PubMed]
26. Goodman E, Hinden BR, Khandelwal S. Accuracy of Teen and Parental Reports of Obesity and Body Mass Index. Pediatrics. 2000;106:52–8. [PubMed]
27. Malik VS, Schulze MB, Hu FB. Intake of sugar-sweetened beverages and weight gain: a systematic review. Am J Clin Nutr. 2006;84:274–88. [PMC free article] [PubMed]
28. Pietilainen KH, Kaprio J, Borg P, et al. Physical inactivity and obesity: a vicious circle. Obesity (Silver Spring) 2008;16:409–14. [PMC free article] [PubMed]