Overweight and obesity are associated with increased body fat, a consequence of positive energy balance over a prolonged period of time that occurs when energy intake exceeds energy expenditure( Reference Ogden, Yanovski and Carroll 1 ). Recent studies suggest that the prevalence of overweight and obesity is high, although stable, in children and adolescents( Reference Olds, Maher and Zumin 2 – Reference Castetbon 4 ). The prevalence of underweight, however, is high and continues to increase( Reference Martin, Rosenberg and Pratt 5 – Reference Wang, Monteiro and Popkin 7 ). Overweight and obesity in youth are associated with type 2 diabetes, metabolic syndrome, poor quality of life, lower physical fitness and self-esteem, and, in later life, with increased risk of CHD and cancer( Reference Reilly, Methven and McDowell 8 – Reference Klenk, Nagel and Ulmer 11 ). Underweight in children and adolescents is associated with a poor quality of life, lower physical fitness, amenorrhea, decreased bone mineral content, negative body image and fatigue, and, in later life, with increased mortality( Reference Stokic, Srdic and Barak 12 – Reference Flegal, Graubard and Williamson 16 ).
Monitoring underweight, overweight and obesity in children and adolescents is essential for accurate tracking and to evaluate the effectiveness of public health recommendations and intervention programmes. Studies of French children and adolescents found that overweight and obesity were stable between 1996 and 2006, although data were limited to Central and Western France( Reference Péneau, Salanave and Maillard-Teyssier 17 ). To our knowledge, obesity and overweight have not been updated in France since 2009. Furthermore, few data are available on the prevalence of underweight in French children and adolescents( Reference Lazzeri, Rossi and Kelly 18 ). Previous studies reported a prevalence of underweight of 7 and 15 % in French boys and girls, respectively, in 2006( Reference Lazzeri, Rossi and Kelly 18 ).
The purpose of the present study was to measure the prevalence of underweight, overweight and obesity in French children and adolescents from 2009 to 2013, to compare the results with findings prior to 2009 and to measure differences in weight categories according to socio-economic status (SES).
Methods
Study design
Data for the present study are from the French health promotion campaign ‘Move, a priority for your health’ (http://www.bougetasante.fr/), the BOUGE Program. The objectives of BOUGE were to: (i) assess physical fitness of children and adolescents; and (ii) promote the benefits of physical activity and physical fitness on the health of youths aged 9–16 years in French schools (http://www.bougetasante.fr/). The programme was developed by a French health-care organization (Fédération Nationale Mutualité Française; http://www.mutualite.fr/) and the National School Sport Union (Union Nationale du Sport Scolaire; http://unss.org/) and consisted of two days during the school year; one day to assess physical fitness (including anthropometric characteristics) and the second day to promote the health benefits of physical activity. Children and adolescents participating in the BOUGE Program were assessed only one time during the study. Each year a different group of students was measured. The study was approved by a research ethics committee (CPP Nord-Ouest IV, Lille, France). All procedures were performed in accordance with the Helsinki Declaration of 1975 as revised in 2008 and the European Good Clinical Practices. As the study did not involve an intervention and data were collected retrospectively by the study organizational structure (http://www.mutualite.fr/), the study was approved by the research ethics committee (CPP Nord-Ouest IV, Lille, France) as an epidemiological study. In this context, written informed consent was not required according to French human research regulations. Data collection was approved by the French National Commission of the Informatics Personal Data (Commission Nationale Informatique et Liberté).
A manual of operations was developed for teachers and participants in order to standardize test procedures (http://eps-bergpfad.fr/Sante_Bouge_Sommaire.html). Included in the manual were rationale of the study, test procedures and how data were collected. Teachers recorded test results into an electronic data system provided by the trial sponsor. An audit of the complete data set was performed and the aberrant data were excluded.
Data were collected in sixteen regions of France in 101 schools. All schools in France were invited to participate in the study. Each school director decided whether or not to participate. If the school agreed to participate, the students aged between 9 and 16 years were invited to participate. In total, 12 082 adolescents (6107 girls, 5975 boys) volunteered to participate. Of this number, 9670 were included in the present study; 2412 adolescents were excluded because of missing or incomplete data.
Measurements
Anthropometric measures
Weight and height, respectively, were measured in shorts and T-shirts without shoes to the nearest 0·1 kg using an electronic scale and the nearest 0·1 cm using a standard physician’s scale in a private room. BMI, a valid estimate of body fatness in children and adolescents( Reference Pietrobelli, Faith and Allison 19 ), was calculated from weight divided by the square of height (kg/m2). Underweight, overweight and obesity were assessed using international age- and sex-specific cut-off points( Reference Cole, Bellizzi and Flegal 20 , Reference Cole, Flegal and Nicholls 21 ).
Socio-economic status
The Human Development Index (HDI), developed by the United Nations Development Programme (http://hdr.undp.org/en), was used to assess SES of the family living in a city. The HDI considers the following factors: (i) a long and healthy life: life expectancy at birth; (ii) education index: mean years of schooling and expected years of schooling; and (iii) standard of living: gross income per capita. A city scores higher HDI when the life expectancy at birth is longer, the education period is longer and the income per capita is higher. The HDI score was collected using the zip code of the school (www.insee.fr).
Statistical analysis
Data are presented as percentage for categorical variables and as mean and standard deviation for continuous variables. Normality of the distribution was checked graphically and by using the Shapiro–Wilk test.
Comparisons of underweight, overweight and obesity between boys and girls were assessed by the χ2 test. Changes in underweight, overweight and obesity from 2009 to 2013 were assessed using the Cochran–Armitage trend test. ANOVA was used to compare SES among different BMI categories (underweight, overweight and obesity).
All statistical tests were performed at the two-tailed α level of 0·05. Data were analysed using the statistical software package IBM SPSS Statistics for Windows, Version 22.0 and Microsoft® Excel 2010.
Results
Mean height, weight, BMI and prevalence rates of underweight, overweight and obesity by age and sex are presented in Table 1. Obesity was significantly greater in boys compared with girls (P<0·05), although there were no significant differences between boys and girls in the prevalence of overweight (Table 1). The prevalence of underweight was higher in girls compared with boys (P<0·01).
* Childhood: 9–11 years in boys and 9–10 years in girls; early adolescence: 12–14 years in boys and 11–13 years in girls; late adolescence: 15–16 years in boys and 14–16 years in girls.
† The χ2 test was performed to assess differences in prevalence rates by sex; Student’s t test was performed to assess differences in anthropometric data by sex. Significant P values are indicated in bold font.
Prevalence rates from 2009 to 2013, for boys and girls combined and by sex, are presented in Table 2. Obesity increased significantly in boys and girls (P<0·05). There were no significant changes in overweight or underweight. While not significant, the prevalence of underweight girls increased from 12·0 % in 2009 to 16·7 % in 2013. The prevalence of underweight boys remained the same, 7·1 % in 2009 and 7·3 % in 2013.
* Cochran–Armitage trend test. Significant P values are indicated in bold font.
Prevalence rates of underweight, overweight and obesity according to SES are presented in Table 3. SES was significantly lower in underweight, overweight and obese boys and girls compared with youths of normal weight (P=0·003). During the course of the study, the prevalence of overweight and obesity in boys was higher in low and middle SES compared with high SES (P<0·001). In girls, the prevalence of underweight, overweight and obesity was higher in low and middle SES compared with high SES (P<0·001).
* Socio-economic status is based on the Human Development Index (HDI), collected using the zip code of the school. A higher HDI score is achieved when the life expectancy at birth is longer, the education period is longer and the income per capita is higher.
† ANOVA was used to compare the score of socio-economic status among different BMI categories. Significant P values are indicated in bold font.
‡ ANOVA was used to compare the prevalence rates of the socio-economic status categories among different BMI categories. Significant P values are indicated in bold font.
Discussion
Short- and long-term health outcomes associated with underweight, overweight and obesity in children and adolescents are important issues influencing public health policy( Reference Flegal, Graubard and Williamson 16 , Reference Lobstein, Baur and Uauy 22 ). Therefore, regular monitoring of BMI is recommended to track data and to assess the effectiveness of intervention programmes( Reference Nutbeam 23 ). To date, such tracking of children and adolescents in France has been inconsistent( Reference Péneau, Salanave and Maillard-Teyssier 17 , Reference Lazzeri, Rossi and Kelly 18 , Reference Klein Plata, Wagner and Haan 24 ).
By applying recommended international standards, the results of the present study suggest that combined overweight and obesity did not change significantly in boys and girls between 2009 and 2013, although the prevalence was higher compared with data published previously, 21·5 % in 2012 v. 15·4 % in 2004( Reference Péneau, Salanave and Maillard-Teyssier 17 ). Differences might be attributed to sampling error because previous data were collected in a specific part of France (i.e. only from two administrative regions( Reference Péneau, Salanave and Maillard-Teyssier 17 )) while the present study included most of the country (i.e. sixteen of twenty-two administrative regions). Separating obesity from overweight, our findings showed a significant increase in obese children and adolescents in both boys and girls.
Data from the present study also suggest that the prevalence of overweight and obesity is higher in children and adolescents with low SES. These findings are in agreement with previous studies in France( Reference Péneau, Salanave and Maillard-Teyssier 17 , Reference Klein Plata, Wagner and Haan 24 ), Europe( Reference Gnavi, Spagnoli and Galotto 25 – Reference Kromeyer-Hauschild, Zellner and Jaeger 27 ) and the USA( Reference Kimm, Obarzanek and Barton 28 ). Similarly, our data show that low SES is also associated with underweight children and adolescents. Interestingly, the phenomenon of overweight and obesity may coexist with underweight in the same population( Reference Doak, Adair and Bentley 29 ).
Low body fat and lean mass, typical of underweight children and adolescents, is associated with poor quality of life, lower physical fitness, amenorrhea, decreased bone mineral content, negative body image and fatigue( Reference Stokic, Srdic and Barak 12 – Reference Mak, Ho and Lo 15 ), and increased mortality as adults( Reference Flegal, Graubard and Williamson 16 ), when compared with adolescents and children of normal weight. Our findings show that the prevalence of underweight French girls increased by 39 % between 2009 and 2013, from 12·0 to 16·7 %, although the change was not significant. This result may be due to the analysis strategy. Using the Cochran–Armitage trend test, we examined linearity changes and not the differences between each year. Nevertheless, the high prevalence of underweight in young girls suggests a need for health intervention to address this problem. Increased prevalence of underweight from 2009 to 2013 might also be the result of increased numbers of children and adolescents who arrived from other countries and entered into the French school system, especially from Africa. For a variety of reasons the anthropometric measures of these children and adolescents may differ from those of European children. However, according to regulatory rules in clinical research, we cannot collect any information about the ethnic diversity.
One of the strengths of the present study was the large sample size representing most of France. The large sample size provided investigators an opportunity to assess underweight, overweight and obesity according to age, sex and SES. Standardized testing and data collection was another strength. A potential weakness is that subjects volunteered and were not selected at random. Even though we collected data on a large sample in many administrative regions (sixteen out of twenty-two regions), the study did not use a stratified sample design. Therefore, we cannot establish that this cohort is fully representative of French children and adolescents. The severe economic downturn in France between 2009 and 2013 could have had an impact on our findings as a consequence of changes in living habits at another time point. As a consequence, the authors recommend studies that incorporate random subject selection.
In summary, data from the present study suggest that obesity increased in French youth between 2009 and 2013, and occurred more frequently in boys and girls of low SES. The prevalence of overweight was high but did not change over the course of the study. Our results also indicate that underweight increased substantially in girls. Although the differences were not significant, they warrant careful monitoring. Based on the study findings, the authors support public health initiatives in French children and adolescents to prevent and treat unhealthy weight issues in this population.
Acknowledgements
Acknowledgements: The authors thank all participating youths and teachers for their collaboration in the study. Financial support: This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors. Conflict of interest: The authors declared no conflict of interest. Authorship: Each author of this article contributed significantly to the study. J.V., J.-B.B., J.M. and Z.U. designed the research; J.V., J.-B.B., P.S.F., L.B., J.M. and Z.U. conducted the research; J.V. and J.-B.B. analysed the data; J.-B.B. and J.V. performed the statistical analysis; J.V., J.-B.B., P.S.F. and L.B. wrote the paper; Z.U. had primary responsibility for the final content; all authors read and approved the final manuscript. Ethics of human subject participation: This study was conducted according to the guidelines laid down in the Declaration of Helsinki and the European Good Clinical Practices. The study was approved as an epidemiological study by the CPP Nord-Ouest IV (Lille, France) research ethics committee. In this context, written informed consent was not required according to French human research regulations. Data collection was approved by the French National Commission of the Informatics Personal Data (Commission Nationale Informatique et Liberté).