Trends in overweight and obesity and changes in the ... - Nature

European Journal of Clinical Nutrition (2007) 61, 404–411

& 2007 Nature Publishing Group All rights reserved 0954-3007/07 $30.00 www.nature.com/ejcn

ORIGINAL ARTICLE

Trends in overweight and obesity and changes in the distribution of body mass index in schoolchildren of Jena, East Germany K Kromeyer-Hauschild and K Zellner Institute of Human Genetics and Anthropology, Friedrich-Schiller-University Jena, Jena, Germany

Objective: Data of Jena children (Germany) show increases in the prevalence estimates of overweight and obesity between 1975 and 2001. Our objective was to determine if these increases contribute to changes in the distribution of body mass index (BMI) in the entire population of children. The decade 1985–1995, which includes the time of German reunification, is of particular interest because of the significant socio-economic changes in East Germany, in which Jena is located. Design and subjects: The analyses were based on data from four cross-sectional growth studies in 7- to 14-year-old children performed in Jena, Germany, in 1975 (n ¼ 2013), 1985 (n ¼ 1534), 1995 (n ¼ 1906) and 2001 (n ¼ 1918). Overweight and obesity were estimated by German reference data and the reference of Cole et al. Mean difference plots were used to investigate the changes in the BMI distributions within sexes between the studies. Results: Between 1985 and 1995, the prevalence of overweight and obesity increased significantly in both sexes, whereas nonsignificant changes were found between 1975 and 1985. Although there is a tendency towards increased overweight between 1995 and 2001, the differences were nonsignificant. Comparing data from 1975 and 1985, the mean difference plots showed virtually no changes in the BMI distribution. In the plots comparing the BMI distribution for 1985 data and 1995 data, the whole distribution tended to shift upwards with a greater shift in the upper BMI range. The entire population of children in Jena gained weight during this period of time. This increase may be linked to a transition towards a more Western lifestyle, that is, usually by lower energy expenditure and nutritional transitions, caused by the process of reunification of Germany in 1989. Conclusions: Weight gain appears to be a characteristic feature of the entire population studied and does not seem to be a separate problem of the obese children. This suggests that reported changes in the prevalence may be underestimating the looming public health crisis.

European Journal of Clinical Nutrition (2007) 61, 404–411. doi:10.1038/sj.ejcn.1602533; published online 20 September 2006 Keywords: childhood obesity; prevalence; BMI distribution

Introduction The prevalence of overweight and obesity is increasing in adults as well as in children worldwide (Livingstone, 2000; Bundred et al., 2001; Magarey et al., 2001; Kain et al., 2002; Dietz, 2004). Epidemiological data show that obesity and

Correspondence: Dr K Kromeyer-Hauschild, Institute of Human Genetics and Anthropology, Friedrich-Schiller-University Jena, Kollegiengasse 10, 07740 Jena, Germany. E-mail: [email protected] Guarantor: K Kromeyer-Hauschild. Contributors: KK-H contributed to the study design, data collection, data analysis and writing of the paper, and KZ contributed to the study design, data collection and assisted in writing the paper. Received 14 July 2006; revised 25 July 2006; accepted 30 July 2006; published online 20 September 2006

overweight are multi-determined chronic health problems resulting from complex interactions between genes and an environment that promotes high-energy intake and lowenergy expenditure, resulting in a positive energy balance (Gunnell et al., 1998; Weinsier et al., 1998; Goran, 2001). In this environment, some individuals are more vulnerable than others to weight gain and developing obesity. Comparison of the entire body mass index (BMI) distribution showed that for both adults and children, the distribution is becoming more skewed in all sex–age groups. The heaviest subgroup of the population was markedly heavier in the third National Health and Nutrition Examination Survey (NHANES III: 1988–1994) than in previous surveys. In younger children and to some extent in adolescents, the principal effect is increased positive skewness at the upper part of the distribution. In addition, for adults, the entire

Overweight and obesity in children from Jena K Kromeyer-Hauschild and K Zellner

405 distribution of BMI is shifted upwards (Flegal and Troiano, 2000). These observations suggest that there is a distinct subgroup that is genetically susceptible to obesity and that for adults some factors causing increases in BMI are affecting the entire population. In most epidemiological studies concerning obesity, prevalence rates were reported. To investigate the interaction between the individuals and the environment, BMI distribution changes at the population level should be studied, in order to identify subgroups at high risk for obesity. To prevent obesity in a child population, strategies that encourage healthful diets and plentiful physical activity will benefit the health of all children, whether at risk of obesity or not. A recent study on Jena school children indicated a rapid increase in the prevalence of overweight and obesity between 1985 and 1995, whereas between 1975 and 1985 only small changes in the prevalence estimates occurred (Zellner et al., 2004). This rapid increase in Jena children may be linked to enormous environmental transitions during the 1990s in East Germany. As a result of the reunification of Germany in 1989, Western lifestyle, characterized by a ready access to an abundance of food and low levels of physical activity, was introduced in East Germany, in which Jena is located. In this study, it was investigated whether this trend was accompanied with changes in the distribution of BMI in the population. In addition, the prevalence rates of overweight and obesity in Jena children calculated with two different sets of reference BMI values – German reference data recommended by the ‘Arbeitsgemeinschaft Adipositas im Kindes- und Jugendalter’ (AGA) (Kromeyer-Hauschild et al., 2001) and international reference values proposed by Cole et al. (2000) – were compared.

Methods The analyses were based on data from four successive growth studies – carried out in 1975, 1985, 1995 and 2001 — on samples of children from Jena. The children were randomly chosen, and the representative sample includes approximately 15–20% of all children aged 7–14 years living in Jena during the periods of the studies. Study approval was obtained from the University ethics committee and from school authorities. Parents gave their informed written consent. All subjects were of German origin and the distribution of socio-demographic characteristics in the samples was representative of the local population indicating no bias selection in participation. Measurements in all studies were performed by a welltrained staff according to standardized guidelines (Martin and Saller, 1957). Height and weight were measured using a Martin anthropometer and a beam balance scale to the nearest 0.1 cm and 0.5 kg, respectively (see Kromeyer-Hauschild et al., 1999 for further methodological details of the studies). The number of subjects with complete height and weight data by age in the cohorts is given in Table 1.

Table 1

Age distribution in the study population

Age (years)a

7 8 9 10 11 12 13 14 Total a

1975

1985

1995

2001

Boys

Girls

Boys

Girls

Boys

Girls

Boys

Girls

71 125 129 157 128 140 122 133 1005

80 124 128 151 137 126 124 124 994

74 125 111 82 83 87 105 113 780

83 116 98 97 77 80 96 106 753

79 118 109 96 158 164 141 124 989

71 120 120 88 149 124 129 110 911

67 100 97 117 138 161 130 154 964

81 100 106 117 130 149 138 130 951

Refers to midpoint of the year interval.

Statistics The BMI was calculated as weight (kg) divided by squared values of height (m2). For the comparison between the four cohorts concerning the prevalence of overweight and obesity, the data were examined according to national reference data – BMI percentiles from German children – as recently recommended by the AGA (Kromeyer-Hauschild et al., 2001) – and international reference data from Cole et al. (2000). These BMI percentiles were derived using the LMS method of Cole and Green (1992). According to the national reference data, the prevalence of overweight and obesity was estimated as the percentage of children whose BMI was above the 90th and 97th age- and sex-specific percentile of this reference, respectively. All individual BMI data were also converted to SD scores using this national reference data, allowing the comparison — adjusted for age — between the four cohorts. The SD scores (SDS) were constructed for the cohorts and sexes separately using the LMS method of Cole and Green (1992). The LMS method allows for the calculation of SDS values (Z-scores) (SDS ¼ [(BMI/M(t))L(t)1]/L(t)S(t), where BMI is the individual BMI value, L(t), M(t) and S(t) are the reference values of L, M and S for the individual age and sex) even in the case that characteristics do not fit a normal distribution curve, such as the BMI. In the study, individual BMI values were converted to SDS values and the SDS values were converted to centiles using normal distribution tables. Analysis of variance followed by Tamhane-T2 post hoc test – which does not require variances homogeneity – was performed for BMI-SDS comparisons between the studies and the sexes. The normality of SDS distribution was evaluated before by the Kolmogorov–Smirnov test. One-sample t-test was used to compare the cohorts to the national reference. Trends in overweight and obesity prevalence (national reference) among studies and sexes were tested by using w2-test for categorical variables. Mean difference plots were used to investigate the changes in the BMI distributions within sex–age groups between the studies. Mean difference plots allowed for qualitative visual comparison of distributions to be made (Flegal and Troiano, European Journal of Clinical Nutrition


406 (P ¼ 0.003) and girls (Po0.001) and obesity in boys (P ¼ 0.007) increased significantly between 1985 and 1995. Although there is a tendency towards increased overweight between 1995 and 2001, the differences were nonsignificant. It should be mentioned that the time periods compared are not the same. The investigations were carried out every 10 years until 1995, whereas there were only 6 years between 1995 and 2001 and this may cause the different changes. Comparing the two references, we found that Cole et al.’s method produced slightly lower estimates for obesity than did the German reference but higher estimates for overweight. When comparing the BMI-SDS values (Figure 1), no differences were found between the sexes (P ¼ 0.993) or between the years 1975 and 1985 (P ¼ 0.870). In 1975 and 1985, the values were significantly lower (all Po0.001) than in 1995 and 2001. A statistically significant difference (P ¼ 0.009) was also found in boys between 1995 and 2001. In 1975, 1985 and 1995, the values found in children from Jena were consistently lower than the national reference (the German reference has a mean of 0 and a standard deviation of 1.0). Although in these 3 years the differences were significant (all Po0.001) to the national reference, no significant difference (P ¼ 0.076) was seen in 2001. Figure 2 shows mean difference plots by single year of age for boys and girls aged 7–10 years. The plots display differences between the distributions, whereas the general shape of the line formed by the points is more important than the individual details. Although there are differences in detail, the patterns for children aged 7–10 years are broadly similar. When 1975 and 1985 were compared, the distribution showed virtually no directed changes. In the plots comparing 1985 and 1995, the whole distribution tended to shift upwards in both sexes, except the lower end of the distribution in 7-year-old boys. The changes were most

2000). With this method, shifts in distributions are investigated by comparing the corresponding percentiles from two distributions. Therefore, we calculated even percentile values for 1975, 1985, 1995 and 2001 (1st, 2nd, 3rd percentile up to the 99th percentile) using Cole’s LMS method. Then, we calculated the individual percentile levels for each subject. If there are subjects with BMI values on the corresponding percentile levels in the compared years, the mean of these percentiles from the 2 years (e.g. 1975 and 1985) as well as the difference between these percentile values (e.g. 1985– 1975) was calculated. Each point on the plots represents the mean (graphed on the x axis) and the difference (graphed on the y axis) for the corresponding percentiles from the compared years. The mean difference plots were presented by single year of age. The number of marks in the plots depends on the number of corresponding percentiles in the compared years. The location of the plotted point on the y axis shows the direction and magnitude of any shift between distributions. The shift is assessed by judging differences from the horizontal line representing zero, whereas general patterns are more important than the individual details. Statistical inferences were drawn at a significance level of 5%. Analyses were performed using SPSS (SPSS Inc., Chicago, IL, USA) software.

Results The prevalence estimates of overweight and obesity by sex and by survey according to the two different definitions are illustrated in Table 2. The table shows comparable time trends in the prevalence rates of overweight and obesity using the two references. Nonsignificant changes in the prevalence estimates in both sexes were found between 1975 and 1985, whereas the prevalence of overweight in boys

Table 2 Comparative prevalence of overweight and obesity among boys and girls in Jena 1975

Overweight Boys German reference International reference Girls German reference International reference Obesity Boys German reference International reference Girls German reference International reference

1995

2001

N

% (95% CI)

N

% (95% CI)

N

% (95% CI)

N

% (95% CI)

40 55

4.0 (2.9–5.4) 5.5 (4.2–7.1)

29 52

3.7 (2.5–5.3) 6.7 (5.0–8.6)

69 75

7.0 (5.5–8.7) 7.6 (6.1–10.1)

86 111

8.9 (7.4–10.9) 11.5 (9.9–13.7)

33 45

3.3 (2.4–4.6) 4.5 (3.3–6.0)

22 45

2.9 (2.0–4.2) 6.0 (4.4–7.9)

64 98

7.0 (5.5–8.8) 10.8 (8.9–12.8)

78 106

8.2 (6.8–10.2) 11.1 (9.2–13.3)

9 7

0.9 (0.4–1.7) 0.7 (0.3–1.4)

6 5

0.8 (0.2–1.7) 0.6 (0.2–1.5)

24 19

2.4 (1.7–3.3) 1.9 (1.2–3.0)

22 17

2.3 (1.4–3.4) 1.8 (1.0–2.8)

8 7

0.8 (0.4–1.6) 0.7 (0.3–1.4)

4 2

0.5 (0.2–1.4) 0.3 (0.03–1.0)

7 6

0.8 (0.3–1.6) 0.7 (0.2–1.5)

17 9

1.8 (1.0–2.9) 0.9 (0.4–1.8)

Abbreviation: 95% CI, 95% exact confidence limits.

European Journal of Clinical Nutrition

1985


407 Boys (Mean / 95% CI) Girls (Mean / 95% CI)

0.0

BMI-SDS

-0.1 -0.2

-0.3 -0.4 1975

1985

1995 2001

Figure 1 Converted BMI values to SD scores using the national reference data for each year of investigation. 95% CI ¼ 95% confidence limits. The national reference (mean ¼ 0) differs significantly from 1975; 1985 and 1995 (Po0.001), not significantly from 2001 (P ¼ 0.076). Significant differences between: 1975–1985– 1995–2001, analysis of variance, Po0.001; 1975–1995, 1975– 2001, 1985–1995, 1985–2001, Tamhane-T2 test, Po0.001; 1995–2001, Tamhane-T2 test, P ¼ 0.009.

pronounced in girls aged 7–8 years. The changes in the distribution of BMI for boys and girls between 1985 and 1995 are characterized by increasing differences with increasing BMI range. Thus, an increasing skewness and large changes in the upper BMI range were found. This shift was also seen in the changes between 1995 and 2001, except in 7-year-old girls. In girls aged 7–8 years, the BMI data showed a slight decrease in the lower part of the distribution during this period of time. Mean difference plots comparing the distribution of BMI in the different intervals are shown in Figure 3 by single year of age for boys and girls aged 11–14 years. There are some broad similarities in the changes seen in the younger children between 1975 and 1985 as well as between 1985 and 1995. Whereas the first interval is characterized by little differences, pronounced increases in skewness with large shifts at the upper end of distribution were found between 1985 and 1995 (except for 14-year-old boys and girls). Between 1995 and 2001, a slight upward shift of the entire distribution was found, especially in boys. In the older age ranges, the variability in the BMI changes is greater than for younger children.

Discussion Data in Jena children (Germany) show increases in the prevalence estimates of overweight and obesity between 1975 and 2001. The prevalence estimates depend on the selection of the reference population. But the used reference does not affect trend estimates over time of overweight and obesity. The trends are similar, independent of criteria. The comparatively low prevalence estimates in all cohorts as well as the converted BMI values to SDS using the German

reference data indicate that Jena children tend to be leaner than their peers in other German regions. Currently, in Germany, the prevalence of overweight is 8–12% among 5to 6-year-olds, 10–18% among 9- to 10-year-olds and 14–17% among 13- to 15-year-olds according to the national reference, and 4 (5- to 6-year-olds) to 8% (13- to 15-yearolds) of the children are obese (Wabitsch et al., 2002; Kromeyer-Hauschild, 2004). The prevalence estimates among German children increased with time and age, and significant disparities between geographical regions were found. Relative to Germany as a whole or to the population in East Germany, Jena is characterized by a higher proportion of upper middle class families with an academic background. Higher social status (highly qualified occupation of father vs manual worker) is a protective factor for overweight in Jena children (Kromeyer-Hauschild et al., 1999). This inverse relationship between social status and overweight was found frequently (De Spiegelaere et al., 1998; Sakamoto et al., 2001; Langnaese et al., 2002) and may be the main explanation for the low prevalence estimates in children from Jena. The changes in the BMI-SDS values between 1975 and 1995 indicate a decreasing deviation from the German reference values, confirming the trend of increasing prevalence estimates. Whereas between 1995 and 2001 a further significant increase in the SDS value was found (the 2001 data set was not significantly different from the German reference data), nonsignificant changes in the prevalence estimates (only a trend to higher estimates) were seen. Prevalence estimates focus only on a small proportion of a given sample, and the mean difference plots focus on changes at the population level. The plots give a more complex picture of population changes and in our study confirm the findings concerning the BMI-SDS values. The observations of population changes are very important, because relatively small increase in the average weight of a population has had a significant effect on the incidence of obesity. An increase in the average BMI in the US population from 26.7 to 28.1 kg/m2 between 1991 and 2000 has led to a marked increase in the number of people with BMI430 kg/m2 (Friedman, 2003). In our study, the mean difference plots show that the whole distribution tended to shift upward in all age groups between 1985 and 1995 and in older ages between 1995 and 2001, suggesting an increase in BMI across the entire population. Increasing weight appears to be a characteristic feature of the population as a whole, and does not seem to be a separate problem only for people suffering from obesity. As the distribution of BMI has shifted in a skewed fashion, the heaviest children have become even heavier. In most studies, the lower part of the BMI distribution has been found to remain unchanged. In the upper BMI range, although, it has been seen that there is a similar pattern of increase among other populations (Troiano and Flegal, 1998; Hulens et al., 2001). In Aachen (former western part of Germany), between 1966 and 1999, a BMI increase was observed in preschool children. The difference was more marked in the upper BMI range (HerpertzEuropean Journal of Clinical Nutrition


408 7 years 1985-1975

BMI difference

3

1995-1985

2001-1995

Boys Girls

2 1 0 -1 14 16 18 20 22 24 26 28

14 16 18 20 22 24 26 28

14 16 18 20 22 24 26 28

8 years 1995-1985

1985-1975

2001-1995

BMI difference

3 2 1 0 -1 14 16 18 20 22 24 26

14 16 18 20 22 24 26

14 16 18 20 22 24 26

9 years 1985-1975

1995-1985

2001-1995

14 16 18 20 22 24 26

14 16 18 20 22 24 26

14 16 18 20 22 24 26

1995-1985

2001-1995

BMI difference

3 2 1 0 -1

10 years 1985-1975

BMI difference

3 2 1 0 -1 14 16 18 20 22 24 26 BMI mean

Figure 2

14 16 18 20 22 24 26 BMI mean

14 16 18 20 22 24 26 BMI mean

BMI distribution mean difference plots in Jena children aged 7–10 years, by age group and time interval.

Dahlmann et al., 2003). In the USA, Flegal and Troiano (2000) found no changes in the lower part of the BMI distribution in younger children and a slight shift upwards in the whole population with increasing age. European Journal of Clinical Nutrition

The marked increase in the prevalence of overweight and obesity in children from Jena between 1985 and 1995 may be linked to transitions towards a more Western lifestyle, characterized by an essentially unlimited supply of con-


409 11 years

BMI difference

3

1985-1975

1995-1985

2001-1995

14 16 18 20 22 24 26

14 16 18 20 22 24 26

1985-1975

1995-1985

2001-1995

14 16 18 20 22 24 26

14 16 18 20 22 24 26

14 16 18 20 22 24 26

Boys Girls

2 1 0 -1 14 16 18 20 22 24 26 12 years

BMI difference

3 2 1 0 -1

13 years 1985-1975

1995-1985

14 16 18 20 22 24 26

14 16 18 20 22 24 26

2001-1995

BMI difference

3 2 1 0 -1

14 years

14 16 18 20 22 24 26

1985-1975

1995-1985

2001-1995

14 16 18 20 22 24 26 BMI mean

14 16 18 20 22 24 26 BMI mean

14 16 18 20 22 24 26 BMI mean

BMI difference

3 2 1 0 -1

Figure 3

BMI distribution mean difference plots in Jena children aged 11–14 years, by age group and time interval.

venient, highly palatable, energy-dense foods and low levels of physical activity, caused by the reunification of Germany in 1989 (see Kromeyer-Hauschild et al., 1999 for further details). Substantial changes in lifestyle (ownership of

automobiles, televisions, telephones, computers, labor-saving machines in households) that are frequently lower energy expenditure have spread throughout the population rapidly. These environmental changes may be an explanation for the European Journal of Clinical Nutrition


410 weight gain across the entire study population. These changes show the effects not only on the overweight and obese but also on the normal and underweight children. Between 1995 and 2001, it was seen among girls that the BMI in the lower part of the distribution slightly sank. The sinking in the BMI appears to compensate for the high amount of weight gain among girls between 1985 and 1995. The reasons for this decrease remain unclear. Possibly, there is a kind of adaptation to the new environment in a subgroup. The differences in the magnitude of the weight gain indicate that there are individual differences (determined by genetic or by socio-economic factors) in the response to the environment. The tendency to an increasing variation found in the study on older children may be linked to an increasing variation in the environment in these age groups (e.g. regarding food habits or leisure activities). Because obesity is difficult to treat, early prevention of obesity in childhood and adolescence is an important goal. All children, especially risk groups, should be encouraged to make behavioral changes within their present environment. Therefore, the focus should be to achieve better identification within the population as a whole, which groups are high or low risk for weight gain (before they are overweight or obese). When looking for the nutritional status in individuals, it is important to assess not only BMI values but also the body composition (Maynard et al., 2001). Even though the BMI is the most widely used measurement to define obesity, it has limitations on an individual level and is only a proxy measurement of body fat (Daniels et al., 1997; Freedman et al., 2005). Two persons with the same amount of body fat can have quite different BMI values, because BMI reflects not only body fat, but also bone, muscle and other lean tissues (Ellis et al., 1999). Measurements of circumferences and skinfold thicknesses are more sensitive indicators of body fat but have a much higher relative measurement error compared to BMI (Sardinha et al., 1999; Bedogni et al., 2003). From a population perspective, the BMI offered a reasonable measure with which to assess body fat (Poskitt, 1995; Dietz and Robinson, 1998). Greater public health attention should be focused on changes over time in the distribution of BMI – not only in the prevalence rates. Weight gain in the entire population demands an investment in the primary and secondary prevention of obesity and overweight in the population to establish a healthy, active lifestyle and to keep children and adolescents within a range of body weight that is considered to be healthy.

References Bedogni G, Iughetti L, Ferrari M, Malavolti M, Poli M, Bernasconi S et al. (2003). Sensitivity and specificity of body mass index and skinfold thicknesses in detecting excess adiposity in children aged 8–12 years. Ann Hum Biol 30, 132–139. Bundred P, Kitchiner D, Buchan I (2001). Prevalence of overweight and obese children between 1989 and 1998: population based series of cross sectional studies. BMJ 322, 326–328.


Cole TJ, Bellizzi MC, Flegal KM, Dietz WH (2000). Establishing a standard definition for child overweight and obesity worldwide: international survey. BMJ 320, 1240–1243. Cole TJ, Green PJ (1992). Smoothing reference centile curves: the LMS method and penalized likelihood. Stat Med 11, 1305–1319. Daniels SR, Khoury PR, Morrison JA (1997). The utility of body mass index as a measure of body fatness in children and adolescents: differences by race and gender. Pediatrics 99, 804–807. De Spiegelaere M, Dramaix M, Hennart P (1998). The influence of socioeconomic status on the incidence and evolution of obesity during early adolescence. Int J Obes Relat Metab Disord 22, 268–274. Dietz WH (2004). Overweight in childhood and adolescence. N Engl J Med 350, 855–857. Dietz WH, Robinson TN (1998). Use of the body mass index (BMI) as a measure of overweight in children and adolescents. J Pediatr 132, 191–193. Ellis KJ, Abrams SA, Wong WW (1999). Monitoring childhood obesity: assessment of the weight/height index. Am J Epidemiol 150, 939–946. Flegal KM, Troiano RP (2000). Changes in the distribution of body mass index of adults and children in the US population. Int J Obes Relat Metab Disord 24, 807–818. Freedman DS, Wang J, Maynard LM, Thornton JC, Mei Z, Pierson RN et al. (2005). Relation of BMI to fat and fat-free mass among children and adolescents. Int J Obes Relat Metab Disord 29, 1–8. Friedman JM (2003). A war on obesity, not the obese. Science 299, 856–858. Goran MI (2001). Metabolic precursors and effects of obesity in children: a decade of progress, 1990–1999. Am J Clin Nutr 73, 158–171. Gunnell DJ, Frankel SJ, Nanchahal K, Peters TJ, Davey SG (1998). Childhood obesity and adult cardiovascular mortality: a 57-y follow-up study based on the Boyd Orr cohort. Am J Clin Nutr 67, 1111–1118. Herpertz-Dahlmann B, Geller F, Bohle C, Khalil C, Trost-Brinkhues G, Ziegler A et al. (2003). Secular trends in body mass index measurements in preschool children from the City of Aachen, Germany. Eur J Pediatr 162, 104–109. Hulens M, Beunen G, Claessens AL, Lefevre J, Thomis M, Philippaerts R et al. (2001). Trends in BMI among Belgian children, adolescents and adults from 1969 to 1996. Int J Obes Relat Metab Disord 25, 395–399. Kain J, Uauy R, Vio F, Albala C (2002). Trends in overweight and obesity prevalence in Chilean children: comparison of three definitions. Eur J Clin Nutr 56, 200–204. Kromeyer-Hauschild K (2004). Aktuelle Aspekte der Gewichtsentwicklung bei Kindern und Jugendlichen. Pa¨diat Prax 64, 371–378. Kromeyer-Hauschild K, Wabitsch M, Kunze D, Geller F, Gei HC, ¨ r den Body-mass-Index fu ¨ r das Hesse V et al. (2001). Perzentile fu Kindes- und Jugendalter unter Heranziehung verschiedener deutscher Stichproben. Monatsschr Kinderheilkd 149, 807–818. Kromeyer-Hauschild K, Zellner K, Jaeger U, Hoyer H (1999). Prevalence of overweight and obesity among school children in Jena (Germany). Int J Obes Relat Metab Disord 23, 1143–1150. Langnaese K, Mast M, Muller MJ (2002). Social class differences in overweight of prepubertal children in northwest Germany. Int J Obes Relat Metab Disord 26, 566–572. Livingstone B (2000). Epidemiology of childhood obesity in Europe. Eur J Pediatr 159 (Suppl 3), S14–S34. Magarey AM, Daniels LA, Boulton TJ (2001). Prevalence of overweight and obesity in Australian children and adolescents: reassessment of 1985 and 1995 data against new standard international definitions. Med J Aust 174, 561–564. Martin R, Saller K (1957). Lehrbuch der Anthropologie. Bd. 1. Stuttgart: Gustav Fischer. Maynard LM, Wisemandle W, Roche AF, Chumlea WC, Guo SS, Siervogel RM (2001). Childhood body composition in relation to body mass index. Pediatrics 107, 344–350. Poskitt EM (1995). Defining childhood obesity: the relative body mass index (BMI). European Childhood Obesity group. Acta Paediatr 84, 961–963.


411 Sakamoto N, Wansorn S, Tontisirin K, Marui E (2001). A social epidemiologic study of obesity among preschool children in Thailand. Int J Obes Relat Metab Disord 25, 389–394. Sardinha LB, Going SB, Teixeira PJ, Lohman TG (1999). Receiver operating characteristic analysis of body mass index, triceps skinfold thickness, and arm girth for obesity screening in children and adolescents. Am J Clin Nutr 70, 1090–1095. Troiano RP, Flegal KM (1998). Overweight children and adolescents: description, epidemiology, and demographics. Pediatrics 101, 497–504.

Wabitsch M, Kunze D, Keller E, Kiess W, Kromeyer-Hauschild K (2002). Adipositas bei Kindern und Jugendlichen in Deutschland. Deutliche und anhaltende Zunahme der Pra¨valenz – Aufruf zum Handeln. Fortschr Med Orig 120, 99–106. Weinsier RL, Hunter GR, Heini AF, Goran MI, Sell SM (1998). The etiology of obesity: relative contribution of metabolic factors, diet, and physical activity. Am J Med 105, 145–150. Zellner K, Jaeger U, Kromeyer-Hauschild K (2004). Height, weight and BMI of schoolchildren in Jena, Germany – are the secular changes levelling off? Econ Hum Biol 2, 281–294.
