The sample provided 10,030 children of Dutch origin and 1,975 imputed cases, resulting in an analysis sample of 12,005 children (5,811 boys, 6,194 girls) aged 0–21 years. In total, data of 4,382 children aged 0–3 years and of 7,623 children aged 4–21 years were available. Data of 146 children aged 22–25 years were included in the dataset for fitting BMI reference charts, but they were excluded for further analyses.
presents the LMS values for BMI by age and sex. It shows that, for example, for boys aged 5 years the median (M) BMI was 15.64 whilst for girls this was 15.62, and that the coefficient of variation (S) of BMI was about 8% in infancy and rose up to 13–14% in adolescence.
LMS values for BMI (kg/m2) in Dutch 0–21 year olds in 2009 by age and sex.
shows the BMI distribution that correspond to the fitted LMS values for both sexes, including the 0, ±1, ±2, and −3 SDS lines. For comparison, the charts include the international cut-off values for thinness grade II and III, overweight and obesity 
. The distribution was highly skewed. From the age of three years on, the distance between the +1 and +2 SDS lines was 1.5 to 3 times as wide as between the −1 and −2 SDS lines at all ages. The +2, +1, −2 and −3 SDS lines corresponded well to the international cut-off values for respectively obesity, overweight, thinness grade II and III, although the IOTF cut-off values for obesity were a bit higher than the +2 SDS line for girls from the age of 11 years onward. In general, the median (0 SDS) curves for boys and girls were very similar, although BMI values for boys up to one year of age were slightly higher and during puberty slightly lower than for girls.
BMI distribution for Dutch boys and girls in 2009.
presents the mean BMI for boys and girls at all ages in 1980, 1997 and 2009. The mean BMI rose in both boys and girls. Since 1997, mean BMI in boys increased with 0 to 3%, and compared with 1980, it was up to 5% higher. In girls, mean BMI rose 0 to 2.4% since 1997 and was up to 4.5% higher than in 1980. The rise in mean BMI was seen in boys aged four years and older and in girlsz from the age of three years onwards, with the largest relative increase occurring around the age of eight to ten years.
Median (P50) BMI in 1980, 1997 and 2009 by age and sex.
compares overweight and obesity prevalence rates in 2009 with 1980 and 1997 for Dutch boys and girls 
. From this table we calculated an increase of 20–40% in overweight and of 40–100% in obesity since 1997. graphs these prevalence rates by age. At the age of two years, overweight and obesity prevalence rates in 2009 were slightly lower than in 1980 and similar to those in 1997. However, after the age of two years, the 2009 prevalence rates exceeded the rates observed in 1980 and 1997. For girls, the gap started to widen from the age of two years, whereas for boys this occurred from the age of four years. The largest gap with the previous studies was found around the ages of seven and eight years, when 15–19% of the children were overweight and 2–3% were obese. At some ages, we observed a three to four fold increase in the amount of overweight since 1980. For obesity, this ratio was even higher.
Table 3 Prevalence rates (%) of overweight (including obesity) and obesity according to IOTF cut-off values  in 1980, 1997 and 2009 by age and sex.
Figure 2 Prevalence of overweight and obesity in Dutch boys and girls according to international cut-off values .
From we calculated an increase of the relative proportion of obesity among overweight boys from 6% in 1980, through 10% in 1997, up to 14% in 2009. In girls, similar figures were found, starting at 7% in 1980, rising to 14% in 1997 and reaching 15% in 2009. Although the rise from 1997 to 2010 was not statistically significant (p
0.08 for boys and p
0.491 for girls), there was a rising trend in the severity of overweight.
presents the effect of geographical region and educational level of the parents on BMI SDS and compares these figures in 2009 with 1997. The mean BMI SDS of children whose parents had a low education was clearly higher than in those of higher educated parents. BMI SDS increased in all groups compared to 1997, but the increase was smallest in children of parents with a high educational level. Across the geographical regions we saw no large differences in BMI SDS in 2009. In 1997, in contrast, the BMI SDS in the major cities was much higher and in the west of the country it was lower than in the other regions. When we looked within the regions, we noticed a significant increase in BMI SDS since 1997 in all regions except for the major cities. In the major cities, the mean BMI SDS in 2009 even lay beneath 1997, although their 95% confidence intervals overlap. This corresponded well with the overweight prevalences that stabilized in the major cities (14.1% in 1997 vs. 14.4% in 2009, p
0.796), but rose substantially in the other regions (9.0% in 1997 vs. 12.9% in 2009, p<0.001).
Mean BMI SDS per geographical regions and parental education in 1997 and 2009.