Comments regarding child obesity definition with BMI
Their data source was 1978-1980 in The Netherlands. Children’s percentiles of Body Mass Index showed this distribution.
The authors believe this BMI data is useful, because it apparently dates back to a time when there was not much obesity in the Netherlands children’s population.
As just pointed out, because of The Netherlands naturally tall population, the absolute values of body mass index aren’t usable elsewhere, but I think the “spread”, from median to 97th or 80th percentile is useful information. Child obesity, if it were defined in the 90-something percentile range, could benefit from these charts.
An extra reference: the CDC’s blurb on Childhood obesity bullet points.