StATS: Growth curves (March 1, 2005)
The New York Times has a nice article about Dr. James Tanner, an expert on childhood growth.
This article contrasts the growth charts developed by Dr. Tanner with growth charts developed by the U.S. Centers for Disease Control and Prevention.
Dr. Tanner and his colleagues at the Institute for Child Health in London revolutionized growth charts in the 1960's by taking into account variations in a child's tempo of growth. Rather than one-size-fits-all curves, the Tanner-derived charts have separate curves, with a separate set of percentiles, for early-, average- and late-maturing boys and girls. Thus they expand the range of normal growth, using curves that are more forgiving of individual variation, especially around the crucial time of puberty.
These subtle differences derive from the kind of data used to create a growth chart. The American charts (and some modern European charts) primarily rely on cross-sectional data: researchers take a large group of children, separate them by age, measure them one time and then plot the distribution of heights and weights for each age group.
In contrast, Dr. Tanner and others believe a more accurate (and flexible) picture of a child's growth emerges from so-called longitudinal studies, where the same children are repeatedly measured over the course of many years of growth, so that individual variations in tempo - those who mature early and those who mature late - can be statistically incorporated into the charts. The data for such charts are more logistically difficult and costly to collect. But some growth experts believe such charts provide a more realistic picture of variation in individual growth patterns.
Even though the CDC approach may be simplistic, it is still relies on a very sophisticated model that allows for changes over time in the spread of the data and in the skewness of the data.
This page was written by Steve Simon while working at Children's Mercy Hospital. Although I do not hold the copyright for this material, I am reproducing it here as a service, as it is no longer available on the Children's Mercy Hospital website. Need more information? I have a page with general help resources. You can also browse for pages similar to this one at Category: Nonlinear regression.