Longitudinal growth standards for preschool children.

The appropriateness of a preschool child's size at a particular age is usually assessed by comparing the child's measurement against growth charts. When the child has measurements at more than one age, the child's percentile status at one age is sometimes compared with the percentile status at a subsequent age, with the purpose of assessing the child's pattern over time or growth rate. This second use of growth charts is not entirely appropriate, for it assumes a high degree of 'tracking' in the population. Healthy children frequently cross percentile lines, the amount of shifting depending upon the age range under consideration, the child's sex, and whether the measurements are length or weight. Thus the correct evaluation of percentile-level changes over time is difficult. We present here conditional standards for length and weight in preschool children, which assess a child's length or weight taking into account previous measurements and/or a current measurement of a second anthropometric variable. The approach can be very flexible as regards ages of measurement, although complete flexibility requires use of a computer. In addition to its clinical application, this approach can be used to predict the future size of a child and may be valuable as an analytic tool in research situations in which control populations are not feasible.