Simple method for developing percentile growth curves for height and weight

Abstract

The present paper demonstrates the ease of use of method I by Preece and Baine ([1978] Ann Hum Biol 5:1-24) in generating smoothed growth curves for both height and weight. Using the National Center for Health Statistics (NCHS) growth curve data, smoothed curves were developed and compared to those produced using the least-squares-cubic-spline method. Based on the lower sum of squares and better fit of shape as indicated by residual examination, it was concluded that the method I curve fitting procedure by Preece and Baine ([1978] Ann Hum Biol 5:1-24) fit centile growth curves for height and weight in 2-18-year-old male and female children as well as, if not better than, the least-squares-cubic-spline method used in developing the 1979 NCHS growth curves. Further, as this paper demonstrates, smoothed curves can be generated on a desktop computer using readily available software (the SOLVER function within Microsoft EXCEL).