Application of the Preece‐Baines growth model to cross‐sectional data: Problems of validity and interpretation

Abstract

The Preece-Baines growth model (PBGM) is a family of curves that conform to the shape of the human growth curve. It is most often used to analyze longitudinal records on individuals, but it can be applied to cross-sectional data from population surveys. The purpose of this study was to determine empirically the validity of applying the PBGM to cross-sectional data to make inferences about the timing and nature of the adolescent growth spurt. Longitudinal records (n = 339) from the Third Harvard Growth Study were analyzed individually using the PBGM. Results for each sex were pooled to characterize longitudinal growth in this population. Ten independent random cross-sectional samples were generated from the same data set. These were divided by sex and the age means for height were analyzed using the PBGM. Comparisons between the pooled longitudinal estimates and those from the random cross-sectional samples indicated that the PBGM can accurately estimate the age at peak height velocity in males, as well as other characteristics of the adolescent growth spurt when using cross-sectional data. In females, spurious results were obtained when cross-sectional data are used. Possible sources of bias are discussed. © 1994 Wiley-Liss, Inc.

Secular changes in the stature and weight of Taiwanese children, 1964–1988

“…Although the application of the Preece-Baines model to cross-sectional means has limitations (Zemel and Johnston, 1994), it does provide an estimate of the timing of the growth spurt for secular comparisons (Tanner et al, 1982).…”

Secular changes in body height and weight of Portuguese boys over one century

“…Velocity curves between the five periods were compared to assess differences in age at peak height velocity and used as an indirect estimate for tempo of growth and the onset of pubertal development. Pubertal growth velocity appears attenuated in population velocity curves estimated from differences in means between annual cohorts, when compared to individual growth velocity curves (Falkner and Tanner, 1986;Zemel and Johnston, 1994). Because most of the data are mixed-longitudinal the attenuation of pubertal growth velocity is likely to be less attenuated.…”

“…Because most of the data are mixed-longitudinal the attenuation of pubertal growth velocity is likely to be less attenuated. Nevertheless, Zemel and Johnston (1994) have shown that velocity curves can be accurately estimated using cross-sectional data. Although all of the Colégio Militar samples are truncated on the left side of height and weight distributions, because raw data were not available for all of the samples, truncated regression could not be used to estimate true height and weight trends.…”

The contribution of adolescent growth to shorter adult statures among girls of Chinese ancestry

“…Mean constant distance and velocity curves for Da-an group computed according to INCPHVAH category (smallest third; middle third; largest third). man longitudinal study listed in Table 8 ( Zacharias and Rand, 1983;Wember et al, 1992;Byard et al, 1993;Zemel and Johnston, 1994). All of the studies cited used the PB1 function, though differences exist in sampling methods and the number of measurements per individual.…”