Synthetic standards for body weight.

Growth charts represent body stature, body weight, and body mass index (BMI) from birth to maturity. Due to secular changes in these parameters, growth charts tend to become outdated, and must be revised from time to time. Recently, we developed alternative strategies that facilitate developing and renewing growth charts, and suggested synthetic standards for body stature. The increasing prevalence of obesity has made it necessary to develop similar techniques also for monitoring body weight and BMI. Two-hundred-and-forty historic and modern growth studies (108 studies of male growth, 132 studies of female growth) were selected from 22 European, 6 American, 3 African, and 6 Asian nations, published between 1831 and 2001. The studies contained annual information on weight and stature, either between birth and 6 years, or between 6 years and maturity, or information on the whole age range between birth and maturity. Since historic studies up to the mid-20th century usually ignore the fact that body weight (in contrast to body stature) is not normally distributed, a group of 92 more recent studies (45 male, 47 female), published between 1943 and 2001, presenting centiles for weight, was chosen for additional analysis. Furthermore, the skewness of body weight distributions, was investigated in original raw data of body weight obtained from five well reputed longitudinal growth studies, performed at Jena, Germany, Lublin, Poland, Paris, France, Prague, Czech Republic, and Zürich, Switzerland. Average body stature and average weight differ markedly between different populations. But within the same population, both parameters are closely interrelated. In males, birth length and weight correlated with r = 0.503, stature and weight correlated with r = 0.873 at the age of 2 years; with r = 0.882 at the age of 6 years; with r = 0.935 at the age of 14 years, and with r = 0.891 at the age of 18 years. Similar results were obtained in females. At birth, length and weight correlated with r = 0.619. Stature and weight correlated with r = 0.863 at the age of 2; with r = 0.912 at the age of 6; with r = 0.935 at the age of 12; and with r = 0.918 at the age of 15 years. Tables of linear regression coefficients for relative stature and weight at all ages enable the reversal of the process of the meta-analysis and allow the generation of synthetic growth references for stature and weight. Synthetic reference charts help in the revision of current growth charts without much additional effort, and may be used for populations for which autochthonous growth standards are not available. [ABSTRACT FROM AUTHOR]