Prediction of adult height from height, bone age, and occurrence of menarche, at ages 4 to 16 with allowance for midparent height

Abstract

Multiple regression equations for predicting the adult height of boys and girls from height and bone age at ages 4 and upwards are presented. There is a separate equation for each half year of chronological age, and for pre and postmenarcheal girls at ages 11 to 14. These are based on longitudinal data from 116 boys and 95 girls of the Harpenden Growth Study and the London group of the International Children's Centre longitudinal study. The bone age used is the revised version of the Tanner Whitehouse standards, omitting the score for carpal bones (RUS age, TW 2 system). Boys aged 4 to 12 are predicted in 95% of instances to within ± 7 cm of true height, and at ages 13 and 14 to within ± 6 cm. Girls aged 4 to 11 are predicted to within ± 6 cm; premenarcheal girls aged 12 and 13 to within ± 5 and ± 4 cm, respectively; and postmenarcheal girls aged 12 and 13 to within ± 4 and ± 3 cm, respectively. Prediction can be somewhat improved by allowing for midparent height. One third of the amount that midparent height differs from mean midparent height is added or subtracted. An alternative system of equations which are based on initial classification by bone age rather than chronological age is given. These have about the same accuracy as the equations based on initial classification by chronological age, but allowance for bone age retardation is less. It is not clear which system is preferable. The equations probably apply to girls complaining of tall stature and boys and girls complaining of shortness and needing reassurance as to normality. In clearly pathological children, such as those with endocrinopathies, they do not apply.