Accurate adult height prediction is of clinical importance in assessin
g the need for pharmacological intervention and in the evaluation of t
he outcome of therapy. The methods currently in use are subject to a w
ide range of error, one source of which is the use of bone age (BA) me
asurements, We have developed a computer model for predicting adult he
ight in pubertal boys without using BA determinations. The model is ba
sed on the existing Infancy-Childhood-Puberty model and calculates the
onset of the pubertal growth spurt. predicted adult height was assess
ed using this new model and four others in a group of normal boys and
in a group of short normal boys receiving growth hormone. Calculated f
inal heights by all the methods were. not significantly different. Inc
orporation of paternal height into the prediction equations increased
the accuracy of the prediction. It was concluded that our new model is
as accurate as existing methods of predicting final height that invol
ve assessing BA.