A new method for dental age estimation in adults.

Several statistical methods have been proposed for estimation of age from teeth in human adults. Due to the smallness of the material or statistical shortcomings in previous reports, improvements of the methods were desirable and seemed possible. As a contribution to such improvement various age-related changes have been studied in a material of 1000 teeth. For each type of change the kind of measurement most strongly related to age was chosen, and a multiple regression analysis was run with age as the dependent variable, using SPSS/PC+ statistical program. Statistically significant formulae were calculated for each type of tooth excluding molars. Because the sex of a deceased body may be unknown and the tooth colour influenced by changes after death, separate formulae excluding these factors were calculated. For the calculated formulae the Pearson correlation (tau) varied from 0.76 in mandibular second premolars (SEX and COLOUR excluded), to 0.91 for maxillary central incisors; a rather promising result compared with other methods. In addition, the intercorrelations between the age-related factors were calculated after control for the effect of age. In some types of teeth, but not in all, some of the factors displayed significant intercorrelation. To compare these formulae with previous methods an unbiased test in another independent material is needed.