Transition analysis 3: New skeletal age indicators

To improve age estimates for adult human skeletons, age-of-transitions from early to late stages are estimated for multiple skeletal structures, many used for the first time for estimation purposes. Classified dichotomously (i.e., early and late stages), individual skeletal structures yield little information about age. But collectively they can improve age estimates for adult skeletons. Skeletons from four continents (N = 1638) provide a large and diverse sample suitable for developing a global age-estimation method. Sixty-nine early to late stage transitions were estimated for 47 cranial and postcranial structures; some structures used as age indicators have more than two morphological stages. Generalized Linear Models (GLM), GLM using log ages, and Generalized Additive Models (GAM) were used to generate age-of-transition curves, with the model that best fits the data presented here. Look-up tables are provided to estimate age. For each transition, the ages when 5, 10, 25, 50, 75, 90, and 95 % of the sample reached the more advanced of a pair of stages are given for both sexes grouped together. Sex-specific age-of-transitions are provided when transitions occur at different times for men and women. In addition to the look-up tables for manually estimating age, transitions from one stage to the next are the empirical backbone of a computer program, TA3-V1.0, for generating estimates through the combination of multiple curves, as outlined in an accompanying paper.