The impact of using new and conventional methods for the age-at-death estimation in a Czech medieval population (Mikulčice, 9th-10th century): the relationship between age-at-death and linear enamel hypoplasia.

Abstract

Recent advances in age-at-death estimation from the skeleton indicate that some of the most commonly used methods based on linear regression provide different results compared to new techniques using Bayesian statistics, and underestimate individuals over 60 years old which leads to biased prehistoric lifespans. The question is how the choice of age-at-death estimation method can influence subsequent comparisons between different populations or further analysis, such as assessment of the effect of early stress on mortality in adult individuals. The aim of our work is twofold: firstly, to test the differences between age estimation methods evaluating one indicator (the auricular surface), namely the original (Lovejoy et al. 1985), revised (Buckberry & Chamberlain 2002) and newly developed (Schmitt 2005) methods, on the Early Medieval adult population from Mikulčice - III^rd church (Czech Republic, Central Europe). The secondary objective is to assess whether the different age distributions based on the different methods have an impact on age-dependent analyses, in this case the relationship between LEH and age-at-death. Our results showed that in the adult population from Mikulčice - III^rd church, the original and revised methods provided different mortality profiles: the proportion of individuals older than 60 years acquired using Lovejoy's method was only 6.7%, while the newer methods increased the proportion to 26.7% (Buckberry & Chamberlain 2002) and 23.9% (Schmitt 2005). The choice of age-at-death estimation, and thus the different age distributions, also resulted in differences in the achieved age of individuals with and without stress markers, and specifically in the significance of the differences found. This finding seeks to draw attention to the fact that inconsistency in the use of different age-estimation methods can influence the results of further analyses and cause problems when comparing burial grounds.