Forensic International Dental Database (FIDBv2) for adult age-at-death estimation in multiple forensic contexts: Strengthening the operationalization of the Lamendin criteria in a global scope model
Leandro H Luna,Roberto C Parra,Gonzalo Garizoain,Pablo A Rodríguez,P Sebastián Giannotti,Daniela Mansegosa,Eric Baccino,Douglas H Ubelaker,Laurent Martrille,Konstantinos Moraitis,Eleni Zorba,Manusmrati Mishra,Rodrigo Retamal,Ademir Franco,Paulo Miamoto,Andrea Baz,Frederic Camarasa,Lucio Condori,Karen Escalante-Flórez,Carmen Hernández Flores,Fredy Peccerelli,Fernando Retana-Milán,Natascia Rinaldo,Claudia Rivera,Clara Valderrama-Leal,Pascal Adalian,Claudia M Aranda
DOI: https://doi.org/10.1016/j.forsciint.2024.112230
Abstract:The aim of this study is to validate the FIDBv2 online procedure for adult age-at-death estimation using root dentine translucency (RDT) and periodontal retraction (PR) of single-rooted teeth in a worldwide sample. The sample includes 4810 teeth of 2559 individuals from 16 countries of America, Europe and Asia. Bias and inaccuracy between documented (DA) and estimated ages (EA) were calculated. Pearson and Intraclass Correlation Coefficients were computed to assess the strength of agreement between pairs of data, while Kolmogorov-Smirnov Z was used to evaluate the statistical significance of the differences. The percentages of correctly estimated cases within different age ranges were obtained to find trends in the reliability of the results. Most of the biases (-4.61-1.31 years) and inaccuracies (4.81-9.72 years) are low. The dispersion of EA increases with age and almost all the DA-EA correlations are above 0.75. DA-RDT and DA-PR correlations are positive, most of the former being high (0.74-0.91), and the latter being low (0.11-0.54). The highest percentages of correct estimations are identified for the ±7.5 and ±10 years ranges, and most comparisons of bias and inaccuracy between countries are non-significant. The high correlations between DA and EA suggest that the method is robust and reliable for a global application. Mean errors are low, with the best results found in the 30-69-year-old cohort. This research supports that the method is effective and accurate for age estimation in forensic contexts worldwide, thus reaffirming it is a generalizable procedure locally and internationally.