The utility of dental patterns in forensic dentistry.

The comparison of antemortem and postmortem dental patterns, which reflect a combination of dental states, can be quantified for human identification. However, the utility derived from the uniqueness of these patterns is limited by variations in oral health status related to population, age, and birth cohort. We analyzed dental pattern diversity from reference datasets that documented differences in oral health. Our analysis was based on full dentitions and partial dentitions available in forensic situations. To analyze the diversity of dental patterns, data from 3166 adults were extracted from the last 3 contemporary Spanish National Oral Health Examination Surveys, corresponding to the years 1993, 2000, and 2005. Each survey comprised 2 adult age groups (35-44 years and 65-74 years), therefore six datasets were available for our study. Our six samples showed substantial variability in oral health status (caries history) and dental code distribution, not only between age groups within the same survey year, but also between different survey years for the same age group. To test the overall diversity of dental patterns in each datasets, pairwise comparisons were performed and the total number of pattern matches was generated. We calculated total and conditional diversity (excluding individuals in whom all teeth were classified as unrestored or missing) for each forensic situation. To test the homogeneity of diversity estimates among the six datasets we used a random effect model that requires a parameter estimate together with its standard error. Total diversity values were low and heterogeneous. However, conditional diversities were high and homogeneous, which allowed all data to be pooled into a single database. Once data were pooled, high combined diversity values (above 0.99) were obtained for each forensic situation. This indicates their usefulness for forensic purposes. We conclude that the conditional diversity value derived from dental patterns is a potentially valuable tool with broad applicability for human identification. 2009 Elsevier Ireland Ltd. All rights reserved.