Questions of logic in Atlas methods of dental age estimation

To review aspects of the underlying methodological procedures in Atlas Methods of Dental Age Estimation (DAE) research publications. Attention is paid to issues of Reference Data supporting the Atlases, details of analytic procedures in the development of the Atlases, the statistical reporting of results of Age Estimation (AE), the problems of expressing uncertainty, and the viability of conclusions in the reporting of DAE studies.

Research reports utilizing Dental Panoramic Tomographs for creating Reference Data Sets (RDS) were studied to unravel the processes of creating Atlases with a view to determining the appropriate procedures for developing numerical RDS and compiling them into an Atlas format to enable DAE of child subjects without birth records.

The five different Atlases reviewed gave several different results in terms of AE. The possible causes of this were discussed - namely inadequate representation of Reference Data (RD) and lack of clarity in expressing uncertainty. It is suggested that the method of compiling Atlases needs to be more clearly defined. The yearly intervals described by some of the Atlases fails to take account of the Uncertainty of Estimates which is usually slightly greater than ±2½ years.

The review of published Atlas design papers in the field of DAE shows a number of different study designs, statistical procedures, and presentational styles, particularly with regard to the statistical procedures and findings. These show that Atlas methods can only be accurate to what amounts to at best a year.

Atlas methods lack the accuracy and precision of other methods of AE an example of which is the Simple Average Method (SAM).¹ This inherent lack of accuracy must be taken into account when using Atlas methods for AE.