Validity, reliability, and reproducibility of plaster vs digital study models: Comparison of peer assessment rating and Bolton analysis and their constituent measurements

Introduction: The objective of this validation study was to compare standard plaster models (the current gold standard for cast measurements) with their digital counterparts made with emodel software (version 6.0, GeoDigm, Chanhassen, Minn) for the analysis of tooth sizes and occlusal relationships—specifically the Bolton analysis and the peer assessment rating (PAR) index and their components. Methods: Dental casts were poured from 24 subjects with 8 malocclusion types grouped according to American Board of Orthodontics categories. Measurements were made with a digital caliper to the nearest 0.01 mm from plaster models and with the software from the digital models. A paired samples t test was used to compare reliability and validity of measurements between plaster and digital methods. Results: Reproducibility of digital models via the concordance correlation coefficient was excellent in most cases and good in some. Although statistically significant differences in some measurements were found for the reliability and validity of the digital models via the average mean of the absolute differences of repeated measurements, none was clinically significant. Grouping of the measurements according to the 8 American Board of Orthodontics categories produced no significant difference (Kruskal-Wallis test). No measurement associated with Bolton analysis or PAR index made on plaster vs digital models showed a clinically significant difference. The PAR analysis and its constituent measurements were not significantly different clinically between plaster and emodel media. Conclusions: Preliminary results did not indicate that digital models would cause an orthodontist to make a different diagnosis of malocclusion compared with plaster models; digital models are not a compromised choice for treatment planning or diagnosis.