Title: 0389 - Validity and Reliability of Two Digital ABO Discrepancy Index Measurements
Diana Perilla (Presenter)
Fundación Universitaria CIEO-UniCIEO
Nathalia Espinel, Fundación Universitaria CIEO-UniCIEO
julieth Angarita, Fundación Universitaria CIEO-UniCIEO
Elena Ariza, Fundación Universitaria CIEO-UniCIEO
Juan Tamara, Fundación Universitaria CIEO-UniCIEO
Sonia Plaza, Fundación Universitaria CIEO-UniCIEO
Objectives: To evaluate the validity (accuracy) and reliability (reproducibility, repeatability) of the digital American Board of Orthodontics (ABO) Discrepancy Index (DI) calculations applied to two types of pretreatment digital models.
Methods: 24 subjects (15.04 ± 0.95 years, 7 females, 17 males) with varying degrees of malocclusion were selected. Total DI scores were measured by 3Shape OrthoAnalyzer software (3Shape Ortho System, Copenhagen, Denmark) over digital models obtained using a surface laser scanner (D-700, 3Shape, Copenhagen, Denmark), an intraoral scanning (TRIOS, 3Shape, Copenhagen, Denmark) to be compared to conventional plaster models (gold standard) manually measured with the ABO measuring gauge. Accuracy was assessed comparing to the gold standard method using the Wilcoxon signed rank sum test. Repeatability was evaluated comparing the three methods measured by the same operator in two different times (T1-T2) using the Spearman/Pearson rho correlation test. Reproducibility was determined by inter-operator agreement with the intraclass correlation coefficient (ICC). Additionally, Bland-Altman plots were also used.
Results: Significant differences (P<0.05) were found in overjet (mm) measured over intraoral scanning models. Overjet (points), diastema (points), crowding (mm) and total DI scored with both digital models presented also statistically significant differences (P<0.05). High levels of repeatability were found both with the surface laser scanner (0.94) and the intraoral scanner (0.95). Reproducibility values were between 0.98 (surface lase scanner) and 0.99 (intraoral scanner). However, the Bland- Altman plots showed a wide error range (-2.65-2.03) with the manual method.
Conclusions: Digital models showed adequate values of validity and reliability. Despite some occlusal traits showed statistically significant differences, the clinical differences were no relevant. The digital methods results showed a trend to overestimate the ID scores.
The submitter must disclose the names of the organizations with which any author have a relationship, the nature of the relationship, and the clinical or research area involved. The following is submitted: NONE