M Restrepo, D F Rojas-Gualdrón, A L de Farias, S Calle-Molina, K V Yupanqui, L Santos-Pinto, D G Bussaneli
MIH was not associated with occlusal alterations in children with mixed dentition. However, severe MIH and atypical caries lesions were associated with posterior crossbite, suggesting that the clinical expression of MIH may be more relevant than its presence alone in relation to specific occlusal manifestations.
PURPOSE: To compare diagnostic accuracy for MIH and its differential diagnosis between undergraduate dental students without previous clinical experience and paediatric dentistry specialists in active clinical practice.
METHODS: A cross-sectional analytical study included 70 participants: 40 s-year undergraduate dental students and 30 paediatric dentistry specialists. All participants attended an 8-h standardised training session on MIH one week before the assessment. Following this standardised training session, all participants assessed 115 validated clinical photographs presented in random order under identical conditions, classifying each image using the MIH Index. Diagnostic accuracy was estimated using the Obuchowski area under the curve (AUC) method for a non-binary gold standard, with 95% confidence intervals (95% CI).
RESULTS: Overall diagnostic accuracy was 0.887 (95% CI 0.876, 0.897) for students and 0.945 (95% CI 0.936, 0.954) for specialists. Most pairwise comparisons showed high discriminative ability, but lower AUC values were observed for diagnostically challenging pairs-particularly hypoplasia versus amelogenesis imperfecta (AUC = 0.76), atypical restorations versus atypical caries (AUC = 0.76), and hypoplasia versus non-MIH hypomineralisation defects (AUC = 0.80)-and among categories representing different clinical manifestations of MIH. Specialists outperformed students in 36 pairwise comparisons, reaching statistical significance.
CONCLUSIONS: Paediatric dentistry specialists achieved higher image-based diagnostic accuracy than undergraduate students, with both groups assessed under identical conditions following the same standardised training. Certain diagnostic pairs remained challenging even for specialists. Within the specialist group, accuracy did not vary significantly by years of experience, prior MIH training, or self-reported confidence.