The validity and reliability of mixed-dentition analysis methods: a systematic review.

BACKGROUND: The authors conducted a literature review to determine the validity and reliability of mixed-dentition analysis (MDA) methods used to predict permanent tooth size in the mixed dentition and to evaluate the effect of relevant variables. TYPES OF STUDIES REVIEWED: The authors searched eight online databases for studies of MDA. They selected 39 articles. Inclusion criteria included assessment of mesiodistal tooth width predictions of canines and premolars, use of study-model measurements with or without radiographs, reliability and validity of MDA values and a minimum sample size of 10. RESULTS: All MDA methods had positively correlated validity and high intrarater reliability. Mean differences were less than 2 millimeters with correlation coefficients that tended to be above 0.6. Correlation coefficients were less variable with multiple linear regression equations (MLREs) than with simple linear regression equations (SLREs) with the study-model method, but the opposite was true with the radiographic method. Polymorphisms based on ethnicity, sex, jaw or side of mouth did not meaningfully influence the validity of SLRE predictions, but some differences were apparent with the use of MLRE predictions. CLINICAL IMPLICATIONS: The results of this review call into question the clinical implications of the multiple variations of MDA that have been described in the literature.