Resumo

SILVA, Amanda Falcão da; VELO, Marilia Mattar de Amoêdo Campos  e  PEREIRA, Antônio Carlos. Importance of reproducibility of methods for Dentistry diagnosis. RFO UPF [online]. 2016, vol.21, n.1, pp. 115-120. ISSN 1413-4012.

Objective: Reproducibility may be defined as the ability of agreement of results when the instrument is applied uniformly and repeatedly over invariant objects. Thus, this study aimed to perform a review on the reproducibility of methods and the statistical tools used in the studies, in order to get important information for better scientific evidence. Literature review: The reproducibility of a diagnosis method may be easily assessed, for example, by repeating independent tests performed to a given number of patients in a relatively short time, ensuring there is no change in the status of the disease initially assessed. Tests may be repeated by the same examiner in the case of intra-examiner agreement, or by different examiners for inter-examiner agreement. Among reproducibility measures available to assess clinical and epidemiological methods of dental caries, the overall percentage of agreement, Kappa statistic, Dice coefficient, and intraclass correlation coefficient (ICC) are highlighted, since they are generally extremely important instruments to verify and improve research accuracy. Final Considerations: The appropriate and recommended use of reliability instruments is of major importance, because it generates answers and guidance regarding the ability and accuracy of the instrument or method used.

Palavras-chave : Diagnosis.; Reproducibility of tests.; Validity of tests..