Reproducibility and Validity of Diagnostic Tests

Every day, irrespective of whether dental professionals are clinicians, researchers, or public oral health service managers, they are faced with decisions that will have direct impact on their processes of planning, implementing, and evaluating their interventions. The good outcome of these processes will depend largely on the use of reproducible diagnostic tests, that is, tests capable of producing similar results when repeated and simultaneously valid measurements are made – that is to say – capable of measuring/diagnosing the clinical condition for which the test was developed. This makes it fundamental to know the concepts of reproducibility and validity, as well as the main methods for calculating them, their limitations, and indications in oral health. The particularities of each sample representative of the studied population, notably the prevalence of the attribute under study, in the conformation of both reproducibility and validity of a diagnostic test deserve special attention. These particularities are factors that hinder the generalization of the results of studies conducted in a certain sample/population, when compared with another that has a different epidemiological profile. In this chapter, the importance of presenting measurements that impart consistency to the results found is shown, for example, presenting the interval of confidence in reproducibility studies, or simultaneous presentation of measurements of validity that complement each other.