Abstract: New diagnostic methods must be evaluated by rigorously designed diagnostic accuracy studies before clinical implementation. Designing a diagnostic accuracy study includes 8 procedures:constructing the research question with the PICOS (P:Patient; I:Intervention; C:Comparison; O:Outcome; S:Study design)framework, identifying an appropriate gold standard, choosing a representative patient sample, estimating the sample size, interpreting results of diagnostic tests and the gold standard blind to the other, setting up the optimal threshold, evaluating the diagnostic accuracy, and finally drafting a report according to the standards for reporting diagnostic accuracy. The accuracy of diagnostic tests includes sensitivity, specificity, predictive value (PV), and likelihood ratio (LR). The LR estimated by diagnostic tests can move clinicians from the pretest probability to a posttest probability. If the clinical setting is similar to that of the study and the patient meets all eligibility criteria of the study, the LR may facilitate the diagnostic process in clinical practice.