诊断试验准确性研究设计及临床应用

Study Design and Clinical Practice of Diagnostic Accucary Test

新的诊断方法在临床开展之前,必须经由严格设计的诊断试验准确性研究进行评价。诊断试验准确性研究设计包括应用PICOS(P:Patient;I:Intervention;C:Comparison;O:Outcome;S:Study design)原则构建研究问题、确定诊断金标准、选择具有代表性的研究对象、估算样本量、同步盲法比较诊断试验与金标准结果、确立最佳截点值、评价诊断准确性以及遵循诊断准确性研究报告规范进行论文报告8个方面。诊断试验的准确性指标包括灵敏度、特异度、预测值和似然比。其中,诊断试验的似然比可帮助医生从验前概率获得验后概率。当医疗环境与研究环境相似、收治患者符合研究入组标准时,应用诊断试验研究的似然比有助于对目标疾病进行诊断与鉴别诊断。

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.