Siamak Sabour
Beheshti University of Medical Science, Iran
Title: Diagnostic (Validity, reliability) and prediction studies, rule of thumb in clinical researches
Biography
Biography: Siamak Sabour
Abstract
Reliability (precision) and validity (accuracy) are two important methodological issues in all fields of researches. The reliability is being assessed by inappropriate tests such as paired t.test and Pearson correlation coefficient which are among common mistakes. For quantitative variable Intra Class Correlation Coefficient (ICC) and for qualitative variables weighted kappa should be used. Regarding reliability or agreement, it is important to know that an individual based instead of group based approach using appropriate tests should be considered. Sensitivity, specificity,)positive predictive value (PPV), negative predictive value (NPV), likelihood ratio positive and likelihood ratio negative as well as diagnostic accuracy and odds ratio are among the tests to evaluate the validity of a single test compared to a gold standard. For prediction studies, we need two different cohort data set or at least one cohort dataset splitting that to develop our prediction model and then to validate it. Briefly, in reliability and validity as well as prediction studies, appropriate statistical tests with correct interpretation and correct study design should be applied by researchers. Otherwise, misdiagnosis and mismanagement of the patients cannot be avoided.