ABSTRACT
Generally, in the interpretation of clinical safety laboratory data, it is extreme values that indicate potential safety issues. We illustrate the application of multivariate extreme value modelling to such data. Applying the methods to a clinical trial dataset, we find unexpected extremal relationships that have potentially important implications for the interpretation of such data.
Subject(s)
Clinical Trials as Topic/methods , Drug-Related Side Effects and Adverse Reactions , Models, Statistical , Clinical Laboratory Techniques , Data Interpretation, Statistical , Humans , Multivariate AnalysisABSTRACT
Most clinical studies collect several safety-related laboratory variables. Generally, it is the extreme values of these variables that indicate potential safety issues. We illustrate the novel application of extreme value modelling to such data, with the aim of predicting the incidence of severe adverse drug reactions. By applying the methods to a clinical trial data set, we identify a dose-response relationship and use Bayesian techniques to identify a potential safety concern by making predictions from the fitted model, despite the small sample size.
