All Times EDT
Keywords: HAP studies; linear mixed-effects model; paired t-test; type I error.
Human abuse potential (HAP) studies are clinical studies to evaluate the abuse potential of a new drug containing central nervous system (CNS) active drug substance. The outcomes are the pairwise differences of endpoints between positive control, test drug, and placebo. Currently the statistical model we use is a linear mixed-effects model and then t test for paired differences if model assumptions do not hold. Nonparametric approach will be applied if paired difference is not normal as well. Nonparametric approaches such as sign test are testing different hypothesis and should be used only if necessary. However, multiple tests for normality of outcomes may inflate the type I error rate. In this study, we used numerical simulation to evaluate the overall type I error rates under different scenarios in the deviation from normality in the mixed model across the entire testing procedure.