Modeling Placebo Response Using Multiple Functional Predictors

• Activity Number: 243
• Type: Contributed
• Date/Time: Monday, August 10, 2015 : 2:00 PM to 3:50 PM
• Sponsor: Mental Health Statistics Section
• Author(s): Thaddeus Tarpey* and Eva Petkova and R. Todd Ogden and Adam Ciarleglio and Bei Jiang
• Companies: Wright State University and New York University School of Medicine and Columbia University and New York University and Columbia University/New York University
• Keywords: placebo response ; functional data ; personalized treatment

Abstract

Many modern approaches in medical treatment studies produce functional observations (e.g. EEG and fMRI) that can be used as functional predictors of outcome in clinical trials. This talk investigates using several functional predictors in a nonparametric regression to model treatment outcome. The primary goal of this analysis is to discover combinations of functional predictors that can distinguish placebo response from specific drug response to aid in personalized treatment decisions. The primary motivation for this work is in treating depression where placebo response rates tend to be high. Data from a large placebo-controlled depression clinical trial using baseline functional EEG measurements will be used to illustrate the methodology.