Study Design and Statistical Methodology in Biomarker Evaluation

• Activity Number: 76
• Type: Contributed
• Date/Time: Sunday, August 4, 2013 : 4:00 PM to 5:50 PM
• Sponsor: Biometrics Section
• Author(s): Qian Li*+
• Companies: NIH, NCCAM
• Keywords: Biomarker Evaluation ; Study Design ; Statistical Methodology ; Classification

Abstract

The break-through development in technology and computing capacity in recent years makes it possible to identify biomarkers at the genetic and proteomic levels as well as with medical imaging. Despite the booming interest in biomarker development, the study design and statistical methodologies that are used in biomarker evaluation in medical literatures are often problematic, which leads to equivocal study conclusions. To help researchers choose valid study design and statistical methods, the common study design and statistical methodologies will be presented by classifying biomarkers into four categories based on their potential clinical uses. Issues of hybrid study designs that are used to evaluate one, or a panel of biomarkers for multiple uses, will be discussed as well.