GET THE APP

Computing exact confidence intervals for informational odds ratio | 22884
Journal of Proteomics & Bioinformatics

Journal of Proteomics & Bioinformatics
Open Access

ISSN: 0974-276X

Computing exact confidence intervals for informational odds ratios in cancer genomic association studies


International Conference on Omics Studies

September 04-05, 2013 Holiday Inn Orlando International Airport, Orlando, FL, USA

Jimmy T. Efird

Scientific Tracks Abstracts: J Proteomics Bioinform

Abstract :

Traditionally, odds ratios (ORs) have been used as the effect measure in cancer studies of genomic association. However, ORs lack collapsibility and will yield biased estimates when adjusting for variables that are not true confounders. In contrast, adjusted informational odds ratios (IOR) are collapsible. While asymptotic confidence intervals (CI) for IORs have adequate nominal coverage for large sample sizes, they do not perform well for small sample sizes. Because IORs are a marginal relative effects estimate, existing exact CIs also are known to be overly conservative. In this presentation we demonstrate how to estimate an exact confidence for IOR that has better nominal coverage than currently available exact procedures.

Biography :

Jimmy T. Efird is an Associate Member of the Leo Jenkins Cancer at Brody School of Medicine. Additionally, he holds a joint appointment as Associate Professor in the Department of Public Health and as Epidemiologist/Chief Statistician (Director, Shared Resources) in the Center for Health Disparities. Dr. Efird received his Ph.D. from Stanford University (Epidemiology with a concentration in Biostatistics). His expertise includes statistical methods for assessing gene-environment interaction, clinical trial design, computing power and sample size for correlated samples, and multiplicity adjustment for confidence intervals. He has over 100 publications in scientific journals and technical proceedings. Additionally, Dr. Efird serves as a Senior Consultant for The NCRR-funded RCMI Translational Research Network Data and Technology Coordinating Center.

Top