Sample Size Calculations for Clinical Trials with Rate of Decline on a Quantitative Trait as the Endpoint

• Activity Number: 536
• Type: Contributed
• Date/Time: Thursday, August 2, 2007 : 10:30 AM to 12:20 PM
• Sponsor: Biometrics Section
• Author(s): Steven Edland*+
• Companies: University of California, San Diego
• Keywords: Clinical Trial ; Sample Size ; Phase III ; Power ; Compound symmetry ; Pilot Data ; Compound symmetry ; Pilot Data

Abstract

While estimates of marginal model parameters by gee are robust to mis-specification of the covariance structure, we caution that power calculation formulas that depend on covariance structure assumptions are not. We illustrate this mathematically and empirically using Monte Carlo simulations informed by data from a recently completed clinical trial of a treatment to slow the progression of Alzheimer's disease. We also caution that covariance estimates from shorter term pilot studies may not be representative of the covariance that will be observed in the clinical trials to be powered, because the covariance often depends on the length of follow-up and interval between observations. We derive an alternative power formula that can be used to power clinical trials of arbitrary length and interval between follow-up that does not require explicit assumptions about the covariance structure.