Power Calculations for Logistic Regression with Exposure Measurement Error

This program provides provides power calculations for logistic regression with a continuous exposure variable and an additional continuous covariate or confounding variable. A classical measurement error model is assumed for this implementation. Please refer to the following reference for details on the methods used:

Power and sample size calculations for generalized regression models with covariate measurement error.

Tor D. Tosteson, Jeffrey S. Buzas, Eugene Demidenko, Margaret Karagas. Statistics in Medicine Volume 22, Issue 7, 2003. Published Online: 10 Mar 2003

Odds ratio for a one standard deviation increase in the true exposure:

Prevalence of the binary outcome with no exposure

Sample size

Two-sided significance level

Correlation between the true and the measured exposure

Odds ratio for a one standard deviation increase in the covariate

Correlation between the covariate and the exposure

Main Page | Questions/Comments |