Using Whole-Cohort Information in Case-Control Studies

• Activity Number: 309
• Type: Contributed
• Date/Time: Tuesday, August 2, 2016 : 8:30 AM to 10:20 AM
• Sponsor: Biometrics Section
• Author(s): Alastair Scott* and Christopher Wild and Gustavo Amorim
• Companies: University of Auckland and University of Auckland and Ghent University
• Keywords: Case-control ; logistic regression ; calibration

Abstract

Standard analyses of data from case-control studies that are nested in a large cohort fail to make use of any baseline data that may be available for cohort members not included in the sub-study. Breslow et al (2013) reviewed methods for using this extra information and looked at the practical impact in a study of coronary heart disease in women drawn from the Women's Health Initiative (WHI).

In this paper, we first review the contributions of Norm Breslow to the field. We then extend some of these results and derive expressions for the reduction in variance from using the extra information. We look at the effect in some simulations and apply the new results to the same sub-study of the WHI.

Reference.

Breslow, N.E., Amorim, G., Pettinger, M.R., and Rossouw, J. (2013). Using the whole cohort in the analysis of case-control data. Statistics in Biosciences, 5, 232-249.