Avoiding bias from weak instruments in Mendelian randomization studies

20114,048 citationsReview

Authors

Stephen Burgess · MRC Biostatistics Unit

Simon G. Thompson · MRC Biostatistics Unit

Abstract

Weak instrument bias is of practical importance for the design and analysis of Mendelian randomization studies. Post hoc choice of instruments, genetic models or data based on measured F-statistics can exacerbate bias. In particular, the commonly cited rule of thumb that F > 10 avoids bias in IV analysis is misleading.

Topics & Keywords

Genetic Associations and Epidemiology Statistical Methods in Clinical Trials Advanced Causal Inference Techniques

Publication Details

Published in: International Journal of Epidemiology

Volume 40, Issue 3, pp. 755-764

DOI: 10.1093/ije/dyr036

Field-Weighted Citation Impact: 6.61

Command Palette

Avoiding bias from weak instruments in Mendelian randomization studies

Authors

Abstract

Topics & Keywords

Publication Details