Bias due to participant overlap in two-sample Mendelian randomization.
Burgess S, Davies NM, Thompson SG. · 2016
Read Bias due to participant overlap in two-sample Mendelian randomization. by Burgess S, Davies NM, Thompson SG. (2016) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/europmc-pmc5082560.txt.
Description
Stephen Burgess, Neil M. Davies, and Simon G. Thompson measure a hidden bias in Mendelian randomization studies, in Genetic Epidemiology in 2016. The method uses genes as natural experiments to judge whether a risk factor causes a disease. Two-sample versions take the gene-risk and gene-outcome data from separate studies, which should keep bias low. But the big consortia behind such data often share participants, and nobody knew what that overlap does.
Through simulations the authors show that, for a continuous outcome, bias grows in a straight line with the share of overlap: half overlap gives half the one-sample bias. With a binary outcome, using only control participants for the risk factor data removes the bias, while using cases too restores it. They supply formulas and a web tool for the expected bias, and advise consortia to publish estimates that leave case participants out.
Catalog record
| CREATOR | Burgess S, Davies NM, Thompson SG. |
|---|---|
| YEAR | 2016 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 53 KB |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |