Research in Translation

Research in Translation Research in Translation articles discuss a particular drug, treatment, or public health intervention in the context of translation from early research to clinical research, or clinical evidence to practice.

See all article types »

Mendelian Randomisation and Causal Inference in Observational Epidemiology

  • Nuala A Sheehan mail,

    To whom correspondence should be addressed.

  • Vanessa Didelez,
  • Paul R Burton,
  • Martin D Tobin
  • Published: August 26, 2008
  • DOI: 10.1371/journal.pmed.0050177

Reader Comments (3)

Post a new comment on this article

Interpretation of Figures 3 and 4

Posted by plosmedicine on 31 Mar 2009 at 00:29 GMT

Author: Eyal Shahar
Position: Professor
Institution: The Univesrity of Arizona
Submitted Date: September 10, 2008
Published Date: September 12, 2008
This comment was originally posted as a “Reader Response” on the publication date indicated above. All Reader Responses are now available as comments.

I believe that in both Figure 3 and Figure 4, the marginal association between Gene 1 and disease is not a valid estimate of the effect of Gene 1 on disease. Gene 2 is a confounder in both cases. There is a back-door, unblocked path from Gene 1 to disease (Gene 1--Gene 2--Exposure--Disease), which creates a non-causal association between Gene 1 and Disease.
Although the segment "Exposure--> Disease" is part of that path, the full path is not a causal path for the effect of Gene 1 on Disease.

I would be interested in the authors' response.

No competing interests declared.