Abstract
In the current era, with increasing availability of results from genetic association studies, finding genetic instruments for inferring causality in observational epidemiology has become apparently simple. Mendelian randomisation (MR) analyses are hence growing in popularity and, in particular, methods that can incorporate multiple instruments are being rapidly developed for these applications. Such analyses have enormous potential, but they all rely on strong, different, and inherently untestable assumptions. These have to be clearly stated and carefully justified for every application in order to avoid conclusions that cannot be replicated. In this article, we review the instrumental variable assumptions and discuss the popular linear additive structural model. We advocate the use of tests for the null hypothesis of ‘no causal effect’ and calculation of the bounds for a causal effect, whenever possible, as these do not rely on parametric modelling assumptions. We clarify the difference between a randomised trial and an MR study and we comment on the importance of validating instruments, especially when considering them for joint use in an analysis. We urge researchers to stand by their convictions, if satisfied that the relevant assumptions hold, and to interpret their results causally since that is the only reason for performing an MR analysis in the first place.
Highlights
In many areas of application, it is important to be able to distinguish a causal association from a non-causal one to assess the relationship between a treatment, or exposure, X, and an outcome Y
As an example of an actively randomised instrumental variable (IV), we can conceive of a trial where unemployed individuals are randomly allocated to either participate in or abstain from a certain programme and their employment status recorded a year later, or where individuals are randomly assigned to a particular treatment and their health status monitored at a later time point
Mendelian randomisation (MR) with multiple independent IVs can be viewed as a meta-analysis where the individual ratio estimates corresponding to each Gk can be combined into a pooled inverse variance weighted (IVW) estimate (Burgess et al 2017a; Thompson et al 2016, 2017)
Summary
In many areas of application, it is important to be able to distinguish a causal association from a non-causal one to assess the relationship between a treatment, or exposure, X, and an outcome Y. As not all participants comply with their allocation—some refuse and others enter the programme or take the treatment even if assigned to the control group—the actual exposure may differ from that dictated by the randomisation. IVs essentially constitute an imperfect way of randomising the actual quantity X of interest. As it is imperfect, any conclusions drawn using an IV are weaker than those from a randomised controlled trial with full compliance. We will focus on the use of genetic IVs and on Mendelian randomisation studies, noting where these differ from other applications. We will use directed acyclic graphs (DAGs) throughout to illustrate the conditional independencies implied by the joint distribution of a set of variables (Dawid 1979; Pearl 2000)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have