Introduction Determining factors in the causal pathway between an exposure and a disease is often the aim of etiologic research. For example, if a gene is associated with a certain disease, one would like to know if this relation can be explained by one or more intermediary variables. A naive approach to explore this is to condition on the intermediate factor and perform a stratified analysis. However, when an association remains, it is not always clear how this should be interpreted. It may be explained by a true direct effect of the exposure on the disease, or it may be explained by different types of error.
Methods We study different situations, using directed acyclic graphs, where despite absence of a direct effect, still a conditional relation between the exposure and the outcome remains. By quantifying the effect on the association parameter in each of the situations, we develop tools for sensitivity analysis of the completeness of adjusting for a mediator. The performance of the sensitivity tools is studied by simulation and by applying the tools on data of the LETS study. In this case-control study the question was whether the relation between blood group and venous thrombosis was mediated through coagulation factor VIII.
Results A remaining association can be caused by different types of measurement error, reverse causation or unmeasured confounding. In the LETS study intra-individual variation explained most of the remaining association.
Conclusion Sensitivity analyses should be performed before the conclusion is made that effects are not completely mediated.
Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.