In epidemiological studies, known confounders of incidence and prognosis have been used to construct propensity scores that effectively mitigate index event bias14. Such approaches are difficult in genetic studies, because there may be a substantial polygenic confounder that can neither be modelled directly nor easily captured by a propensity score11. Recently, the implications of index event bias have been discussed in the contexts of genetic association discovery15 and Mendelian randomisation2. Although the magnitude of bias appears small in currently typical settings, it is unclear how GWAS will be affected as studies increase in magnitude and polygenic analyses combine effects over thousands of variants.
