Awareness of index event and related biases2,10,11 has grown as attention turns to follow-up of GWAS. Our interest in this problem arose within the GENetIcs of sUbSequent Coronary Heart Disease (GENIUS-CHD) consortium30, which aims to identify risk factors for recurrent coronary events in patients with coronary heart disease. In a simulation study15, we showed that index event bias could be small in GWAS. Here, we have confirmed this for Crohn’s disease, but have shown an example in IPF where a strong effect on susceptibility appears to create a substantial bias that reverses the survival effect. This illustrates how index event bias can have variable effects in different studies, and reinforces the need to adjust for it to be confident in any genetic associations with prognosis.
