Figure 2.
Illustration of immortal time bias in the Kiri et al. observational cohort study of inhaled corticosteroids in patients discharged with COPD.29
The top patient received the first prescription of ICS during the year after discharge and was excluded. The time between cohort entry and the first ICS prescription is immortal and unexposed (thick line) and, with the subsequent exposed time, should have been included in the analysis to avoid immortal time bias.