Skip to main content
Fig. 3 | BMC Medical Research Methodology

Fig. 3

From: A novel framework for classification of selection processes in epidemiological research

Fig. 3

Selection during exposure at the population-level, illustrated by causal diagrams with subscripts of the nodes indicating time. a. A known determinant (C0) does not influence exposure initiation (E0) but exposure continuation (E1) b.E0 causes a side effect (A1) that influences E2. c.E0 and an unknown determinant (U0) cause an index or competing event (boxed S1) that precludes the disease outcome (D2) d.E0 and U0 cause an early disease event (boxed D1) that leads to depletion of susceptibles that precludes later disease events (D2) to occur. Induced inverse associations are marked with dashed lines

Back to article page