Skip to main content
Fig. 2 | BMC Medical Research Methodology

Fig. 2

From: The hazard of using the Poisson model to cope with immortal time bias in the case of time-varying hazard

Fig. 2

Risk ratios estimated by Poisson model (RRP) by varying the shape parameter of the Weibull distribution for generating survival times, that is by moving from scenario A to scenario B (left panel) and from scenario A to scenario C (right panel) The left panel was obtained by assuming that survival times followed a Weibull distribution with λ = 0.75 and υ ϵ (0.1,1). The right panel was obtained by assuming that survival times followed a Weibull distribution with λ = 0.1 and υ ϵ (1,3). The true exposure effect (RRT) was set to 0.75

Back to article page