Skip to main content
Fig. 4 | BMC Medical Research Methodology

Fig. 4

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

Fig. 4

Risk ratios estimated by Poisson model (RRP) by varying the time to and prevalence of exposure Exposure status was simulated for each patient from a Binomial distribution with the following probabilities: 0.15, 0.25, and 0.40. Exposure times were then simulated from a Gamma distribution with scale parameter equal to 0.1 and shape parameter from 0 to 50. The true exposure effect (RRT) was set to 0.75

Back to article page

BMC Medical Research Methodology

ISSN: 1471-2288

Contact us