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

Fig. 2

From: Impact of a non-constant baseline hazard on detection of time-dependent treatment effects: a simulation study

Fig. 2

Hazard functions, cumulative hazard curves and survival curves for lag until effect non-PH scenario. Lag period lengths investigated were tlag = 0, 1, 3 and 10 months within the maximum follow-up time t = 50, with the setting tlag = 0 representing PH. The lag period instantaneous change point times from control group hazard to treatment group hazard are indicated by the vertical gray lines. Decreasing times for treatment effectiveness as a result of increasing lag times are indicated by the decreased shading of the dashed lines used for the treatment group. Decresing, constant and increasing event rate scenarios are indicated by the green, purple and blue lines respectively

Back to article page