Fig. 2

Survival curves for: (A) ever and never use of post-diagnosis low-dose aspirin and (B) current, past and never use of post-diagnosis aspirin, with estimated restricted mean survival time (RMST) at 5-years after the start of follow-up. The survival curves and RMST were adjusted using regression standardization (i.e. one survival curve for each level of the exposure for each individual was predicted from the model using the observed covariates and the given exposure level, and then averaged at each exposure level). The difference in RMST provides an estimate of the difference in survival time at 5-years after the start of follow-up associated with exposure levels