Fig. 1
From: Comparison of methods for estimating the attributable risk in the context of survival analysis
Estimation of the risk of invasive breast cancer attributable to ever use of menopausal hormone therapy at baseline as a time function, E3N cohort, 1992-2008. The dark solid and dark dashed curves pertain to the point estimates by KM and COX, respectively, the dark circles to the point estimates by the 4-year interval PCH; the light solid and light dashed curves, as well as the light circles, show the corresponding 95% confidence intervals. The WKM curves are not displayed because they almost coincided with the KM curves at the chosen scale