The predicted cumulative risk curves (as a function of time t) were calculated using ‘survfit.coxph’ within each stratum of sex as2\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\begin{array}{*{20}{c}} {1 - \widehat S\left( t \right) = 1 - {\mathrm{exp}}\left( { - \widehat {H}_0\left( t \right){\mathrm{exp}}\left( {{\mathbf{x}}^{\mathrm{T}}\hat {\mathbf{\upbeta }}} \right)} \right),} \end{array}$$\end{document}1−S^t=1−exp−H^0texpxT𝛃^,where \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\widehat S\left( t \right)$$\end{document}S^t is the cumulative survival at time t, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\widehat {H_0}(t)$$\end{document}H0 ^(t) is the estimated baseline cumulative hazard at time t, x is the vector of the predictor variables set to the values of interest, and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\widehat {\mathbf{\upbeta }}$$\end{document}𝛃^ is the vector of the estimated log hazard ratios for each predictor.
