A model-independent method to determine restricted mean survival time in the analysis of survival curves
Autori
V. Damuzzo, M. Chiumente, L. Agnoletto, L. Leonardi, D. Mengato, A. Messori
Rivista
SN Comprehensive Clinical Medicine
Topic
analisi statistiche e meta-analysis HTA
Impact factor
0,1
0,1
Abstract
The restricted mean survival time (RMST) is a relatively new paramater proposed to improve the analysis of survival curves. As opposed to the median, the RMST has the advantage of capturing the overall shape of the survival curve, including the so-called “right tail”. One limitation of RMST lies in the mathematical complexity of its calculation (model-dependent analysis). In the present report, we describe a model-independent method that simplifies the calculation of RMST. The estimation approach (trapezoidal rule) is the same as that commonly employed in pharmacokinetics. In the analysis of 6 survival curves, the performance of the model-independent method was virtually the same as that of model-dependent methods.
The restricted mean survival time (RMST) is a relatively new paramater proposed to improve the analysis of survival curves. As opposed to the median, the RMST has the advantage of capturing the overall shape of the survival curve, including the so-called “right tail”. One limitation of RMST lies in the mathematical complexity of its calculation (model-dependent analysis). In the present report, we describe a model-independent method that simplifies the calculation of RMST. The estimation approach (trapezoidal rule) is the same as that commonly employed in pharmacokinetics. In the analysis of 6 survival curves, the performance of the model-independent method was virtually the same as that of model-dependent methods.
