Summarizing and Communicating on Survival Data According to the Audience: A Tutorial on Different Measures Illustrated with Population-Based Cancer Registry Data [Corrigendum]

Abstract

Belot A, Ndiaye A, Luque-Fernandez MA, et al. Clin Epidemiol. 2019;11:53–65. Page 57, Number of life years lost (NLYL) section, left column, the sentence “where the quantity 1–FP(t) can be replaced by SP(t), ie, the classical survival function using the population mortality rates λP” placed immediately after equation 14 indicates that the quantity 1–FP(t) could be replaced by SP(t). This is wrong as the one-to-one relationship between hazard and risk applies only in all-cause mortality setting but not in competing risks settings (such as the cause-specific or relative survival setting). Indeed, FP (t) depends on both hazards (cancer and other causes) through the overall survival S(t). The authors therefore request the reader to ignore this sentence and apologize for this error. Read the original article  

Keywords

• survival
• competing risks
• relative survival setting
• conditional survival
• restricted mean survival time
• net survival
• crude probability of death
• number of life years lost