Why comparing survival curves between two subgroups may be misleading

TL;DR

This paper reveals potential pitfalls in comparing survival curves between subgroups, highlighting the relationship with predictive values and cautioning against misleading conclusions in prognostic test validation with censored data.

Contribution

It establishes a link between survival estimates and predictive values, emphasizing the need for careful interpretation in subgroup survival comparisons.

Findings

Current methods can be misleading in subgroup survival analysis

Survival estimates relate directly to predictive values in classification

Implications for validation of prognostic diagnostic tests

Abstract

We analyse an issue when comparing survival curves between two subgroups. We show that there is a direct relationship between estimates of subgroups' survival at a time point and positive and negative predictive values in the binary classification settings. Our findings present a case where current methods of comparing survival curves between subgroups may be misleading. We think that this ought to be taken into account during the validation of prognostic diagnostic tests that predict two prognostic subgroups for a given disease or treatment, when the validation data set consists of censored data.