Hage A, Hage F. Kaplan-Meier Survival, Actuarial Survival, Censoring, and Competing Events - What is What?
Ann Thorac Surg 2022;
114:40-43. [PMID:
35367199 DOI:
10.1016/j.athoracsur.2022.03.044]
[Citation(s) in RCA: 5] [Impact Index Per Article: 2.5] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 11/03/2021] [Revised: 03/12/2022] [Accepted: 03/19/2022] [Indexed: 11/25/2022]
Abstract
Survival analyses, most commonly Kaplan-Meier curves, are frequently utilized in the field of cardiovascular medicine to analyze and graphically illustrate the differences in outcomes between two or multiples study groups in randomized-controlled trials. While Kaplan-Meier curves provide a nice representation of the survival (or the occurrence of other events of interest) of one or several groups of patients, they are commonly misused, especially in the setting of interval censoring/actuarial survival and/or competing events. Here, we sought to provide the reader with a simple example that clarifies some of these concepts. Many randomized-controlled trials (RCTs) and observational studies in the field of cardiovascular medicine utilize some sort of survival analysis, the most common being the reporting of Kaplan-Meier curves and a corresponding P value. Unfortunately in many instances, the terms Kaplan-Meier Survival and Actuarial Survival are used by authors interchangeably, where in fact, these two terms represent very different concepts that go beyond semantics and therefore need to be distinguished. This is also the case for other important concepts related to survival analysis such as right censoring, interval censoring, and competing events, and this is despite the presence of extensive articles on the proper use and interpretation of survival analyses for both statisticians and physicians without formal statistical background.(1) Here, we seek to provide the reader with a simple example that clarifies some of these concepts and emphasizes understanding the salient differences between them.
Collapse