External validation of a Cox prognostic model: principles and methods.
Royston P, Altman DG. · 2013
Read External validation of a Cox prognostic model: principles and methods. by Royston P, Altman DG. (2013) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/europmc-pmc3667097.txt.
Description
A risk model earns trust only when tested on fresh patients, Royston and Altman argue in a 2013 BMC Medical Research Methodology article. Their subject is the Cox model, a standard method that foretells survival over time. Outside checks for such models were thin in print. A sound check must test two things: how well the model sorts patients by risk, and how close its foretold chances run to what happens.
How much can be checked depends on what the first study printed. The authors give methods for three levels of detail, from the risk score alone up to the full survival curve. The hardest task, judging accuracy of foretold chances, gets a way to rebuild the missing base curve. They run the methods on two breast cancer data sets, one to build and one to test.
Catalog record
| CREATOR | Royston P, Altman DG. |
|---|---|
| YEAR | 2013 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 68 KB |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |