Limitation of site-stratified cox regression analysis in survival data: a cautionary tale of the PANAMO phase III randomized, controlled study in critically ill COVID-19 patients
Christian E Sandrock, Peter X K Song · 2024
Read Limitation of site-stratified cox regression analysis in survival data: a cautionary tale of the PANAMO phase III randomized, controlled study in critically ill COVID-19 patients by Christian E Sandrock, Peter X K Song (2024) 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/pmc-11654277.txt.
Description
Christian Sandrock and Peter Song warned in 2024 about a statistical trap in survival analysis, using a COVID-19 drug trial as their case. Trials often judge a drug by one preset test, but the way data are grouped can hide a real effect. In the trial they discuss, grouping results by hospital site dropped almost seventeen percent of outcome data from the main test. Power fell with the lost data.
Other groupings, by country or region, or pooling small sites, kept all patients in the count. Those versions pointed to a benefit from the drug. The authors show with math how the site grouping creates the artifact. Their caution aims at trial designers who set analysis plans before data arrive.
Catalog record
| CREATOR | Christian E Sandrock, Peter X K Song |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 17 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |