Preprint / Version 1

Methods Matter

Beware of collider stratification bias when analyzing recurrent injuries

##article.authors##

  • Ian Shrier
  • Steven Stovitz
  • Chinchin Wang
  • Russell Steele

DOI:

https://doi.org/10.31236/osf.io/7njf8

Keywords:

causal inference, collider stratification bias, rehabilitation, subsequent injuries

Abstract

A major objective of sport medicine research is to assess causes of injuries. Studying causes of subsequent injuries, including the effects of rehabilitation programs, presents unique challenges to provide appropriate guidance for treatment. In this commentary, we highlight the issue of “collider stratification bias”. Collider stratification bias is a major concern whenever we restrict populations based on variables that have multiple causes, which is common in surveillance programs studying subsequent injuries.

Downloads

Posted

2021-04-21