Clinical trials must cope better with multiplicity

Nat Med. 2012 Nov;18(11):1607. doi: 10.1038/nm1112-1607.

Abstract

Clinical trials typically address more than one question. But in attempting to protect against misleading results that are due to chance when multiple interrelated tests are run simultaneously, researchers sometimes apply overly strict statistical devices that mask true effects. They should give more consideration to choosing the type of statistical analysis that fits best.

MeSH terms

  • Clinical Trials as Topic / statistics & numerical data*
  • Humans
  • Patients
  • Physicians
  • Research Design