Approaches to multiplicity in publicly funded pragmatic randomised controlled trials: a survey of clinical trials units and a rapid review of published trials.

For the remaining 20 studies effectiveness was required for at least one primary outcome; seven of these RCTs adjusted for multiplicity (either a formal procedure (Bonferroni or Holm) [ 8 ]; hierarchical testing methods; p -values between 0.025 and 0.05 considered to have borderline significance; or performing a post-hoc analysis with 97.5% confidence intervals).

also in 7,017 other papers