Table 1

Comparisons between superiority and non-inferiority trials

CharacteristicsSuperiority trialsNon-inferiority trials
Null hypothesis New treatment not superior to standard treatment/placeboNew treatment inferior to standard treatment
Alternative hypothesis New treatment is better than standard treatment/placeboNew treatment non-inferior to standard treatment
Non-inferior margin Not applicablePredetermined
Sequential testing Not possibleCan be performed
Significance level P<0.05 (two-sided)P<0.025 (one-sided)
Comparator Standard treatment or placeboStandard treatment (seldom placebo)
Possible outcomes 1. New treatment superior (or inferior) to standard treatment/placebo
2. Inconclusive
See figure 1
Subgroup analyses PossiblePossible