The objective of non-inferiority trials is to compare a novel treatment to an active treatment with a view of demonstrating that it is not clinically worse with regards to a specified endpoint. It is assumed that the comparator treatment has been established to have a significant clinical effect (against placebo). These trials are frequently used in situations where use of a superiority trial against a placebo control may be considered unethical.
Non-inferiority is most easily assessed using a confidence interval (CI) approach.
Firstly we specify a non-inferiority margin, d. This is the maximum difference we are prepared to tolerate in a given direction if the new treatment is not to be considered (clinically) inferior.
If a 95% confidence interval for the difference between treatment means lies above or below this boundary value (in a favourable direction) then non-inferiority is deemed to have been established.
- Suppose a test treatment, T, is considered inferior if it produces a measurement of 3 points or more below some control treatment, C, on average.
- If a 95% CI for the difference between means, T – C, lies above d = -3 then the test treatment is deemed not inferior to control.
- The diagram below illustrates possible outcomes:
The choice of inferiority margin, d, affects the sample size calculation and the conclusion of the study. A general rule of thumb is that this quantity must be considerably smaller (1/2 or 1/3) than the minimal clinical difference we might use to calculate sample size in a superiority trial.
Some web links and articles:
- Good Enough: A Primer on the Analysis and Interpretation of Non-inferiority Trials, Kaul
- A Regulatory Perspective on Choice of Margin and Statistical Inference Issue in Non-inferiority Trials, H. M. James Hung
- Non-inferiority trials are unethical because they disregard patients’ interests. Bertele
- Reporting of Noninferiority and Equivalence Randomized Trials: An Extension of the CONSORT Statement, Piaggio
- Guideline on the choice of the non-inferiority margin
- Active-control clinical trials to establish equivalence or noninferiority: methodological and statistical concepts linked to quality, Gomberg-Maitland
- A randomized, double-blind, placebo-controlled noninferiority trial of amoxicillin for clinically diagnosed acute otitis media in children 6 months to 5 years of age, Le Saux
- Acute treatment of moderate to severe depression with hypericum extract WS 5570 (St John’s wort): randomised controlled double blind non-inferiority trial versus paroxetine, A Szegedi
- Book: The pros and cons of noninferiority trials, Pocock S.J.1published by Blackwells