Table 1 Classification of other methods used to define the margins in the systematic review
Subtype | Definition | Frequency (n = 68) |
|---|---|---|
Expert opinion | The non-inferiority margin was chosen based on expert opinion. It also includes defining the margin based on: (1) the Delphi approach, (2) a threshold (e.g., superiority margin) for clinical efficacy that was chosen from the literature and considered by the authors, researchers, or experts as a clinically acceptable non-inferiority margin | 42 (62%) |
Based on literature review | 1. Non-specific literature review: the choice of the margin was based on a review of the literature without indicating how the review was conducted, and what types of historical data were reviewed | 3 (5%) |
2. Based on historically controlled data: the margin was defined based on the assessment of historical experience from non-concurrently controlled trials. The relative efficacy of the historical experience of active comparator was assessed against a historical group (e.g., placebo group, spontaneous cure rate group, or outcome without treatment group) | 2 (3%) | |
3. Review of other types of literature: the margin was defined based on the assessment of other types of literature (e.g., observational studies) | 2 (3%) | |
The margin was used in other non-inferiority trial(s) with similar design | A similar margin was used in other non-inferiority trial(s) of drugs that are used to treat the same indication, regardless of whether the active comparator was used in these trials or not | 9 (13%) |
Regulatory consultation/guideline | The choice of the margin was based on one of the following: recommendations by a regulatory authority, following a guideline from a regulatory authority without indicating how the margin was exactly defined, or used a margin that is provided by one of the regulatory guidelines without indicating how exactly it was defined (neither by the authors nor by the guideline) | 5 (7%) |
Based on the efficacy of the experimental drug from the previous clinical trials | The margin was defined based on the efficacy of the experimental drug itself from the previous trials | 5 (7%) |