Background: The interpretation of the results of active-control trials regarding the efficacy and safety of a new drug is important for drug registration and following clinical use. It has been suggested that non-inferiority and equivalence studies are not reported with the same quantitative rigor as superiority studies.
Methods: Standard methodological criteria for non-inferiority and equivalence trials including design, analysis and interpretation issues were applied to 18 recently conducted large non-inferiority (15) and equivalence (3) randomized trials in the field of AIDS antiretroviral therapy. We used the continuity-corrected non-inferiority chi-square to test 95% confidence interval treatment difference against the predefined non-inferiority margin.
Results: The pre-specified non-inferiority margin ranged from 10% to 15%. Only 4 studies provided justification for their choice. 39% of the studies (7/18) reported only intent-to-treat (ITT) analysis for the primary endpoint. When on-treatment (OT) and ITT statistical analyses were provided, ITT was favoured over OT for results interpretation for all but one study, inappropriately in this statistical context. All but two of the studies concluded there was "similar" efficacy of the experimental group. However, 9/18 had inconclusive results for non-inferiority.
Conclusion: Conclusions about non-inferiority should be drawn on the basis of the confidence interval analysis of an appropriate primary endpoint, using the predefined criteria for non-inferiority, in both OT and ITT, in compliance with the non-inferiority and equivalence CONSORT statement. We suggest that the use of the non-inferiority chi-square test may provide additional useful information.