OBJECTIVES: To assess non-inferiority (NI) of aclidinium bromide 400 µg BID (AB400), to tiotropium bromide 18 µg (TIO18) and 5 µg (TIO5) QD in patients with COPD. METHODS: A systematic literature search identified 21 RCTs: TIO18 (16 trials), TIO5 (3 trials) and AB400 (2 trials). All trials were analysed using a Bayesian network meta-analysis (NMA) and relative treatment effects between all regimens were obtained for change from baseline (CFB) in FEV1 trough, SGRQ total score, and TDI focal score, at 24 weeks. The posterior distributions of the relative treatment effects were used to estimate the probability of non-inferiority, based on the NI margins.  Two approaches used for margin selection (a) NIMCID: 50% of the Minimum Clinically Important Difference (MCID) (b) NINMA: 50% of the difference between TIO18 and placebo (PLA), based on the NMA results. AB400 was considered non-inferior if it could maintain at least 50% of this difference with a probability >99%. RESULTS: For FEV1 the MCID is 100mL, while the CFB for TIO18 vs. PLA is 104.1mL giving NIMCID=50mL, NINMA=52.05mL. AB400 is shown to be non-inferior to TIO5 and TIO18, for both margins. For SGRQ, a change of 4 points is the MCID and CFB for TIO18 vs. PLA is 2.65, thus NIMCID = 2 and NINMA= 1.325. AB400 was shown to be non-inferior to TIO5 and TIO18 for both margins. A difference of 1 unit is the MCID for TDI, while the CFB for TIO18 vs. PLA is 0.90 unit, giving NIMCID = 0.5, NINMA = 0.45. For these margins, the probability of aclidinium been non-inferior is 97% for the NIMCID, and 96% for the NINMA. CONCLUSIONS: This analysis suggests that treatment with AB400 in COPD is non-inferior to TIO18 and TIO5 with respect to lung function, health status, and to breathlessness with a probability >95%.