FIGURE 2 Plot of least-square means (LSM)±se of change from baseline of trough forced expiratory volume in 1 s (FEV1) with mixed-effect model repeated measure (MMRM) full analysis set. R: 50 μg glycopyrronium (GB) once dailly + fixed-dose combination (FDC) 12 μg formoterol fumarate (FF)/250 μg fluticasone propionate (FP) twice daily; T: FDC 12.5 μg GB/12 μg FF/250 μg FP twice daily. p-values of NI for T versus R at weeks 2, 4, 8, and 12 were <0.05, <0.05, 0.08 and <0.05, respectively. p-values of SP for T versus R at weeks 2, 4, 8, and 12 were 0.10, 0.09, 0.07 and 0.13, respectively. p-values of change from baseline for T at weeks 2, 4, 8, and 12 were <0.001, 0.003, 0.008 and <0.001, respectively. p-values of change from baseline for R at weeks 2, 4, 8, and 12 were <0.001, <0.001, <0.001 and <0.001, respectively. LSM of change from baseline of trough FEV1 (L) was calculated with MMRM method with treatment, centre, visit, and treatment-by-visit interaction as fixed-effect factors and lung function FEV1 (L) at baseline as covariate. NI: noninferiority; SP: superiority.