Table 6.

Change from baseline in weighted mean FEV ₁ (L) difference in treatment effect compared between 12–24-hour and 0–12-hour at Day 7 (mITT population)

	Placebo	UMEC QD
		15.6 μg	31.25 μg	62.5 μg	125 μg
	N = 60	N = 60	N = 57	N = 59	N = 60
N	54	58	53	56	56
Column vs. placebo	N/A	0.018	−0.007	0.021	0.010
Absolute difference
95% CI	N/A	(−0.040, 0.076)	(−0.067, 0.053)	(−0.038, 0.080)	(−0.048, 0.069)
Ratio*	N/A	1.176	0.943	1.162	1.059
	UMEC BID				Tiotropium QD
	15.6 μg		31.25 μg		18 μg
	N = 56		N = 58		N = 56
N	52		55		53
Column vs. placebo	0.019		0.028		−0.036
Absolute difference
95% CI	(−0.041, 0.078)		(−0.030, 0.087)		(−0.095, 0.023)
Ratio*	1.147		1.215		0.793

*Relative to 0–12-hour weighted mean FEV₁.

Absolute differences >0 indicate a larger treatment effect between 12–24 hours; analysis performed using a mixed model with covariates trough mean baseline, trough period baseline, treatment, period, time and time by treatment interaction as fixed effects and subject as a random effect.

Note: The column versus placebo difference was calculated as the difference in change from baseline in weighted mean FEV₁ between active and placebo at 12–24 hours minus the difference in change from baseline in weighted mean FEV₁ between active and placebo at 0–12 hours. The ratio is the difference in change from baseline in weighted mean FEV₁ between active and placebo at 12–24 hours divided by the difference in change from baseline in weighted mean FEV₁ between active and placebo at 0–12 hours.

BID, twice daily; CI, confidence interval; N/A, not applicable; QD, once daily; UMEC, umeclidinium bromide.