Table 2– Results for least squares mean change from baseline trough forced expiratory volume in 1 s on day 29.

Treatment	Subjects n	Change from baseline mL	Difference from placebo mL	p-value versus placebo	Difference from salmeterol mL#	p-value versus salmeterol#
Placebo	71	-7
Salmeterol	43	71	77 (1–153)	0.046#
GSK961081
Twice daily
100 μg	47	167	173 (100–247)	<0.001	96 (14–179)	0.023
200 μg	46	243	249 (175–323)	<0.001	172 (89–255)	<0.001
400 μg	49	251	258 (185–330)	<0.001	181 (98–263)	<0.001
Once daily
100 μg	45	148	155 (80–229)	<0.001	78 (-7–162)	0.071
400 μg	41	209	215 (139–291)	<0.001	138 (53–223)	0.002
800 μg	48	270	277 (204–350)	<0.001	200 (117–282)	<0.001

Data are presented as least squares mean or least squares mean (95% CI), unless otherwise stated. For the primary end-point, p-values for GSK961081 doses were compared to placebo at α=0.025 due to separate closed step-down procedures. Least squares means were adjusted for age, sex, smoking status, reversibility stratum, overnight site stratum, concurrent inhaled corticosteroid use, baseline and treatment. ^#: inferences involving salmeterol were post hoc analyses.