Skip to main content

Table 2 Results from the analyses of the primary, secondary and selected other endpoints (ITT population)

From: Umeclidinium/vilanterol versus fluticasone propionate/salmeterol in COPD: a randomised trial

Endpoint UMEC/VI 62.5/25 mcg (N = 358) FP/SAL 500/50 mcg (N = 358)
Primary endpoint
wm 0–24 h FEV1 on Day 84, L
n 332 337
LS mean (SE) 1.618 (0.0122) 1.539 (0.0121)
LS mean (SE) change from baseline 0.166 (0.0122) 0.087 (0.0121)
Treatment difference (95 % CI) 0.080 (0.046–0.113)
p < 0.001
Secondary endpoint
Trough FEV1 on Day 85, L
n 344 353
n a 333 338
LS mean (SE) 1.600 (0.0126) 1.511 (0.0125)
LS mean (SE) change from baseline 0.151 (0.0126) 0.062 (0.0125)
Treatment difference (95 % CI) 0.090 (0.055–0.125)
p < 0.001
Other endpoints (selected)
Peak FEV1 0–6 h
Day 1
n 358 358
n a 358 358
LS mean (SE) 1.712 (0.0083) 1.678 (0.0083)
LS mean (SE) change from baseline 0.266 (0.0083) 0.231 (0.0083)
Treatment difference (95 % CI) 0.034 (0.011–0.057)
p = 0.003
Day 84
n 358 358
n a 335 340
LS mean (SE) 1.773 (0.0131) 1.676 (0.0130)
LS mean (SE) change from baseline 0.327 (0.0131) 0.229 (0.0130)
Treatment difference (95 % CI) 0.097 (0.061–0.134)
p < 0.001
Time to onset on Day 1 (increase in FEV1 ≥ 0.100 L above baseline)
n 358 358
Median time to onset, min 17 60
Hazard ratio (95 % CI) 1.3 (1.1–1.5)
p = 0.002
  1. Analysis of the primary endpoint was performed using ANCOVA with covariates of baseline FEV1, smoking status and treatment. Analysis of the secondary endpoint was by MMRM analysis including covariates of baseline FEV1, smoking status, day, treatment, day by baseline interaction and day by treatment interaction, where day is nominal
  2. CI confidence interval, FEV 1 forced expiratory volume in 1 s, FP/SAL fluticasone propionate/salmeterol, ITT intent-to-treat, LS least squares, MRMM mixed-effect model repeated measure model, SE, standard error, UMEC umeclidinium, VI vilanterol, wm weighted mean
  3. aNumber of patients with analysable data at the current time point