The primary objective of the study was to demonstrate the non-inferiority of indacaterol vs. tiotropium with respect to 24 hour post dose (trough) FEV1 after 12 weeks of treatment in patients with severe COPD. Trough FEV1 was defined as the average of the 23 hours 10 min and the 23 hours 45 min post dose values. Trough FEV1 was analyzed using a mixed model for the PPS-S. The model contained treatment as a fixed effect with the baseline FEV1, FEV1 prior to inhalation and FEV1 15 min post-inhalation of salbutamol/albuterol (components of SABA reversibility at Visit 2), FEV1 prior to inhalation and FEV1 60 min post-inhalation of ipratropium (components of anti-cholinergic reversibility at Visit 3) as covariates. Smoking history (current or ex-smoker) was included as a factor in the model.

Per-Protocol Set for Spirometry (PPS-S) This set included Full Analysis Set (FAS) patients without any major protocol deviations or non-protocol deviation criteria that could affect the respective analysis of efficacy.

COPD exacerbations were defined as :Worsening of 2 or more major symptoms for at least 2 consecutive days: dyspnea; sputum volume; suputum purulence AND requiring treatment with systemic corticosteroids and/or antibiotics OR Worsening of any 1 major symptom together with any 1 of the following minor symptoms for at least 2 consecutive days: Sore throat; colds; fever without other cause; increased cough; increase wheeze AND requiring treatment with systemic glucocorticosteroids and/or antibiotics. The rate was analyzed using a linear model assuming a negative binomial distribution for the PPS-E. The time at risk for a patient was defined as the length of time the patient was in the study and the log(length of time in the study) was used as the offset variable in the model.

Per-Protocol Set for Exacerbations (PPS-E). This set included Full Analysis Set (FAS) patients without any major protocol deviations or non-protocol deviation criteria that could affect the respective analysis of efficacy.