Investigate the Impact of Early Treatment Initiation With Tiotropium in Patients Recovering From Hospitalization for an Acute COPD Exacerbation 2

Change from baseline of trough Forced Vital Capacity (FVC) at 12 weeks on study drug.

Percentage (number) of patients with adverse clinical event on study, which is defined as the combined endpoint of Chronic obstructive pulmonary disease (COPD) exacerbations per Boehringer Ingelheim (BI) definition, all-cause re-hospitalization, or all-cause mortality.

Change from baseline of trough FEV1 (forced expiratory volume in 1 second) at 12 weeks on study drug.

Trough FEV1 is defined as the FEV1 measurement prior to the next dosing of study drug and approximately 24 hours after last inhalation of drug.

Change from baseline of trough Forced Vital Capacity (FVC) at 12 weeks on study drug.

Percentage (number) of patients with COPD exacerbation on study was analysed for the combined study.

A COPD exacerbation was defined as a complex of lower respiratory events/symptoms (increase or new onset) related to the underlying COPD with duration of three days or more, requiring a change in treatment where a complex of lower respiratory events/symptoms was defined as at least two of the following:

1. Shortness of breath; 2) Sputum production (volume) ; 3) Occurrence of purulent sputum; 4) Cough; 5) Wheezing; 6) Chest tightness.

Onset of exacerbation was defined by the onset of first recorded symptom.The end of exacerbation was decided by the investigator based on clinical judgment.

A required change in treatment included either prescription of antibiotics and/or systemic steroids; and/or a newly prescribed maintenance respiratory medication (i.e., bronchodilators including theophyllines and PDE4-inhibitors).

Percentage (number) of patients with all-cause hospitalization outcome event occured during the study was analysed for the combined study.

All-cause hospitalization included all hospitalizations, except planned hospitalizations for elective procedures.

Hospitalizations occurring on the same day as discharge were not considered a separate admission.

Percentage (number) of patients with 30-day hospital readmission rates outcome events was analysed.

Days to hospital readmission were calculated as:Hospital readmission days = Readmission date - Date of hospital discharge + 1.

The 30-day hospital readmission analysis summarized the frequency of patients with hospital readmission and readmission days >1 and <31 days using the TS.

Number of COPD exacerbation per patient year outcome event occured during the study was analysed descriptively by calculating average occurrence (number of events per patient year drug exposure) by treatment group for on study period using the TS.

Total patient year exposure of COPD was calculated by aggregating the time to min(treatment stop +30, last contact) for on-treatment analysis, or time to last contact for on-study analysis.

Number of all-cause hospitalization per patient year outcome event occured during the study was analysed descriptively by calculating average occurrence (number of events per patient year drug exposure) by treatment group for on study period using the TS.

Total patient year exposure of all-cause hospitalization was calculated by aggregating the time to min(treatment stop +30, last contact) for on-treatment analysis, or time to last contact for on-study analysis.

Time to event: Time to recovery based on EXACT-PRO total score. The percentage of observed patients recovered by end of study was reported.

Time to recovery was assessed with the EXACT-PRO questionnaire. EXACT-PRO total scores were transformed to smooth scores for determining time to recovery and all other endpoints related to the EXACT questionnaire. The day-2 score was transformed to the mean of the total scores recorded on Day 1, 2, and 3. Similarly, each subsequent day's score was transformed to the mean score using a rolling 3-day average.

Analysis based on Kaplan Meier estimate