Methods	Design: A randomised, double-blind, active-control, parallel group trial. The trial included 35 centres across the United States, of which the majority were primary care centres
Participants	Population: 255 adults with a clinical history of COPD were randomised to tiotropium + formoterol (124) and tiotropium (131) Baseline Characteristics: Mean age 64 years. COPD severity mild to severe. 67% men. Inclusion Criteria: Male and non-pregnant female patients aged >40 years WHO had a clinical history of COPD were enrolled in this study. Each patient had a post-bronchodilator FEV1 < 70% and >30% predicted normal or >0.75 L, whichever was less, at run-in, and a FEV1 to FVC ratio (FEV1/FVC) of < 0.70 at screening and run-in. Daytime and/or nighttime symptoms of COPD, including dyspnoea, must have been present on ≥4 of the 7 days before the baseline visit Exclusion Criteria: A current or previous history of asthma or other significant medical condition that may have interfered with study treatment as assessed by the investigator, smoking cessation within the previous 3 months, ventilator support for respiratory failure within the previous year, the use of oxygen (≥2 L/min or for >2 h/d), initiation of pulmonary rehabilitation within the previous 3 months, the requirement for nasal continuous positive airway pressure or bilevel positive airway pressure, clinically significant lung disease other than COPD (i.e., bronchiectasis, sarcoidosis, pulmonary fibrosis, tuberculosis), sleep apnoea, chronic narrow-angle glaucoma, symptomatic prostatic hyperplasia or bladder neck obstruction, and the need for chronic or prophylactic antibiotic therapy
Interventions	1. Formoterol (Foradil Aerolizer) 12 μg twice daily and tiotropium (Handihaler) 18 μg once-daily in the morning delivered via 2 separate inhalers 2. Formoterol-matched placebo twice-daily and tiotropium 18 μg once-daily delivered via 2 separate inhalers
Outcomes	Primary: The normalized area under the curve (AUC) for FEV1 measured 0 to 4 hours post-morning dose (FEV1 AUC 0-4 h) at the last visit. Secondary: Changes from baseline in trough (average of values obtained 10 and 30 min pre-dose) FEV1 and FVC, weekly morning and evening peak expiratory flow (PEF), symptom severity scores, transition dyspnoea index (TDI), and health related quality of life (St. George’s Respiratory Questionnaire, SGRQ) scores, number and severity of exacerbations, the global therapeutic response, discontinuations because of worsening COPD, and percentages of patients achieving targeted improvements in the SGRQ and TDI scores, use of rescue albuterol, nocturnal awakenings requiring rescue albuterol, changes in study or concomitant medications, and adverse events
Notes	Co-medication: Continued use of prior stable inhaled corticosteroid (27%) regimens and systemic corticosteroids for the treatment of exacerbations was permitted throughout the study. All patients were provided with albuterol for use as rescue medication Run-in: Following screening, prohibited medications (i.e., beta-agonists, beta-blockers, cromolyn sodium, ipratropium bromide, leukotriene antagonists, cytotoxic agent, and theophylline) were withdrawn. Patients previously using TIO or FORM discontinued the drugs at least 4 weeks or 48 hours before screening, respectively. Patients completed a 2-week run-in period using placebo and as-needed rescue albuterol
Risk of bias
Bias	Authors’ judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Patients were randomised sequentially as they qualified for the study according to a pre-generated computer code labelled on the medication kit
Allocation concealment (selection bias)	Unclear risk	A pre-generated computer code was labelled on the medication kit
Blinding of participants and personnel (LABA+TIO versus TIO) (performance bias)	Unclear risk	Not described
Blinding of outcome assessment (detection bias) Exacerbations	Unclear risk	Not described
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	The number of withdrawals in the different groups was relatively low but uneven (LABA + tiotropium (14.5%), and tiotropium + placebo (6.1%))
Selective reporting (reporting bias)	Low risk	Results for all listed primary and secondary outcomes were reported