Korn 2011.

Methods	Design: double‐blind, parallel, randomised, controlled trial. 12 weeks' duration. Additional inhaled bronchodilators other than albuterol were discontinued. Intention‐to‐treat analysis was performed. Participants were recruited from the USA, Europe and India Run‐in: 2 weeks
Participants	Population: 1123 participants with a diagnosis of moderate to severe COPD were randomly assigned. Mean age was 62.8 years. Mean FEV1 was 51.8% predicted. In the indacaterol and salmeterol arms, 45.8% and 46.1% of participants were taking concomitant inhaled corticosteroids, respectively Inclusion criteria: Adults > 40 years with at least a 10‐pack‐year smoking history FEV1/FVC < 0.7, FEV1 30% to 80% predicted post bronchodilator Exclusion criteria: respiratory tract infection or COPD exacerbation during previous 6 weeks, diagnosis of asthma, concomitant pulmonary disease, long‐term oxygen therapy, type 1 or uncontrolled type 2 diabetes, cancer with less than 5‐year survival, lung cancer, QTc abnormalities, live vaccine in previous 30 days
Interventions	1. Indacaterol 150 mcg 2. Salmeterol 50 mcg
Outcomes	Primary endpoint:‐ FEV1 standardised area under the curve (AUC) from 5 minutes to 11 hours 45 minutes at week 12 Secondary endpoints: 24‐hour trough FEV1 at week 12, FEV1 and FVC measured over 24 hours, Transitional Dyspnoea Index (TDI) and rescue medication use Follow‐up on days 1, 2, 28, 29, 84, 85 Values were reported at baseline; additional information was obtained from study authors for post‐treatment data
Notes	Study was supported by Novartis, and some study authors were Novartis employees. Novartis directly supplied data for: trough FEV1, serious adverse events and mortality.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Randomisation was performed using an automated interactive voice response system
Allocation concealment (selection bias)	Low risk	Allocation was performed via automated interactive voice response system with participants and assessors blinded to allocation; double‐dummy design
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Allocated interventions were not known by participants or by personnel during the study
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Outcome assessors were blinded to treatment allocation
Incomplete outcome data (attrition bias) All outcomes	Low risk	Rates and reasons for dropout were similar across all arms
Selective reporting (reporting bias)	Low risk	All prespecified outcomes were reported