“…With 530 evaluable patients at week 12 in each group, the study will have 80% power to detect a mean difference of 50 mL in change in predose morning FEV1, assuming a standard deviation of 290 mL, using an unpaired t-test with two-sided significance level of 0.05. The difference of 50 mL was chosen based on previous long-term clinical trials showing that ICS/LABA combinations show this level of superiority over LABA alone, and this is associated with differences in health status and exacerbation rates [14,15,16,21].…”