Figure 2
Adjusted treatment differences for the change from baseline for primary, and selected secondary and other endpoints (intent-to-treat population). *Data were analysed using a closed step-down statistical hierarchy, whereby failure to achieve significance at any point in the hierarchy meant that statistical significance could not be inferred for subsequent endpoints in the hierarchy. Solid lines indicate a significant treatment difference; dashed lines indicate differences which are not significant (evening PEF) or for which significance cannot be inferred (all remaining endpoints); FEV1 = forced expiratory volume in one second; PEF = peak expiratory flow; ACT = Asthma Control Test; AQLQ = Asthma Quality of Life Questionnaire; CI = confidence interval.