Backman 2001.

Methods	Setting: multicentre study (nine countries) Design: parallel group trial Length of intervention period: 4 weeks Randomisation: PACT Allocation concealment: adequate Masking: Open label Excluded: Yes (post‐run in) Withdrawals: Stated Jadad score: 3
Participants	277 randomised (FP: 137; BUD: 140). Mean age (range): FP: 46.3 (19‐81); BUD: 44.3 (18‐87); Baseline FEV1 (L): FP: 2.44 (SD 0.77); BUD: 2.46 (SD 0.76); FEV1 % predicted: FP: 75 (SD 12); BUD: 75 (SD 15); +ve atopic (%): FP: 61; BUD: 62; Duration of asthma (N): <1‐5 years: FP: 24; BUD: 36; >5‐10 years: FP 28; BUD: 23; >10 years: FP: 85; BUD: 81 Inclusion criteria: >18 years of age; documented history of reversible airways obstruction; FEV1 (% predicted) between 50‐90%; stable phase of disease; requirement for between 400 and 1200mcg/day of BDP, flunisolide or BUD. Exclusion criteria: Use of Diskus or Tubruhaler in previous 6 months; no treatment for airways disease in previous 4 weeks.
Interventions	FP 500 (via Diskus) versus BUD1200 (via Turbuhaler). 2‐week baseline period where PEF (am and pm); symptoms and rescue medication usage were measured
Outcomes	FEV1; am PEF; pm PEF; serum cortisol; adverse events; device handling
Notes	Details concerning randomisation method provided by Glaxo Wellcome
Risk of bias
Bias	Authors' judgement	Support for judgement
Adequate sequence generation?	Low risk	See Appendix 2
Allocation concealment?	Low risk	See Appendix 2
Blinding? All outcomes	High risk	Open label design