Brambilla 1994.

Methods	Setting: multicentre study France, hospital outpatient clinic Length of intervention: 12 weeks Design: parallel group, dose down‐titration Randomisation: yes, method not stated Allocation concealment: unclear Masking: none Excluded: not stated Withdrawals: stated Baseline characteristics: comparable Jadad score: 2
Participants	146 adults Inclusion criteria: Recurrent acute exacerbations of breathlessness and wheezing 15% or greater variability in FEV1 Bronchial responsiveness to methacholine or histamine (PD20 FEV1 2mg or less) Receiving inhaled BDP 1000‐2000 mcg/d for at least 6 months Exclusion criteria: Airways disease other than asthma Smoker > 5 packs cigarettes per year Significant co‐existent disease
Interventions	BDP: 800 to 2000 mcg total daily dose via MDI BUD: 800 to 1600 mcg total daily dose via Turbohaler DPI
Outcomes	FEV1 Morning PEFR Evening PEFR Number of puffs of inhaled corticosteroid per day Number asthma exacerbations/week Number night awakenings requiring beta2 agonist use/week Daily beta2 agonist use (puffs/day) Withdrawal due to asthma exacerbation Incidence of local oral side effects
Notes	Reply from author but unable to clarify details of randomisation procedure. Dose down titration study design: patients were randomised to receive either BDP or BUD at a dose considered to be equivalent to their pre‐study dose of BDP. BDP 250 mcg/puff via MDI was considered equivalent to BUD 200 mcg/actuation via Turbuhaler DPI. During the study daily dose of ICS was reduced according to pre‐defined criteria based on experience of asthma symptoms in the week prior to assessment.
Risk of bias
Bias	Authors' judgement	Support for judgement
Allocation concealment (selection bias)	Unclear risk	B ‐ Unclear