Schuh 2006.

Methods	Design: randomised controlled trial
Participants	Children with mild to moderate asthma Randomised: fluticasone 35, placebo 34 Age (mean (SD)) (years): fluticasone 9.0 (2.6), placebo 9.2 (3.4) Sex (M/F): fluticasone 25/10, placebo 20/14 Baseline lung function; mean % predicted FEV1 (SD): fluticasone 63.0 (10.8), placebo 61.5 (10.7) Inclusion criteria: children aged 5 to 17 years, diagnosed with acute asthma by the ED paediatrician, with FEV1 50% to 79% predicted on ED presentation Exclusion criteria: children presenting with a first episode of wheezing, persistent vomiting, airway instability, those who had received oral corticosteroids within 7 days, co‐existent cardiopulmonary/neuromuscular disease, varicella contact within 21 days of the study entry, previous intensive care unit treatment for asthma, inability to communicate in English
Interventions	Intervention: fluticasone propionate 2 mg (8 inhalations, 250 μg each) via MDI/VHC with a mouthpiece and prednisolone placebo syrup in the ED Control: 8 inhalations of fluticasone placebo and 2 mg/kg (maximum 60 mg) of active prednisolone syrup Co‐interventions: all children received nebulised albuterol (dose of 0.15 mg/kg in first 10 patients, then decreased to 0.075 mg/kg because of tremor and tachycardia noted in the first 10 patients attributed to the high‐efficiency nebuliser) and ipratropium bromide, 250 μg per dose, with a mouthpiece 20 min before the experimental therapy, immediately after the initial dose of the experimental therapy (0 min), and at 60, 120, 180 and 240 min
Outcomes	Change in % predicted FEV1 from baseline to 240 min and 48 h in the 2 groups, RR and oxygen saturation, hospital admissions, adverse events
Notes	ICS versus systemic corticosteroid subgroup Trial added in 2012 update
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Computer‐generated randomisation scheme, random block sizes of 10
Allocation concealment (selection bias)	Low risk	The research pharmacist supplied the corresponding MDI/syrup in a double‐dummy setup to the study nurses who enrolled participants and delivered the experimental therapy. Randomisation codes were secured in the pharmacy until enrolment and analysis decisions had been completed. The study nurses were unable to determine which intervention a given patient had received
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Double‐blind, double‐dummy
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Double‐blind, double‐dummy
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	60 children performed spirometry at 240 min, 30 in the intervention group and 30 in the control group (of 35 and 34 enrolled) 1 parent changed her mind regarding participation at 120 min, 5 children vomited the syrup and were too nauseated for FEV1 assessment, and 3 children were too sleepy to cooperate with FEV1 assessment
Selective reporting (reporting bias)	Unclear risk	No apparent indication of selective reporting bias