Baren 2001.

Methods	STUDY DESIGN: Parallel group LOCATION, NUMBER OF CENTRES: North America, single centre. DURATION OF STUDY: 8 weeks COMPLIANCE: Not assessed CONFOUNDERS: Even distribution between groups in terms of baseline lung function, age, sex and maintenance therapies.
Participants	N SCREENED: 197 N RANDOMISED: 192 N COMPLETED: 178 M = 64/F = 128 MEAN AGE: 31 BASELINE DETAILS: Ethnicity: Asian: 7; Black: 146; Hispanic: 3; White: 18; Insurance: Government/HMO: 40%; Government/military: 4%; HMO: 22%; Private: 22%; None: 13%. PEFR: 246 l/min; respiratory rate: 21.3; Inhaler use in previous 24 hrs (puffs): 4.7 INCLUSION CRITERIA: Aged between 16‐46 years; attendance at emergency department with symptoms of acute asthma EXCLUSION: Admission to hospital; unable to speak English; unwilling/unable to provide informed consent
Interventions	Education group: On discharge, participants were provided with a pack containing oral steroids, transportation vouchers to attend a primary care follow up; asthma information card; written instructions on use of vouchers and medication. Attempts made to contact all intervention group participants to remind them to attend a primary care follow up. Control group: Participants discharged with short course of oral steroids; further instructions and medication at discretion of discharging physician FOLLOW‐UP PERIOD: Participants were followed up for two months.
Outcomes	Scheduled attendance at primary care physician/clinic; relapse (re‐presentation at ED within 21 days of discharge); withdrawal/loss to follow‐up.
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Computer‐generated block randomisation schedule
Allocation concealment (selection bias)	Low risk	Prepared by third party. 'Study packages were prepared and sealed by 2 investigators not involved in patient enrolment.'
Blinding (performance bias and detection bias) All outcomes	Unclear risk	Study participants aware of treatment group assignment. Information on study outcome assessor blinding not available.
Incomplete outcome data (attrition bias) All outcomes	High risk	Differential loss to follow‐up. 11/93 in control group withdrew versus 3/94 in intervention group.
Selective reporting (reporting bias)	Unclear risk	Could not determine this reliably
Other bias	Low risk	No other sources of bias could be identified.