Stegeman 1996.

Methods	Study design: parallel RCT Duration of study: September 1990 to June 1993 Duration of follow‐up: to December 1994
Participants	Country: Netherlands Setting: multicentre (11 sites) Inclusion criteria: 3 groups of patients Group 1: necrotising GN and upper or lower airways disease consistent with WG Group 2: biopsy‐proven WG limited to the airways Group 3: ANCA positive patients fulfilling American College of Rheumatologists criteria for WG but not for Groups 1 or 2 Number: treatment group (41); control group (40) Median age, range (years): treatment group (56, 21 to 82); control group (57, 25 to 83) Sex (M/F): treatment group (30/11); control group (28/12) Exclusion criteria: allergy or adverse reactions to co‐trimoxazole or one of its components; long term (> 6 weeks) antibiotic treatment; impaired kidney function (CrCl < 30 mL/min/24 hours)
Interventions	Treatment group TMP/SMX: 160/800 mg twice daily for 24 months Control group Placebo tablets for 24 months
Outcomes	Death Remission at 24 months Infections/patient/years
Notes	Compliance assessed by tablet counts Supported by a grant from the Dutch Kidney Foundation (89.0872) Roche Pharma Ltd., Reinach, Switzerland provided the TMP/SMX and matched placebo
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Stratified according to disease group. sequence generation not reported but all parties blinded
Allocation concealment (selection bias)	Low risk	Treatment assignment not known to investigators
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Treatment assignment not known to investigators, patients or physicians
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Insufficient information to permit judgement
Incomplete outcome data (attrition bias) All outcomes	Low risk	All patients accounted for
Selective reporting (reporting bias)	Low risk	Study and review outcomes reported
Other bias	Low risk	Funding source stated