Gelfand 1976.

Study characteristics
Methods	Design: randomised, double‐blind, single‐centre, cross‐over trial Exclusions postrandomisation: none reported Losses to follow‐up: 2 participants did not complete their series of treatment courses because the main study was terminated Duration of study: 6–11 months Unit of randomisation: participant
Participants	Country: US Setting: outpatient Number: 5 women (aged 25–38 years) and 4 men (aged 28–63 years) with HAE were selected. Each had an attack frequency ≥ 1 per month. Diagnosis established on basis of characteristic clinical history, low serum C4 and low C1‐INH activity Age (mean): 34.9 (SD 11.4) years Sex: 4 men (44.4%); 5 women (55.6%) Inclusion criteria: attack frequency ≥ 1 per month; diagnosis of HAE. Exclusion criteria: not stated.
Interventions	Danazol capsules 200 mg 3 times/day Placebo capsules 3 times/day 9 people were randomised each 28‐day period to receive either danazol or placebo capsules for 28 days, for a total of 93 courses (46 courses in which participants were taking danazol, 47 courses where the participants were taking placebo).
Outcomes	Rate of HAE attacks, functional C1‐INH, C4 concentrations
Funding	Funded and carried out by NIH.
Declarations of interest	No conflicting interest information provided.
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Randomisation generation method not stated.
Allocation concealment (selection bias)	Unclear risk	Unclear whether allocation concealment took place.
Blinding of participants and personnel (performance bias)	Low risk	Participants and personnel were blinded to allocation.
Blinding of outcome assessment (detection bias)	Unclear risk	Unclear if outcome assessors were blinded.
Incomplete outcome data (attrition bias) All outcomes	Low risk	No evidence of attrition.
Selective reporting (reporting bias)	Unclear risk	Protocol was not published.
Other bias	Low risk	We identified no other sources of bias.