Bonduelle 1991.

Methods	Single centre, parallel‐group design with no blinding Number of women randomized: n = 30 (15 in each treatment group) Withdrawals post‐randomization: n = 6 (2, selection criteria not met; 4, did not return), 5 from danazol group and 1 from NET group Loss to follow‐up: n = 8 (during treatment), 4 from each group, all because of adverse events
Participants	Patients (with mean ages 39 and 32) recruited from the Menstrual Disorders Clinic at the Royal Infirmary, Glasgow, UK Inclusion criteria: complaint of MBL requiring > 5 pads/tampons/day for > 6 days in cycle; presence of flooding or clots on any day of the cycle; presence of secondary anaemia; excessive MBL proving socially and domestically disruptive. Exclusion criteria: underlying pathology (from history, examination and D&C within the last year)
Interventions	NET 5 mg 3 times/day from day 19‐day 26 Danazol 200 mg, daily Duration: 3 cycles (1 pre‐treatment)
Outcomes	Duration of menstruation (days) Prevalence of side effects
Notes	Considered by authors to be a "pilot study". Poor‐quality trial because of high proportion of participants excluded or lost to follow‐up No source of funding given
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Randomization technique not stated
Allocation concealment (selection bias)	Unclear risk	Not stated
Blinding of participants and personnel (performance bias) PBAC, satisfaction, Adverse effect, response to treatment, bleeding	High risk	Open trial, with different dosage schedule. All outcomes likely to be influenced by lack of blinding
Blinding of participants and personnel (performance bias) Haematin alkaline	Unclear risk	Alkaline haematin was not measured.
Blinding of outcome assessment (detection bias) PBAC, satisfaction, Adverse effect, response to treatment	High risk	Open trial, all outcomes likely to be influenced by lack of blinding
Blinding of outcome assessment (detection bias) Haematin Alkaline	Unclear risk	Alkaline haematin was not measured.
Incomplete outcome data (attrition bias) All outcomes	High risk	30 women randomized, 6 lost post randomization and 8 during treatment
Selective reporting (reporting bias)	Unclear risk	No protocol available
Other bias	Low risk	Similar at baseline: yes Disclaimer: not stated CoI: not stated Source of funding: not given