Nakajima 1996.

Methods	Design: RCT parallel group, Blindness: double Duration of study: 8 wks
Participants	Single or multicentre: Multicentre Diagnostic criteria: DSM IIIR Co‐morbidity allowed: they did not assess for depression (thus not clear whether some patients were depressed) Setting: In patients and out patients Age: 18 to 69 years Other characteristics: Men and women
Interventions	No. of arms: Three, fluvoxamine 100‐150mg (ie low dose) vs fluvoxamine 200‐300mg (medium doese) vs placebo
Outcomes	Primary: Change in YBOCS total score Proportion of responders defined as improved or much improved
Notes	15
Risk of bias
Bias	Authors' judgement	Support for judgement
Allocation concealment?	Unclear risk	B ‐ Unclear