Methods	Allocation: random, no further details. Blindness: double, no further details. Duration: 52 weeks. Design: parallel. Location: multicentre. Countries: USA, Canada.
Participants	Diagnosis: (DSM-IV) schizophrenia or schizoaffective disorder. N=414. Age: 18-55 years (mean=39 years). Sex: 282 M, 132 F. History: duration ill not reported, age at onset not reported. Setting: in- and outpatient.
Interventions	Haloperidol: flexible dose. Allowed dose range: 2-19 mg/day. Mean dose: 8.2 mg/day. N=97. Olanzapine: flexible dose. Allowed dose range: 5-20 mg/day. Mean dose: 12.3 mg/day. N=159. Risperidone: flexible dose. Allowed dose range: 2-10 mg/day. Mean dose: 5.2 mg/day. N=158
Outcomes	Leaving the study early: any reason, adverse events, inefficacy. Global State: relapse. Mental State: PANSS total score, PANSS positive subscore, PANSS negative subscore, depression (MADRS), anxiety (Hamilton anxiety scale). Cognitive Functioning: Neurocognitive Composite Score. Adverse effects: open interviews, EPS (akathisia, tremor, use of antiparkinson medication, AIMS, BAS, SAS), sedation, weight change, laboratory (cholesterol, prolactin, urine analysis)
Notes
Risk of bias
Bias	Authors’ judgement	Support for judgement
Adequate sequence generation?	Unclear risk	Random, no further details.
Allocation concealment?	Unclear risk	No further details.
Blinding? Objective outcomes	Low risk	Objective outcomes such as laboratory measures or death are unlikely to have been much affected by problems of blinding
Blinding? Subjective outcomes	Unclear risk	Double, no further details. Whether blinding was successful has not been examined, but the compounds differ quite substantially in side-effects. This can be a problem for blinding
Incomplete outcome data addressed? All outcomes	High risk	The attrition rate was high (62.8%). The last-observation-carried-forward method was used to account for people leaving the study early. It assumes that a participant who discontinued the study would not have had a change of his condition if he had remained in the study. This assumption can obviously be wrong
Free of selective reporting?	High risk	Only those adverse events that occurred in at least 10% of the participants were reported. This procedure can miss rare, but important adverse events
Free of other bias?	High risk	The study was sponsored by the manufacturer of olanzapine.