Methods	Allocation: two step randomisation before and after the availability of ziprasidone. Blinding: double, identical capsules. Duration: 26 weeks. Design: parallel. Location: not reported.
Participants	Diagnosis: (DSM-IV) chronic schizophrenia. N=444. Age: 18-65 years (mean olanzapine=40.0 years, mean quetiapine=40.1 years, mean risperidone=41.8 years, mean ziprasidone=41.3 years). Gender: 308 M, 136 F. History: duration of illness not reported, age at onset not reported. Setting: in- and outpatient.
Interventions	Olanzapine: flexible dose, allowed dose range: 7.5-30 mg/day, mean dose=20.5 mg/day. N=108. Quetiapine: flexible dose, allowed dose range: 200-800 mg/day, mean dose=565.2 mg/day. N=95. Risperidone: flexible dose, allowed dose range: 1.5-6.0 mg/day, mean dose=4.1 mg/day. N=104. Ziprasidone: flexible dose, allowed dose range: 40-160 mg/day, mean dose=115.9 mg/day. N=137
Outcomes	Leaving the study early: any reason, adverse events, inefficacy. Global State: CGI. Mental State: PANSS total score. Death: suicide. Adverse effects: open interviews, EPS (akathisia), cardiac effects (ECG), prolactin-associated side-effects, weight gain, laboratory (prolactin, glucose, cholesterol)
Notes
Risk of bias
Bias	Authors’ judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Random, 2 step randomisation before and after the availability of ziprasidone, Participants were re-randomised to a different medication than they had received in previous phase 1
Allocation concealment (selection bias)	Unclear risk	No further details.
Blinding (performance bias and detection bias) Subjective outcomes	Unclear risk	Double, identical capsules. Whether blinding was successful has not been examined, but both compounds differ quite substantially in side-effects. This can be a problem for blinding
Blinding (performance bias and detection bias) Objective outcomes	Low risk	Objective outcomes such as laboratory measures or death are unlikely to have been much affected by problems of blinding
Incomplete outcome data (attrition bias) All outcomes	High risk	The attrition rate was extremely high (72. 5%). Efficacy data analysis was based on mixed models, but it is unclear whether any statistical method can account for such high dropout rates
Selective reporting (reporting bias)	High risk	Use of antiparkinson medication was permitted but data on this was not available
Other bias	Unclear risk	Patients had a history of intolerance to atypical antipsychotic treatment but baseline data on this were not provided