Dorevitch 1999.

Methods	Allocation: randomised. Blindness: double. Duration: 2 hours.
Participants	Diagnosis: DSM‐IV diagnosis of paranoid schizophrenia (N = 19), schizoaffective disorder (N = 7), bipolar disorder (N = 2). N = 28. Age: range 20‐60 years. Sex: 13 males, 15 females. History: "actively psychotic inpatients". (p.142). Excluded: Not reported. Setting: Psychiatric hospital, Tel Aviv.
Interventions	1. Haloperidol: dose 5 mg/IM (mean number of IMs not reported). N = 13. 2. Flunitrazepam: dose 1 mg/IM (mean number of IMs not reported). N = 15.
Outcomes	Global state: need for seclusion or restraint. Aggression: OAS. Adverse effects.
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	"assigned by a table of random numbers." (p.142).
Allocation concealment (selection bias)	Unclear risk	Not stated.
Blinding of participants and personnel (performance bias) All outcomes	Unclear risk	Described as double‐blind, however further details regarding the blinding of participants and personnel are not reported.
Blinding of outcome assessment (detection bias) All outcomes	Low risk	"All ratings were completed by the same rater (N.K.), who was blind to the study medications." (p.143).
Incomplete outcome data (attrition bias) All outcomes	Low risk	No evidence of incomplete outcome data.
Selective reporting (reporting bias)	Unclear risk	It is not clear whether all side effects are reported.
Other bias	Unclear risk	Small short study.