Van Praag 1976.

Methods	Allocation: randomised. Blindness: double (implied). Duration: 10 days (preceded by 5 to 10 days wash‐out). Setting: inpatient, single‐centre.
Participants	Diagnosis: acute psychosis (n = 3 organic psychosis; n = 13 psychogenic psychosis; and n = 8 acute schizophrenia). N = 28*. Age: 19 ‐ 69 years, mean ˜ 35 years. Sex: 15 F, 9 M. History: acute, inpatient. Included: according to independent views of attending physician and psychiatrist/ supervisor, the participant showed signs of delusions and/ or hallucinations (Lowe 1973). Excluded: deep melancholia (i.e. deep endogenous depression with delusions of guilt, sin and/or poverty). Consent: not stated.
Interventions	1. Perphenazine: 2, 4 and 8 mg per capsule, 3 capsules daily according to constitution, range 0 to 24 mg daily, mean 21 mg daily, n = 15. 2. Clozapine: 50, 100 and 200 mg per capsule, 3 capsules daily according to constitution, range 0 to 600 mg daily, mean 300 mg daily, n = 13. All daily doses determined by a psychiatrist not engaged in scoring; flexible dosing scheme.
Outcomes	Leaving the study early. Adverse events. Unable to use ‐ Global state: no better or worse (no data). Mental state: BPRS (only significance tests). Laboratory tests (no data).
Notes	Only 24 participants were evaluated and described.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Randomised: no further details.
Allocation concealment (selection bias)	Unclear risk	No details.
Blinding (performance bias and detection bias) All outcomes	Low risk	Double‐blind: implied ‐ identical capsules. Quote, "raters and patients were unaware of the type of medication or its dosage; nor were they informed of the biochemical results" (p549).
Incomplete outcome data (attrition bias) All outcomes	Low risk	Four participants (n = 3 receiving perphenazine; n = 1 receiving clozapine) failed to complete the study. All cases were due to presence of severe motor unrest, anxiety or aggression non‐responsive to permitted medication. ITT used.
Selective reporting (reporting bias)	Unclear risk	No means or SD reported for scale data.
Other bias	Unclear risk	Funding: not stated. Raters: not stated to be independent of treatment.