Marjerrison 1964.

Methods	Allocation: randomly assigned, no further details. Blindness: double (multiplicity of forms, individual coded bottles). Design: 4 parallel groups, single centre. Duration: phase I, five‐months; phase II, two‐months. Analysis: intention‐to‐treat not performed in the trial. Country: Canada.
Participants	Diagnosis: **schizophrenia but 12 with other diagnoses distributed among the groups. N=88. Age: mean 48 years. Sex: M 38, F 40. Setting: hospital. History: chronic psychotics; highly treatment resistant.
Interventions	1. Trifluoperazine: dose phase I, 29 mg/day, phase II, 27 mg/day. N=16. 2. Chlorprothixene: dose phase I, 200 mg/day, phase II, 270 mg/day. N=8. 3. Placebo. N=34. 4. Usual Phenothiazines: dose variated. N=30 (drugs were chosen clinically).
Outcomes	Leaving the study early. Global state: (numbers of subjects receiving barbiturate medication). Use of antiparkinson agents. Unable to use ‐ Mental state: (PRP; Ward Inventory scores).
Notes	** Mixed diagnosis.
Risk of bias
Bias	Authors' judgement	Support for judgement
Allocation concealment?	Unclear risk	B ‐ Unclear