Kurland 1961.

Methods	Allocation: randomised. Blindness: double. Duration: 6 weeks, preceded by 48 hours wash‐out. Setting: hospital, single‐centre, Baltimore, Maryland (US).
Participants	Diagnosis: "predominantly schizophrenia" ‐ no further details. N = 238 (n = 209 relevant) Age: 18‐61 years, mean ˜ 39 years. Sex: ratio men to women was 1:2 ‐ no further details. History: inpatient, newly admitted. Included: between 18 to 65 years old; referred on basis of target symptoms of anxiety, agitation and restlessness. Excluded: alcoholism, admission under court order, chronic brain syndrome, major organic diseases, senility. Consent: not stated.
Interventions	1. Perphenazine: initially 5 mg three times daily parenterally for 2 days, then 8 or 16 mg or a multitude of 16 mg orally for 40 days, average 30.83 mg/day, n = 31. 2. Prochlorperazine: initially 5 mg three times daily parenterally for 2 days, then 10 or 25 mg or a multitude of 25 mg orally for 40 days, average 45.38 mg/day, n = 29. 3. Triflupromazine: initially 25 mg three times daily parenterally for 2 days, then 25 or 50 mg or a multitude of 50 mg orally for 40 days, average 110.46 mg/day, n = 31. 4. Mepazine: initially 25 mg three times daily parenterally for 2 days, then 25 or 50 mg or a multitude of 50 mg orally for 40 days, average 135.45 mg/day, n = 27. 5. Chlorpromazine: initially 25 mg three times daily parenterally for 2 days, then 100 or 200 mg or a multitude of 200 mg orally for 40 days, average 401.35 mg/day, n = 32. 6. Promazine: initially 50 mg three times daily parenterally for 2 days, then 100 or 200 mg or a multitude of 200 mg orally for 40 days, average 438.92 mg/day, n = 29. 7. Placebo: active (phenobarbital) and inert (saline/ lactose) initially saline three times daily parenterally for 2 days, then lactose orally for 40 days, average 6 capsules/day, n = 30. [8. Phenobarbital: initially 65 mg three times daily parenterally for 2 days, then 32.5 or 65.0 mg or a multitude of 65.0 mg orally for 40 days, average 183.64 mg/day, n = 29] ‐ not included in data and analysis (not an antipsychotic).
Outcomes	Leaving the study early: any reason; due to adverse events. Response: no better or worse. Unable to use ‐ Mental state: MSRPP (no SD), PRP (no data), Psychiatric Scale of Target Symptoms (unpublished scale). Adverse events (no data).
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Randomised: quote, "the choice of the particular drug to be used in any case was based on a predetermined random selection of the compounds" (p4).
Allocation concealment (selection bias)	Unclear risk	Quote, "the physician knew a patient had six chances in eight of being given a phenothiazine compound, but did not know which drug was being given, or whether the patient was receiving one of the placebos" (p4).
Blinding (performance bias and detection bias) All outcomes	Unclear risk	Double‐blind: identical capsules; quote, "drugs were dispensed in standard unmarked capsules... coloured pink to mask all identifying consistencies and colours of the drugs" (p4). Treatment information was “secured from the pharmacist and recorded in the patient’s chart by a member of the nursing staff” (p5). Quote, "all clinical and research personnel working with patients were bound to the double‐blind stipulation imposed by the research plan" (p6).
Incomplete outcome data (attrition bias) All outcomes	High risk	Quote, "those patients on whom data were incomplete, together with those not receiving medication for at least then days, were excluded from the study" (p6). Unclear which participants dropped‐out of which treatment arm. One publication makes reference to N = 277 included in the study and presents data for those. Whereas the subsequently‐published paper refers to N = 238 included after screening people eligible for study. ITT used.
Selective reporting (reporting bias)	Unclear risk	No SD or P values reported for scale data.
Other bias	Unclear risk	Funding: Research Grant No. MY‐2152 of the National Advisory Mental Health Council, National Institutes of Health, US Public Health Service and administered by Friends of Psychiatric Research Inc, Spring Grove State Hospital, Baltimore, Maryland (US). The study involved "preliminary correspondence with the drug companies"; further, the "ward physician could also drop a patient from the study if her felt he should later the plan of treatment because of the ineffectiveness of the drug" (p5). Raters: not stated to be independent of treatment.