Kurland 1961.

Methods	Allocation: assigned by "predetermined random selection". Blinding: double, "standard unmarked capsules", "coloured pink to mask all identifying consistencies and colours of drugs". Duration: 6 weeks. Design: parallel. Country: USA.
Participants	Diagnosis: 'predominantly schizophrenic in character'. History: newly admitted, target symptoms of hyperactivity, anxiety, tension, overt aggression. N=277. Sex: 1:2 ratio M:F. Age: mean 31 yrs, range 18‐61. Setting: hospital.
Interventions	1. Chlorpromazine: dose 25 mg/day IM (days 1‐2) then min. 300 mg/day (no max.). N=33. 2. 'Positive' placebo (phenobarbital): dose 65 mg/day IM (days 1‐2) then min. 97.5 mg/day (no max.). N=37. 3. 'Negative' placebo (saline/lactose): dose IM (days 1‐2) then oral. N=37. 4. Promazine: dose 50 mg/day IM (days 1‐2) then min. 300 mg/day (no max.). N=32. 5. Mepazine: dose 25 mg/day IM (days 1‐2) then min. 75 mg/day (no max.). N=34. 6. Triflupromazine: dose 25 mg/day IM (days 1‐2) then min. 75 mg/day (no max.). N=36. 7. Prochlorperazine: dose 5 mg/day IM (days 1‐2) then min. 30 mg/day (no max.). N=32. 8. Perphenazine: dose 5 mg/day (days 1‐2) then min. 24 mg/day (no max.). N=36.
Outcomes	Leaving study early. Adverse effects. Global improvement. Unable to use ‐ Mental state: MSRPP, Psychotic Reaction Profile, Psychiatric Scale of Target Symptoms (no SD).
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	"The choice of the particular drug to be used in any case was based on a predetermined random selection" no further details reported.
Allocation concealment (selection bias)	Unclear risk	No information reported.
Blinding of participants and personnel (performance bias) All outcomes	Low risk	"Double‐blind" "The drugs were dispensed in standard unmarked capsules [...] coloured pink to mask all identifying consistencies and colours of the drugs".
Blinding of outcome assessment (detection bias) All outcomes	Low risk	"Double‐blind" "Ratings were blind, in that raters did not know which drug the patients was receiving".
Incomplete outcome data (attrition bias) All outcomes	High risk	"Those patients whom data were incomplete, together with those not receiving medication for at least ten days, were excluded from the study". "Of the 238 patients who met all population restrictions, 187 remained  in the project at least 10 days [...] Fifty‐nine patients completed the prescribed six‐week treatment course."
Selective reporting (reporting bias)	High risk	No SDs reported for the MSRPP, Psychotic Reaction Profile and Psychiatric Scale of Target Symptoms.
Other bias	Unclear risk	Funded by a research grant from the National Advisory Mental Health Council, National Institute of Health, US Public Health Service. Drugs provided by Smith Kline & French Laboratories, Squibb Laboratories, Shering Corporation and Warner‐Chilcott Laboratories.