Luckey 1967.

Methods	Randomisation: “random assignment”. Blinding: double‐blind. Duration: two‐week washout period (with placebo); followed by 12 weeks on active drug. Design: parallel. Location: Minneapolis Veterans Administration Outpatient Clinic. Setting: outpatients.
Participants	Diagnosis (study completers, N = 9): paranoid schizophrenia (N = 4), schizoaffective schizophrenia (N = 3), undifferentiated schizophrenia (N = 2). N = 26. Gender: “most were male”, no further details available. Age: not indicated. History: duration stable: not indicated, duration of illness: "at least one year immediately prior the study", number of previous hospitalisations: not indicated, age at onset: not indicated, severity of illness: not indicated, baseline antipsychotic dose: not indicated.
Interventions	1. Haloperidol: flexible dose, dose range: maximum dose 15 mg/day, mean dose: not indicated. N = 13. 2. Trifluoperazine: flexible dose, dose range: maximum dose 30 mg/day, mean dose: not indicated. N = 13.   dosage scheme: “a ratio of 2mg trifluoperazine to 1mg haloperidol” “we expected to increase the initial dosage by one capsule [2,5mg haloperidol or 5mg trifluoperazine] at each evaluation time until the dosage was four capsules [10mg haloperidol or 20mg trifluoperazine].” Rescue medication: benztropine mesylate (Cogentin) (1 to 2mg/day).
Outcomes	Examined: Clinically important response to treatment: Global clinical judgement. Mental state general: Brief Psychiatric Rating Scale (BPRS) (no SDs available). Leaving the study early due to any reason. Adverse effects: checklist of side effects: at least one movement disorder. Unable to use: Behaviour: Minnesota Multiphasic Personality Inventory (MMPI) (no raw data available).
Notes	Study participants were 26 chronic schizophrenic outpatients. “Of the 26 patients who began the study, 21 completed one month or more of active medication.” “we made use of data on any patient who completed one month or more of active medication”.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	“Random assignment”. No further details.
Allocation concealment (selection bias)	Unclear risk	No further details.
Blinding (performance bias and detection bias) (performance bias)	Low risk	“Double‐blind”. “Haloperidol, trifluoperazine and placebo were prepared in identical capsules.”
Blinding (performance bias and detection bias) (detection bias)	Unclear risk	“Double‐blind”. No further details.
Incomplete outcome data (attrition bias) All outcomes	High risk	The overall‐attrition was very high (17 of 26 participants; 65.4%). 8 of 13 participants left the trial early in the haloperidol‐group (61.5%) and 9 of 13 participants in the trifluoperazine‐group (69.2%). A completers‐only analysis was used regarding the main outcome (global clinical judgment) and for the other outcomes the analysis was based on the results of one month active drug treatment.
Selective reporting (reporting bias)	High risk	Outcome data reporting was incomplete (no SDs for the BPRS; no raw data for the MMPI). Only the most prevalent adverse effects were reported.
Other bias	High risk	“Due to the large number of dropouts and the overall limitations imposed by the small population, none of the results were statistically significant.” “The dosage ratio of 2mg trifluoperazine to 1mg haloperidol was probably too low and may have put haloperidol at a disadvantage.”