Calver 2015.

Methods	Allocation: randomised. Blindness: double. Duration: 2 hours.
Participants	Diagnosis: mental illness (N = 114), drug‐induced psychosis (N = 70), intoxication (N = 17), threatened self‐harm (N = 6), other/unknown (N = 19) (p.225)*. N = 228. Age: major than 18 years. Sex: 144 males, 84 females. History: admitted involuntarily to the psychiatric intensive care unit from the psychiatric emergency care centre (p.223). Excluded: patients willing to take oral medication for sedation without physical restraint or seclusion. Setting: psychiatry intensive care unit.
Interventions	1. Haloperidol: fixed dose 10 mg/IM. N = 110. 2. Droperidol: fixed dose 10 mg/IM. N = 118.
Outcomes	Tranquillisation or asleep, time to sedation. Need for additional benzodiazepines. Adverse events: respiratory rate less than 12 breaths/min, systolic blood pressure less than 90 mmHg, heart rate less than 60 bpm, oxygen saturation less than 90% or the presence of extrapyramidal side effects.
Notes	*when available, outcomes are reported for the mental illness sub‐group; gently provided by the authors (contacted 5th July 2016).
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	"Block randomisation was used. Microsoft Excel was used to randomly create blocks of four (ABAB, AABB, etc.) or six (ABABAB, AAABBB, etc.)" (p.224).
Allocation concealment (selection bias)	Low risk	"Pre‐packed treatment kits [...] contained either droperidol (10mg in 2ml) or haloperidol (10mg in 2ml) [...] transferred into vials identical" (p.224).
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Double‐blind study.
Blinding of outcome assessment (detection bias) All outcomes	Low risk	"labels A or B and not the drug names were known to the investigator. This investigator analysed the data independently and presented this to the other investigators" (p.224).
Incomplete outcome data (attrition bias) All outcomes	Low risk	No evidence of incomplete outcome data.
Selective reporting (reporting bias)	Unclear risk	"In breach of the study protocol midazolam was given nine times simultaneously with the study drug" (p.226); no further evidence of selective reporting.
Other bias	High risk	Short study.