Miller 2002.

Methods	Prospective randomized controlled trial.
Participants	Patients aged 18 to 70 years with an acute anterior shoulder dislocation.
Interventions	The IAL group received an intraarticular injection of 20 ml of 1% lignocaine into the affected shoulder. The IVAS group received an 2 mg midazolam and 100 µg fentanyl intravenously, respectively.
Outcomes	Rate of successful reduction, pain as rated on a visual analogue scale, time required for the reduction, time from reduction until discharge from the emergency department and cost.
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	High risk	Quote: “those with a odd medical record number received a local intra‐articular injection of lidocaine whereas those with an even medical record number received intravenous sedation”.
Allocation concealment (selection bias)	High risk	Published study does not provide information regarding allocation concealment. Comment: the nature of the interventions is such that allocation concealment is not feasible.
Blinding (performance bias and detection bias) All outcomes	High risk	Published study does not provide information regarding allocation concealment. Comment: the nature of the interventions is such that blinding is not feasible.