Table 4.

GRADE quality of evidence ratings

	Certainty assessment							Certainty
Variable	Number of studies	Study design	Risk of bias	Inconsistency	Indirectness	Imprecision	Other considerations
Opiate consumption	5	Randomized trials	Not serious	Very seriousa	Not serious	Not serious	Publication bias strongly suspectedb	⨁○○○ Very low
Pain intensity	4	Randomized trials	Seriousc	Not serious	Not serious	Not serious	Publication bias strongly suspectedb	⨁⨁○○ Low
ED admission	1	Randomized trials	Very seriousd	Not serious	Not serious	Not serious	Publication bias strongly suspectedb	⨁○○○ Very low
ED length-of-stay	1	Randomized trials	Not serious	Not serious	Not serious	Not serious	Publication bias strongly suspectedb	⨁⨁⨁○ Moderate
Patient satisfaction	2	Randomized trials	Not serious	Not serious	Not serious	Not serious	Publication bias strongly suspectedb	⨁⨁⨁○ Moderate
Extrapyramidal side effects	3	Randomized trials	Very seriouse	Not serious	Not serious	Not serious	Publication bias strongly suspectedb	⨁○○○ Very low

Notes:

^aTwo study drugs were assessed: haloperidol (two studies) and droperidol (three studies). Significant variability existed in control groups: conventional therapy (one study), morphine plus cyclizine (one study), tramadol (one study), morphine alone (one study), and dexamethasone and ondansetron (one study).

^bSmall number of studies resulting in likely type II error. Absence of negative or neutral studies suggests evidence of publication bias.

^cRisk of bias high in one study, low in three studies.

^dThe study was judged to have a high risk of bias.

^eOne study at high risk, one study at moderate risk, and one study at low risk for bias.

Abbreviations: ED, emergency department; GRADE, Grading of Recommendations, Assessment, Development and Evaluations.