Twiss 2006.

Methods
Participants
Interventions
Outcomes
Notes	See Table 8
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	"Randomized [...] by picking a slip of paper from one of two boxes (one for women and one for men so that gender could be matched)"
Allocation concealment (selection bias)	Low risk	"After signing informed consent, participants were randomized into control and experimental groups by picking a slip of paper". Informed consent was taken prior to randomisation.
Blinding (performance bias and detection bias) Subjective outcomes	High risk	Participants aware of group allocation.
Blinding (performance bias and detection bias) Objective outcomes	Unclear risk	Unclear if intubation time was assessed blind or was automated.
Incomplete outcome data (attrition bias) All outcomes	High risk	"There were 44 participants in the control group and 42 in the experimental group. [...]. Participants for this study continued to be recruited as others dropped out until a total of 60 participants completed the study. [...]. The elimination of 26 participants because of unforeseen complications during and after surgery is a limitation in this study and threat to the internal validity of results. These complications required extended intubation periods and eliminated participants from the normal weaning protocol or were unable to complete the state anxiety measure on the third postoperative day." 30% of participants were withdrawn (14 from the music group, and 12 from the control group).
Selective reporting (reporting bias)	Unclear risk	Not known.
Protection from contamination?	Low risk	Interventions administered on an individual basis.