Edge 2003.

Methods
Participants
Interventions
Outcomes
Notes	See Table 6
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	"Patients were randomly placed in the hospital rooms by the hospital administration upon their admission to the hospital" ‐ Method of randomisation not described.
Allocation concealment (selection bias)	High risk	Patients were recruited into the study on their final day of hospital stay (after they had been allocated to a room).
Blinding (performance bias and detection bias) Subjective outcomes	High risk	Patients aware of group allocation.
Blinding (performance bias and detection bias) Objective outcomes	Low risk	Length of stay and drug requests objectively recorded.
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	"Eleven patients who were approached could not be used in the study. Three patients explained that they would just prefer not to participate. Two of the patients were considered legally colorblind. [...] One patient was extremely confused and could not understand the information being explained to him. [...] One patient felt uncomfortable with having the principal investigator examine his medical records [...]. One patient was considered extremely depressed by the nurse manager and was not approached to participate in the study and two patients could not speak English [...]. Lastly, one patient was extremely nervous and worried that the anxiety test would reveal that that she should have to stay in the hospital for an extended amount of time. It was therefore the researcher’s opinion that she altered her answers on the test to make it appear as though she had no anxiety in her life." It is unclear from which rooms withdrawals came. Data of remaining included patients appears complete.
Selective reporting (reporting bias)	High risk	Insufficient data provided for extraction: Anxiety, pain medication, and length of stay.
Protection from contamination?	Low risk	Patients assigned to individual rooms.