Anderson 2004.

Methods	RCT of peppermint oil, IPA or normal saline aromatherapy to treat PONV Setting: PACU acute hospital, USA
Participants	33 patients aged 18 years + having surgery under general or regional anaesthesia, or deep IV sedation, who reported nausea in PACU. Treatment groups did not differ in the percentage having general anaesthesia, the type of surgery, age or gender distribution. Exclusions: patients who were unable to give informed consent; patients who did not require anaesthesia services
Interventions	On the participant's spontaneous report of PON, they were instructed to take three slow deep breaths to inhale the vapours from a pre‐prepared gauze pad soaked with either peppermint oil (n = 10), IPA (n = 11), or normal saline placebo (n = 12) held directly under their nostrils. After 2 min the participant was asked to rate their nausea by VAS and given the choice to continue aromatherapy or have standard IV antiemetics. At 5 min post the initial treatment, the participant was again asked to rate their nausea and if they would like to continue aromatherapy or have standard IV antiemetics.
Outcomes	Severity of nausea as measured on 100 mm VAS at 2 min and 5 min after treatment. VAS from 'no nausea' to 'worst possible nausea'. Choosing to use 'rescue' antiemetics Satisfaction with management of nausea, as measured by 100 mm VAS with range from 0 = extremely dissatisfied to 100 = fully satisfied
Notes	Possible lack of accuracy with some participants self‐recording data in PACU if they had poor or blurred vision. Authors Lynn Anderson and Dr Jeffrey Gross emailed to request further information on group sizes, which was supplied by Dr Gross. Supported by the Department of Anesthesiology, University of Connecticut School of Medicine.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	"...group assignments were made in a randomised, double‐blind fashion" Comment: probably done. Nurses administering treatment were unaware of contents of each package of treatment materials. Patients who had consented to participate entered study when they spontaneously reported nausea.
Allocation concealment (selection bias)	Low risk	"A random number generator determined the contents of each serially numbered bag." "...prepared by an individual not otherwise involved in the study..." Data "analysed by investigator unaware of treatment allocation". Comment: probably done
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Staff administering treatment blinded by use of "lightly scented" surgical masks. Comment: probably done
Blinding of outcome assessment (detection bias) All outcomes	High risk	Participants were self‐reporting subjective assessment of nausea and were not blinded. Comment: due to the strong aroma of the peppermint oil, it would be impossible to blind the participant receiving this to their allocation once treatment commenced. Probably not done
Incomplete outcome data (attrition bias) All outcomes	Low risk	Comment: outcomes reported for all participants
Selective reporting (reporting bias)	Low risk	Comment: results reported for all stated outcomes
Other bias	Low risk	Comment: study appears to be free of other sources of bias