Ahles 1999.

Methods	Study design: RCT Country: USA Total: 35 randomised (massage group n = 16, no‐massage group n = 19)
Participants	Population: people undergoing autologous BMT; diagnosis ‐ breast cancer, Non‐Hodgkin's lymphoma, acute myelogenous lymphoma, Hodgkin's disease, acute myelogenous leukaemia, or ovarian cancer Setting: Dartmouth‐Hitchcock Medical Center after providing informed consent for participation in the study Mean age: 41 years
Interventions	Intervention: 9 x 20‐minute massages during hospital stay. Mean length of hospital stay was 3 weeks at the time that the study was conducted Comparison: usual care Country of training: not reported Years in practice: not reported
Outcomes	Outcome measures at pretreatment (day ‐7), mid‐treatment (range = day ‐1 to day +7), and pre‐discharge State‐Trait Anxiety Inventory (STAI) Beck Depression Inventory (BDI) Brief Profile of Mood States (POMS) Numerical scales (0‐10) of emotional distress, fatigue, nausea, and pain Psychophysiological measures ‐ DBP, SBP, respiration, and pulse
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Quote: "patients were randomly assigned", but provided no further information
Allocation concealment (selection bias)	Unclear risk	Not reported
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Performed the same outcome assessment in both groups Quote: "the overall effects of massage therapy on anxiety, depression, and mood were assessed pre treatment, mid treatment, and pre discharge by having patients in both groups complete the following measures"
Incomplete outcome data (attrition bias) All outcomes	Low risk	Quote: "one patient in the control group did not complete all of the assessment because of significant medical conditions"
Selective reporting (reporting bias)	Unclear risk	No selective outcome reporting Quote: "the POMS is an 11‐item adjective checklist which provides a summary measure of general distress or mood. The brief POMS has been shown to be highly correlated with the original 65‐item POMS, a reliable and valid measure of mood states"
Size	High risk	n = 35
Other bias	High risk	Imbalance found in the baseline measurement of nausea and distress between the massage and standard care group; however, no significant between‐group differences in participant characteristics and any class of medication