Jane 2011.

Methods	Study design: RCT Country: Taiwan Total: 72 randomised (massage group n = 36, no‐massage group n = 36)
Participants	Population: participants with metastatic bone pain aged ≥ 18 years; oriented to person, place, and time; able to speak and read Chinese; radiologically diagnosed with evident bone metastases via a bone scan; reportedly experiencing at least moderate metastatic bone pain, with an intensity of P4 on a 0‐10 scale Setting: oncology unit in Chang Gung Memorial Hospital (CGMH), a 3500‐bed‐capacity teaching medical centre in northern Taiwan Mean age: 49.9 years
Interventions	Intervention: massage therapy lasting 37‐50 minutes, mean 40 minutes Comparison: 45‐minute social attention intervention designed to provide the same amount of time and attention as the message therapy Country of training: not described Years in practice: first author was trained in massage therapy during a 4‐month period at the University of Washington in 2003 and had previous experience providing massage for 30 people with cancer
Outcomes	Outcome measures on day 2 (T1) to day 4 (T3): Present Pain Intensity (PPI)‐VAS Mood‐VAS Relaxation‐VAS Sleep‐VAS Symptom Distress Scale (SDS)‐VAS Demographic profile (DP) and medical profile (MP)
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Used computer random number generator
Allocation concealment (selection bias)	Low risk	Central allocation by computer
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Performed the same outcome assessment in both groups (see tables 1, 2, 3). 4 outcomes (PPI‐VAS, Mood‐VAS, Relaxation‐VAS, Sleep‐VAS) assessed for both the massage therapy and social attention groups
Incomplete outcome data (attrition bias) All outcomes	Low risk	Analysed the same number of participants in each group (see Figure 1; 36 allocated to massage therapy and 36 analysed, 36 allocated to social attention and 36 analysed)
Selective reporting (reporting bias)	Unclear risk	Quote: "the Mood‐VAS was adapted from the Linear Analog Self‐Assessment‐Profile of mood States, a 6‐item vertical form of VAS"
Size	Unclear risk	n = 72
Other bias	Low risk	No significant bias found Quote: "no significant differences between MT [massage therapy] and SA [social attention] groups on demographic and medical characteristics, nor on any of the 4 outcome measures gathered at baseline"