Baldwin 2008.

Methods	Randomised controlled trial. Duration 12 months. Intervention 6 weeks and follow‐up to 12 months.
Participants	Adults (n = 358: 256 men, 102 women; median age 66, range 24 ‐ 88 years) with locally advanced or metastatic cancers of the gastrointestinal tract (n = 277) or non‐small‐cell lung cancer or mesothelioma. All patients had lost weight at the start of the trial (mean 9.8% (SD 6%) in lung patients and 11.2% (SD 6.4%) GI patients). 153 participants were alive at 12 months: No intervention group: 47 deaths and 2 withdrawals. Dietary advice group: 52 deaths and 2 withdrawals. Nutritional supplements group: 55 deaths and 2 withdrawals. Dietary advice and supplements group: 44 deaths and 1 withdrawal.
Interventions	This was a 2 x 2 factorial trial. Group 1 received no additional intervention (n = 96). Group 2 received dietary advice to increase intake by 600 kcals per day) (n = 90). Group 3 received an oral nutritional supplement providing 588 kcals per day (n = 86). Group 4 received dietary advice to increase intake by 600 kcals per day and an oral nutritional supplement (n = 86).
Outcomes	Survival, QoL, weight, handgrip strength, energy intake.
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Randomisation was performed by an independent trials centre using a computer‐generated list. Participants were stratified for performance status and site of disease.
Allocation concealment (selection bias)	Low risk	Group allocation was concealed until participants had signed consent to participate.
Blinding (performance bias and detection bias) Clinical outcomes	High risk	The outcome assessors were not blinded.
Blinding (performance bias and detection bias) Functional outcomes	High risk	The outcome assessors were not blinded.
Blinding (performance bias and detection bias) Nutritional outcomes	High risk	The outcome assessors were not blinded.
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	153 participants were alive at 12 months: No intervention group: 47 deaths and 2 withdrawals. Dietary advice group: 52 deaths and 2 withdrawals. Nutritional supplements group: 55 deaths and 2 withdrawals. Dietary advice and supplements group: 44 deaths and 1 withdrawal. Reasons for withdrawal not known for all patients and therefore risk of bias.
Selective reporting (reporting bias)	Low risk	The study is published as an abstract at present. The data on survival, change in weight and change in energy intake are presented in the abstract but not in a format that would enable entry into a meta‐analysis, therefore original data have been provided by the authors for this review. The numbers of participants completing assessment of energy intake was only 31 of 358 and so these data should be interpreted with caution.
Other bias	Low risk	Baseline characteristics for the 4 groups were similar.