Persson 2002.

Methods	Randomised controlled trial. Duration 24 months.
Participants	Adults (n = 142, age range 42 ‐ 89 years) with <5% weight loss who were newly diagnosed with colorectal or gastric cancer.
Interventions	Nutritional counselling to increase food intake to Nordic Nutrition Recommendations and a prescription for nutritional supplements if wanted (n = 67) versus standard care (n = 70).
Outcomes	Survival*, weight*, BMI*, energy intake*.
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Information from the author, random sequence generated on computer by independent centre.
Allocation concealment (selection bias)	Low risk	Information from the author, allocation performed by independent centre, allocation concealed until patient recruited.
Blinding (performance bias and detection bias) Clinical outcomes	Unclear risk	Not stated.
Blinding (performance bias and detection bias) Functional outcomes	Unclear risk	Not stated.
Blinding (performance bias and detection bias) Nutritional outcomes	Unclear risk	Not stated.
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	Information from author. 137 patients randomised in the study, at 24 months there were 25 deaths, 5 withdrawals and 3 patients excluded in the intervention group and 26 deaths, 14 patients withdrawn and 1 exclusion in the control group.
Selective reporting (reporting bias)	Low risk	The data on mortality is unclear in the paper. The data on weight change is presented partly in text and partly in figures and not suitable for direct entry into meta‐analysis. The data on energy intake is presented as % recommendations. All data included in the review has been obtained from the author including data at 4 time‐points.
Other bias	Low risk	Baseline variables given, groups similar at baseline.