Burden 2011.

Study characteristics
Methods	RCT. Parallel design with 2 arms. Duration: pre‐operative intervention (mean (range, SD) time to surgery was 37 (7 – 371 days, 54.7 days); participants followed up until 3 months after operation. Location: single centre in the UK.
Participants	Inclusion criteria: diagnosed with colorectal cancer and scheduled for surgery with pre‐operative weight loss > 1 kg/3 ‐ 6 months. Exclusion criteria: pregnant, enrolled in another trial, unable to give informed consent, had an inoperable tumour. Number randomised: 125 participants (intervention group n = 59 (analysed n = 54), control group n = 66 (analysed n = 62). Gender split: intervention group 63% male; control group 61% male. Age: mean (SD), intervention group 64.5 (13.9) years; control group 65.3 (2.7) years. Nutritional status: 83 (71%) participants had lost weight in 3 – 6 months preceding surgery; the % of usual body weight lost was in the range 1% – 31% (mean (SD) 5.8% (6.5%)).
Interventions	Intervention: participants received dietary advice and ONS in the form of 400 mL of oral supplement and dietary advice (consisting of increasing energy and protein from food, based on an information leaflet*). Control: participants received dietary advice alone in the form of dietary advice as described above. *The author explained: "The leaflet was Build yourself up: for patients undergoing surgical procedures". It included the eat well plate, foods to include in your diet going through the food groups, There was a section on increasing energy and protein content of the diet. At the end it included high calorie snacks. Dietary counselling was not provided in the sense of dietetic practice i.e. assessment of habitual intake and tailored advice. The research assistant talked the patient through the written leaflet. So this advice was usual care as it is usually given out in pre‐op clinics by clinic nurses".
Outcomes	Primary outcome: number of post‐operative complications. Secondary outcomes: use of postoperative antibiotics, length of hospital stay, energy and protein intake, complications, grip strength.
Publication details	Language: English. Funding: NHS fellowship award and Central Manchester foundation trust small awards. Publication status: peer‐reviewed journal.
Notes	Of the 54 participants randomised to the treatment group, 50 completed a diary to record compliance to the intervention (2 full cartons of supplement daily). It was reported that 36 (72%) of participants managed 100% of the intervention, 8 (16%) managed 50% (at least 1 carton daily) and 6 (12%) managed < 25% of the intervention. Tha author provided change scores for energy intake and protein intake on our request.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Numeral sequence of random blocks generated by independent statistician.
Allocation concealment (selection bias)	Low risk	Sequentially numbered brown opaque envelopes.
Blinding (performance bias and detection bias) Clinical outcomes	Low risk	Incomplete blinding, personnel blinded, participants not blinded. It is unlikely that post‐op complications is influenced by lack of blinding.
Blinding (performance bias and detection bias) Functional outcomes	Unclear risk	Functional outcomes were not measured in this study.
Blinding (performance bias and detection bias) Nutritional outcomes	High risk	It is possible that there is bias associated with dietary intake recorded by 24 h recall because participants were unblinded.
Blinding of participants and personnel (performance bias) All outcomes	High risk	Incomplete blinding, personnel blinded, participants not blinded. It is possible that assessment of some outcomes was influenced by lack of blinding
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Judged to be unclear because low risk for clinical outcomes and high risk for nutritional outcomes.
Incomplete outcome data (attrition bias) All outcomes	Low risk	5 of 54 (9%) in the intervention group and 4 of 62 (6%) in the control group did not have surgery so were excluded from the analysis.
Selective reporting (reporting bias)	Unclear risk	No study protocol identified, thus unable to judge whether all planned outcomes were reported. Data on change in grip strength were not reported.
Other bias	Unclear risk	Baseline characteristics were compared and were broadly similar in the two groups. Dietary intake was measured by unstructured dietary recalls; this method is not very reliable. The SD of the mean change in energy intake was twice as high as the mean change, indicating a skewed distribution.