Norman 2008b.

Study characteristics
Methods	RCT. Parallel design with 2 treatment arms. Duration: 3 months from hospital discharge. Location: single centre in Germany.
Participants	Inclusion criteria: adults with benign gastrointestinal disorders admitted to hospital and classified as malnourished according to SGA criteria. Exclusion criteria: malignant disease, renal insufficiency (serum creatinine 41.3mg/dl), and life expectancy < 3 months or age < 18 years. 2008: Number randomised: 101 participants (intervention group, n = 48, control group, n = 48). Attrition: 21 dropouts, 10 intervention group (withdrew before baseline) and 11 lost to follow‐up in the control group. Gender split: not stated. Age: mean (SD), intervention group 52.2 (16.5) years; control group 53.6 (16.8) years. Nutritional status: all malnourished according to SGA (grade B or C). 2011: Number randomised: 160 participants, 114 completed the study (intervention group, n = 60; control group, n = 54). Gender split: 57 males, 57 females. Age: mean (SD) total cohort 50.6 (16.1) years. Nutritional status: all malnourished according to SGA (grade B or C).
Interventions	Intervention: participants received dietary advice and ONS in the form of dietary counselling from a dietitian to increase energy and protein intake from food and up to 3 x 200 mL Fresubin protein energy drinks. Control: participants received dietary advice alone in the form of dietary counselling to increase energy and protein intake from food.
Outcomes	2008: Energy intake*, weight*, height, BMI*, TSF*, MUAC*, body composition (BIA), handgrip strength*, length of stay, number of readmissions*, number of prescribed drugs on discharge, peak expiratory flow. 2011: QoL, cost.
Publication details	Language: English. Funding: grant from Fresenius Kabi, Bad Homburg. Publication status: peer‐reviewed journal.
Notes	A first study was published in 2008; new data were published in 2011. Quality‐adjusted life years were calculated by adopting the area under the curve method. QoL was assessed with Short‐Form (SF)‐36 Health Survey and SF‐36 values were transformed into health‐status utilities.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Participants were randomised according to a computer‐generated randomisation list kept by a co‐worker not involved in the study.
Allocation concealment (selection bias)	Low risk	Central allocation, web‐based by a co‐worker not involved in the study.
Blinding (performance bias and detection bias) Clinical outcomes	Low risk	Not reported, in the 2008 paper the main outcome was number of readmissions. It is unlikely that this may have been influenced by knowing the group allocation.
Blinding (performance bias and detection bias) Functional outcomes	High risk	Although not reported, intervention participants received supplements and control participants did not. Functional outcomes could have been influenced by lack of blinding.
Blinding (performance bias and detection bias) Nutritional outcomes	High risk	Although not reported, intervention participants received supplements and control participants did not. Nutritional outcomes could have been influenced by lack of blinding.
Blinding of participants and personnel (performance bias) All outcomes	High risk	Not described and likely that researchers and participants were aware of group allocation as this was a nutritional intervention. 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 functional and nutritional outcomes.
Incomplete outcome data (attrition bias) All outcomes	Low risk	2008: 21 dropouts: Dietary counselling and supplement group: 10 withdrew before baseline; Dietary counselling alone: 11 lost to follow‐up. Also, in the dietary counselling and supplement group 8 known to not take the supplement, but included in the ITT analysis. In the dietary counselling group, 4 reported consuming nutritional supplements during the study period. 2011: 160 participants were recruited for the study, of which 120 completed the study. Dietary counselling and supplement group: 12 withdrew before the start, 8 lost to follow‐up. Dietary counselling alone: 20 lost to follow‐up, 6 did not complete SF 36 QoL questionnaires. ITT analyses for 60 participants in intervention group and 54 participants in control group.
Selective reporting (reporting bias)	Unclear risk	No study protocol identified, thus unable to judge whether all planned outcomes were reported. Data on mean change (SD) for weight and grip strength were extracted from the paper. Data on TSF and MUAC were not presented but were assessed and so have been obtained from author. Details of hospital admissions are not reported clearly and therefore have been clarified with the authors. For the 2011 manuscript 6 participants did not complete SF36 correctly and all were assigned to the control group.
Other bias	Low risk	Baseline characteristics described in text as not different and data given for some variables.