Persson 2007.

Study characteristics
Methods	RCT. Parallel design with 2 treatment arms. Duration: median 4.3 months, range 3.6 ‐ 6.9 months. Location: Sweden.
Participants	Inclusion criteria: elderly people admitted to hospital for trauma or acute illness, at risk of malnutrition defined by MNA score < 10. Exclusion criteria: malignant disorders, terminal illness or with severe cognitive dysfunction. Number randomised: 108 participants (intervention group, n = 51; control group, n = 57). Attrition: 54 dropouts. Gender split: not reported. Age: mean (SD), intervention group 85 (5.9) years, control group 85 (6.1) years. Nutritional status: mean (SD) BMI, intervention 19.8 (1.9); control 20.6 (3.0).
Interventions	Intervention: participants received dietary advice and ONS in the form of individualised counselling to increase food intake, plus a nutritional supplement and a multivitamin supplement. Control: participants received no dietary advice and no ONS in the form of brief written dietary advice.
Outcomes	Weight *, BMI*, handgrip strength*, energy intake*, activities of daily living, cognitive function, peak expiratory flow, QoL.
Publication details	Language: English. Funding: the Swedish Research Council, Karolinska Institutet and by grants from S. Persson Family Foundation and Sempers Foods AB. Publication status: peer‐reviewed journal.
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Quote: randomisation performed by drawing files from a sealed box.
Allocation concealment (selection bias)	Low risk	Quote: randomisation performed by drawing files from a sealed box.
Blinding (performance bias and detection bias) Clinical outcomes	Low risk	Not blinded but clinical outcomes unlikely to be influenced by lack of blinding
Blinding (performance bias and detection bias) Functional outcomes	High risk	Not blinded but functional outcomes may have been affected by lack of blinding.
Blinding (performance bias and detection bias) Nutritional outcomes	High risk	Not blinded but nutritional outcomes may have been affected by lack of blinding.
Blinding of participants and personnel (performance bias) All outcomes	High risk	Not blinded 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	Not blinded and some outcomes likely to have been affected by lack of blinding.
Incomplete outcome data (attrition bias) All outcomes	High risk	ITT analysis using data obtained at inclusion carried forward and used at follow‐up for those who were still alive but not examined. Treated as protocol analysis also included. Attrition fully reported. 54 (50%) participants did not complete the study; 22/51 (43%) in the intervention group and 32/57 (56%) in the control group including 6/51 (12%) deaths in the intervention group and 12/57 (21%) deaths in the control group. The remaining participants withdrew consent or declined follow‐up; 8 participants in the control group had the intervention prescribed during the study. The high attrition rate and imbalance between groups might have influenced outcome assessment.
Selective reporting (reporting bias)	Unclear risk	No study protocol identified, thus unable to judge whether all planned outcomes were reported.Outcomes reported but not in a format suitable for meta‐analysis. Data on change in weight and handgrip strength are presented as mean (SD) at the start and end of the intervention and have therefore been obtained from authors. Data on mortality extracted from the paper.
Other bias	Unclear risk	Stated in text that baseline characteristics not different but data not shown.