Beck 2012.

Study characteristics
Methods	RCT. Parallel design with 2 arms. Duration: 8 weeks intervention with 26 weeks follow‐up. Location: University Hospital of Herlev in Denmark (but participants drawn from 3 municipalities).
Participants	Inclusion criteria: identified as at nutritional risk according to the level 1 screen NRS 2002; were 65+ years of age, living in three municipalities (Herlev, Roedovre or Gladsaxe), hospitalised for a minimum of 2 days in the geriatric medicine wards of the University Hospital of Herlev. Exclusion criteria: suffered from senile dementia or terminal disease; could not understand the Danish language; resident in nursing homes; or unable to or willing to give informed consent. Number randomised:152 elderly participants (over 65 years of age) randomised, 124 completed study. Gender split: intervention group 54 females (74%) and 19 males (26%); control group 57 females (72%) and 22 males (28%). Age: data not reported (all greater than 65 years). Nutritional status: all at nutritional risk assessed using NRS 2002.
Interventions	Intervention: participants received dietary advice plus ONS if required in the form of comprehensive nutritional assessment by a dietitian, followed by 3 home visits with individualised nutritional counselling by a registered dietitian complemented with 3 follow‐up visits conducted by the GP. Control: participants received no dietary advice and no ONS in the form of 3 follow‐up visits by GPs alone.
Outcomes	Primary outcome: risk of readmissions*. Secondary outcomes: functional status (hand grip strength*, chair stand, mobility, disability and tiredness in daily activities, rehabilitation capacity), nutritional status (weight*, BMI, energy* and protein intake*), need of social services (home care, home nursing, meals‐on‐wheels) and mortality*.
Publication details	Language: English. Funding: grants from the Health Insurance Foundation, the Tryg Foundation and the General Practitioners’ Foundation for Development of General Practice. These are all non‐commercial and had no role in study design, or in the collection, analysis, interpretation and publication of the data. TDC provided cell phones for scientific research assistants and registered dietitians and, as the others, had no role in study design, or in the collection, analysis, interpretation and publication of data. Publication status: peer‐reviewed journal.
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote from paper: "Participants were randomized at discharge. Each allocation using generated random numbers was written on paper and concealed in a serially numbered, opaque envelope. The scientific research assistants opened the next envelope after recruiting each participant and then contacted the GP and, if relevant, the registered dietitians. Hence the scientific research assistants, who collected the outcome data, knew which group a participant was in. The principal investigator was the only one blinded for the intervention". Randomisation ratio not described.
Allocation concealment (selection bias)	Low risk	Quote from paper: "Each allocation using generated random numbers was written on paper and concealed in a serially numbered, opaque envelope."
Blinding (performance bias and detection bias) Clinical outcomes	Low risk	Comment: Not blinded. Scientific research assistants were aware of group assignment. Outcome not likely influenced by lack of blinding.
Blinding (performance bias and detection bias) Functional outcomes	High risk	Comment: Not blinded. Hence the scientific research assistants, who collected the outcome data, knew which group a participant was in. Outcomes could have been influenced by lack of blinding.
Blinding (performance bias and detection bias) Nutritional outcomes	High risk	Comment: Not blinded. Hence the scientific research assistants, who collected the outcome data, knew which group a participant was in. Outcomes could have been influenced 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	High risk	Not blinded, most of the outcomes were functional and nutrition parameters; knowing the group allocation could have influenced outcome assessment.
Incomplete outcome data (attrition bias) All outcomes	Low risk	Comment: 34/152 (22%) attrition. Withdrawals were balanced across groups: 18 (23%) in control group, and 16 (22%) in intervention group. Otherwise no missing data. It was even possible to obtain follow‐up data from some of those who withdrew.
Selective reporting (reporting bias)	Low risk	Study protocol identified Clintrials.gov. All specified outcomes reported.
Other bias	Low risk	Comment: baseline characteristics were similar between groups.