Endevelt 2011.

Study characteristics
Methods	RCT. Parallel design with 3 arms. Duration: 6 months. Location: Haifa, Israel
Participants	Inclusion criteria: adults aged 75 and older, community dwelling at nutritional risk ((serum cholesterol below 160 mg/dL or serum albumin below 3.5 mg/dL or total lymphocyte count of less than 1800) AND (MNA‐sf below 10 or more than 10% weight loss in 6 months)). Exclusion criteria: diagnosis of cancer or liver disease, clinical depression, cognitive impairment (MMSE < 23) and inability or unwillingness to sign and informed consent. Number randomised: 127 people agreed to participate, 68 participants randomised but 59 people not randomised due to communication and language difficulties or unwillingness to have home visits by a dietitian. Total cohort, n = 127; Group 1, n = 59; Group 2, n = 35; Group 3, n = 33. Gender split: Group 1, 24 male, 35 female; Group 2, 13 male, 22 female; Group 3, 12 male, 21 female. Age: mean (SD) Group 1, 84.5 (5.6) years; Group 2, 84.2 (6.0) years; Group 3, 84.7 (4.7) years. Nutritional status: mean (SD) BMI: Group 1, 27.4 (5.2) kg/m²; Group 2, 27.3 (5.0) kg/m²; Group 3, 27.0 (5.2) kg/m².
Interventions	Intervention group 1: participants received dietary advice plus ONS if required in the form of dietetic intervention treatment where each participant had 5 meetings with the clinic dietitian in their homes over a period of 6 months (baseline, 2 weeks, 1 month, 2 months, 6 months); the intensity varied according to the severity of the undernutrition and the content of the meetings was protocolled. Intervention group 2: advice by the primary care physician and a booklet on nutrition. Intervention group 3: participants received no dietary advice and no ONS in the form of standard care.
Outcomes	MNA, food frequency questionnaire (vitamins, minerals, protein), biochemical measurements, health care costs*, cognition (MMSE), depression (GDS‐sf), ADL (Barthel).
Publication details	Language: English. Funding: grant from the National Institute for Health Policy Israel (NIHP). Publication status: peer‐reviewed journal.
Notes	The sample size of intervention group 2 was too small according to the sample size calculation.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	The procedure of randomisation was not reported. From the 127 individuals who agreed to participate, only 68 participants were randomised to either dietetic intervention or medical intervention. The other 59 individuals were not included in the randomisation process due to communication and language difficulties or unwillingness to have home visits by a dietitian.
Allocation concealment (selection bias)	Unclear risk	Not reported.
Blinding (performance bias and detection bias) Clinical outcomes	Low risk	However, this is unlikely to influence clinical outcomes.
Blinding (performance bias and detection bias) Functional outcomes	High risk	Assessments were performed by trained interviewers. Details on blinding are not reported. The design suggests that the study was unblinded. Functional outcomes could have been influenced by lack of blinding.
Blinding (performance bias and detection bias) Nutritional outcomes	High risk	Assessments were performed by trained interviewers. Details on blinding are not reported. The design suggests that the study was unblinded. 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	Unclear risk	Not reported.
Selective reporting (reporting bias)	Low risk	Study protocol identified Clintrials NCT00316966. All specified endpoints reported but not all endpoints were reported in a way that made them usable for meta‐analysis.
Other bias	Low risk	Baseline characteristics were similar, except for a trend for higher education in group 1.