Huynh 2015.

Study characteristics
Methods	Prospective RCT. Parallel design. Duration: 12 weeks. Location: multicentre trial at 9 private and 4 public hospitals across India.
Participants	Inclusion criteria: aged ≥ 18 years, males and non‐pregnant, non‐lactating females, admitted within 36 hours to either the medical or the surgical wards, and who were diagnosed with moderate or severe malnutrition based on the modified SGA. Exclusion criteria: active tuberculosis, acute hepatitis, HIV infection, diabetes, dementia, brain metastases, active malignancy, severe renal or liver failure, burn injury,, clinically significant ascites, oedema, eating disorders or psychological condition interfering with nutritional intake, taking progesterone, steroids or growth hormone. Diagnosis groups: neurological disorders, respiratory medicine, cardiovascular medicine, gastrointestinal disorders, genitourinary and haematological, trauma and orthopaedic diseases, various internal medicine (infection, malaria), others. Number randomised: 212 adults. Gender split: intervention group 57% males, control group 55% males. Age: mean (SD) age intervention group 40.9 (19.6) years; control group 39.0 (16.4) years. Nutritional status: all but 4 participants were moderately or severely malnourished according to modified SGA. These 4 were randomised according to ITT principles, and allocated to intervention group (n = 1) or control group (n = 3).
Interventions	Intervention: participants received dietary advice and ONS in the form of 3 sessions of dietary counselling (administered at baseline, weeks 4 and 8) plus 2 servings of oral nutritional supplement per day for 12 weeks. Control: participants received dietary advice alone in the form of 3 sessions of dietary counselling (administered at baseline, weeks 4 and 8). Dietary counselling was provided to both groups by hospital dietitians who were trained in using a standardised methodology. In each dietary counselling session, energy and nutritional requirements were calculated for each group to achieve their individualised energy goals. The dietitians instructed the participants on the methods for improving their nutritional intake using home foods for the control group and home foods in conjunction with oral nutritional supplement for the intervention group. At hospital discharge and follow‐up visits at weeks 4 and 8, participants in the control group were given instructions on consuming small frequent meals and using protein rich foods and a high‐energy food‐source. In addition to being advised on using home foods in meal preparation, participants in the intervention group were instructed to consume the oral nutritional supplement between meals.
Outcomes	Primary outcome: change in weight over 12 weeks. Secondary outcomes: change in BMI, modified SGA score, pre‐albumin, albumin, haemoglobin, total protein and C‐reactive protein over 12 weeks, change in dietary intake (energy intake and macronutrient intake) and functionality using handgrip strength.
Publication details	Language: English. Funding: Abbott Nutrition provided funding for the study and was responsible for the study design, monitoring, data analysis, manuscript preparation and submission. Publication status: peer‐reviewed journal.
Notes	Outcomes are reported at 4 weeks, 8 weeks and 12 weeks. For meta‐analysis, only the 12 weeks outcomes were used. The authors were contacted to provide change data for weight, BMI, dietary intake and grip strength
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "Eligible patients were randomised in a 1:1 ratio to the control or oral nutritional supplement group. Sealed envelopes containing the patient group assignment were prepared from randomisation schedules generated by Abbott Nutrition for each site."
Allocation concealment (selection bias)	Low risk	Quote: "As eligible patients were enrolled, they were assigned a subject number sequentially starting with the first envelope."
Blinding (performance bias and detection bias) Clinical outcomes	Low risk	Unknown whether the study was blinded or not. Nevertheless, this is unlikely to influence biochemical parameters.
Blinding (performance bias and detection bias) Functional outcomes	Unclear risk	Quote: "Anthropometric measurements were performed by study staff trained in standardised methods of conducting the measurements." It is unknown whether these persons were aware of group allocation.
Blinding (performance bias and detection bias) Nutritional outcomes	High risk	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 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	Judged to be unclear because low risk for clinical outcomes and unclear or high risk for functional and nutritional outcomes.
Incomplete outcome data (attrition bias) All outcomes	Low risk	Screened and randomised n = 212, allocated to each group n = 106, withdrew consent n = 5 (intervention n = 2, control n = 3), not completed n = 54 (intervention n = 30, control n = 24) ‐ most important reasons: lost to follow‐up, withdrew consent. No differences between groups.
Selective reporting (reporting bias)	Low risk	Study protocol identified; all relevant outcomes reported.
Other bias	Unclear risk	Baseline characteristics comparable between the two groups, except for weight (mean (SD)) intervention group 46 (9.6) kg, control 48.5 (10.5) kg).