Diouf 2016.

Study characteristics
Methods	RCT. Parallel design with 2 arms. Duration: intervention during hospitalisation and then 9 weeks following discharge. Location: single centre in Senegal.
Participants	Inclusion criteria: people living with HIV/AIDS (any stage, taking or not‐taking antiretroviral treatment). Exclusion criteria: psychiatric illness, diabetes, physical disability, unable to eat. Number randomised: 65 participants (intervention group n = 32, control group n = 33). Gender split: 32% males, 68% females. Age: mean (SD) intervention 40 (12) years; control 42 (12) years. Nutritional status: severe malnutrition (BMI < 16 kg/m²): intervention group n = 34% and control group n = 24%.
Interventions	Intervention: participants received dietary advice and ONS in the form of standard hospital diet supplemented with supplement of 200 g/day* plus dietary counselling to improve diet at home (after discharge); supplementation was continued for 9 weeks at home. Control: participants received dietary advice alone in the form of standard hospital diet alone plus dietary counselling to improve diet at home (after discharge). *200 g of supplement made up of 100 g ready to use food mixed with 100 g of rice porridge. Ready to use food is composed of peanut butter and skimmed milk powder fortified with a vitamin‐mineral complex commercialised by Nutriset. The rice porridge (9.1 g rice flour per 100 mL water) was prepared extemporaneously, mixed with the ready to use food and served immediately.
Outcomes	Body weight, fat‐free mass, % body fat, dietary intake. Individual dietary intakes were measured and compared to the Recommended Dietary Allowances. Body composition was determined using Bio‐Impedance Analysis.
Publication details	Language: English. Funding: Universite Cheikh Anta Dio de Dakar, Senegal. UNICEF Senegal provided the products. Publication status: peer‐reviewed journal.
Notes	Both well‐nourished and malnourished participants were included. At our request, authors provided change scores for weight, fat‐free mass and body fat. Energy and protein intakes were only measured during 7 consecutive days in 10 randomly selected participants from each group during hospitalization; therefore results on nutritional intake are not presented in this review.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	The randomisation and assignment to group was performed by the senior researcher using a computer‐generated random number list (EPI INFO 6.0; Centers for Disease Control and Prevention, Atlanta).
Allocation concealment (selection bias)	Low risk	The randomisation and assignment to group was performed by the senior researcher using a computer‐generated random number list (EPI INFO 6.0; Centers for Disease Control and Prevention, Atlanta).
Blinding (performance bias and detection bias) Clinical outcomes	Unclear risk	No clinical outcomes reported.
Blinding (performance bias and detection bias) Functional outcomes	High risk	The study was unblinded. Functional outcomes could have been influenced by lack of blinding.
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	High risk	The study  was unblinded. Outcomes could have been influenced by lack of blinding.
Incomplete outcome data (attrition bias) All outcomes	Low risk	High drop‐out rates, but same numbers of dropouts across both groups (intervention group n = 12, control group n = 16). During hospitalisation, 6 participants died in the intervention group and 8 in the control group so 51 completed this stage (intervention group n = 26, control group n = 25). After 9 weeks of home monitoring, 6 participants dropped out of the intervention group (3 participants withdrew and 3 died) and 8 dropped out of the control group (4 participants lost to follow‐up and 4 died). Final analysis of 37 participants, 20 in intervention group and 17 in control group.
Selective reporting (reporting bias)	Low risk	Study protocol identified; all outcomes reported.
Other bias	Low risk	Trend towards less females, lower CD4 counts and lower Hb (all non‐significant) in the intervention group. Nutritional intake from the standard hospital diet was comparable between groups.