Alo 2014.

Study characteristics
Methods	RCT ("quasi randomised" ‐ no further details). Parallel design with 2 arms. Duration: intervention and follow‐up: 6 months. Location: Nigeria.
Participants	Inclusion criteria: diagnosed with HIV and receiving HAART, providing consent, resident with Abakaliki town (Nigeria), without opportunistic infection. Exclusion criteria: pregnancy. Number randomised: 84 (intervention group 42, control group 42); attrition: not described. Gender split: 26 (31%) males, 58 (69%) females. Age: mean (SD) years: intervention group 33.8 (7.7) years; control group 35.3 (10.2) years. Nutritional status: BMI (kg/m²) at baseline: intervention group 23.1 in males and 21.9 in females; control group 23.3 in males and 20.3 females.
Interventions	Intervention: participants received dietary advice in the form of dietary counselling and individualised food prescriptions based on locally available food and counselling on food hygiene. Presription based on easily available and affordable foods, to contain all food groups and assessment of requirements. Control: participants received no dietary advice but the details were not described.
Outcomes	BMI, haemoglobin.
Publication details	Language: English. Funding: none declared. Publication status: peer reviewed journal.
Notes	Emailed authors for details of how randomisation carried out, any attrition, mean change in BMI and weight data.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Study described as quasi‐experimental. No information on random sequence generation in the paper. Quote: "Participants were selected using a simple random sampling technique and were then randomized" Additional information from authorsQuote: "Two nurses representing the intervention group and the control group respectively picked patients enrolment numbers from the ballot bag. All the patients enrolment numbers picked by the nurse representing intervention group automatically became the intervention group and the same with the control group." Therefore judged as low risk.
Allocation concealment (selection bias)	Low risk	No information on allocation concealment in the paper. Information provided by authors indicated that 'drawing lots' was used, therefore judged as high risk.
Blinding (performance bias and detection bias) Clinical outcomes	Low risk	Not described but assessment of clinical outcomes unlikely to be influenced by lack of blinding.
Blinding (performance bias and detection bias) Functional outcomes	Unclear risk	No functional outcomes assessed.
Blinding (performance bias and detection bias) Nutritional outcomes	Low risk	Not described but only BMI assessed which is unlikely to be influenced by lack of blinding.
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Not described and likely that researchers and participants were aware of group allocation as this was a nutritional intervention. It is unlikely that assessment of outcomes was influenced by lack of blinding.
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Not described but unlikely that assessment of the 2 outcomes reported would be influenced by lack of blinding.
Incomplete outcome data (attrition bias) All outcomes	Low risk	No attrition, (information provided by authors). Quote: "We did not record any attrition by death or withdrawal from study or lost to follow up. Patients were ambulatory patients and so not chronically ill. There was available grant to trace lost to follow up patients to their homes by treatment supporters, so it was easy to follow up patients even when they did not come to clinic on their monthly appointments".
Selective reporting (reporting bias)	Unclear risk	No protocol identified. All outcomes described in the methods reported in the results.
Other bias	Unclear risk	Baseline characteristics compared (gender, age and BMI). No differences noted between groups. No comparison of stage of disease and treatment characteristics which might have had an impact on the outcomes of interest.