Phiri 2012 MWI.

Methods	Trial design: Randomized double‐blind, placebo‐controlled trial Multicentre trial: Yes Trial duration: June 2006 to August 2009
Participants	Recruitment: Children were recruited from four hospital in southern Malawi: Queen Elizabeth Central hospital (Blantyre), Chikwawa District hospital, Thyolo District hospital, Zomba Central hospital. Inclusion criteria: Age: 4 to 59 months Anaemia: admitted with and treated for severe malarial anaemia Convalescent children surviving hospital stay (received transfusion and completed the course of intravenous quinine) with Hb > 5g/dL Weight > 5 kg Able to switch to oral medication Sitting unaided Exclusion criteria: Sickle‐cell anaemia Blood loss due to trauma Haematological malignancy Known bleeding disorder Hypersensitivity to AL Treatment with AL within a week of admission Non‐residency in trial area Previous participation in the trial Participation in another clinical trial Known need for medication prohibited during the intervention period Surgery scheduled during the trial Other co‐morbidities: 8% of children were infected with HIV Sample size: 1431 randomized, 1414 allocated to groups; analysed 4310
Interventions	Presumptive treatment for all participants: Six doses of AL as part of the standard 3 day course in hospital. Children < 15 kg received one tablet; children > 15 kg received 2 tablets, once every 12 hours for 3 days. Total number of intervention groups: 2 AL Placebo Dose and timing of intervention: AL or placebo: Children < 15kg received one tablet; children > 15kg received 2 tablets, once every 12 hours for 3 days at 1 month and 2 months post discharge Duration of intervention period: 3 months Place and person delivering intervention: The first daily dose (both at 1 and 2 months pd) was given by trial team member who visited the home every morning for 3 day The second dose was left with the parent or guardian to give in the evening (adherence assessed the following morning) Co‐interventions: ITN use: similar in both groups; 35% used treated net; 16% used untreated net; and 48% used no net (self‐reported) Other: none Additional treatments: Rescue treatment for acute malaria Discharge ‐ month 3: oral quinine for 5 days Month 4 to 6 (extended follow‐up period): AL Treatment for severe malaria Discharge – month 3: intravenous quinine and oral quinine for 5 days Month 4 to 6 (extended follow‐up): intravenous quinine and AL Children with severe disease were admitted to hospital Children with recurrent severe anaemia received blood transfusions Bacterial and other infections treated as per physician’s discretion Co‐interventions equal in each arm? (if not, describe): Yes
Outcomes	Primary outcome: Composite outcome of all‐cause mortality and hospital readmission because of all‐cause severe anaemia (Hb < 5 g/dL or clinical indication for blood transfusion) or severe malaria (readmittance to hospital due to confirmed malaria treated with parenteral quinine) between 1 and 6 months Secondary outcomes: All‐cause mortality Hospital readmission because of all‐case severe anaemia or severe malaria All‐cause hospital admission All‐cause sick child clinic visits Clinic visits because of microscopically confirmed non‐severe malaria Measurement time points: Observation time divided into 3 periods: Discharge ‐ month 1 pd (before IPT) 1 to 3 months pd (IPTpd period) 4 to 6 months pd (extended follow‐up period) All children seen at 6 months for assessment of Hb and malaria parasitaemia Passive case detection from discharge to month 6 How were outcomes assessed? 6 month follow‐up by passive case detection: mothers were asked to bring children to a study clinic if they had fever or were unwell. Standardized forms used to record information; Hb, temp and malaria smear
Notes	Country: Malawi Setting: Urban/peri‐urban/rural Transmission area: Perennial Source of funding: Netherlands African Partnership for Capacity Development and Clinical Interventions against Poverty‐related Diseases; UBS Optimus foundation; Gates Malaria Partnership. Conflict of interest stated: Authors declared no conflict of interest
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Computer‐generated list of random numbers, "stratified by hospital and weight group (< 15 kg and 15 kg or more) in randomly varying block sizes of two, four, or six".
Allocation concealment (selection bias)	Low risk	Sequentially numbered envelopes containing AL or placebo.
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Authors did not mention that placebo and AL were identical tablets, but described the trial as "double‐blind".
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Masking was maintained and the code only broken once all data sets were closed.
Incomplete outcome data (attrition bias) All outcomes	Low risk	All participants accounted for. Loss to follow‐up rates similar across groups (7% at 6 months).
Selective reporting (reporting bias)	Low risk	All pre‐specified outcomes reported on.
Other bias	Low risk	No.