Davis 1996.

Methods	Design: RCT Study dates: not mentioned Malaria transmission pattern and local antimalarial drug resistance: not applicable Adverse event monitoring: daily self‐reported diary. Three medical check ups for laboratory and other tests
Participants	Number enrolled: 106 randomized, 95 completed all study procedures Inclusion criteria: "healthy adult staff and students at teaching hospitals in Perth, Western Australia" Exclusion criteria: "Those with a past history of psychiatric conditions, or neurological, cardiac, hepatic or renal disease were excluded, as were pregnant or breastfeeding females and those with a known allergy to, or taking medication known to interact with quinolone drugs. None of the subjects had taken mefloquine in the 3 months before the study" Country of recruitment: Australia Country of malaria exposure: not applicable Duration of follow up: 7 weeks Type of participants: non‐immune non‐travellers
Interventions	1. Mefloquine (1 x 250 mg tablet), with placebo dose followed 1 week later by 250 mg mefloquine weekly, active treatment duration 4 weeks 2. Placebo, 1 tablet weekly, duration 5 weeks
Outcomes	Included in the review: 1. Measure of adherence to the drug regimen 2. Adverse events: other outcome measures (symbol digit modalities test, digit span forwards and backwards test, ECG, hearing loss at 6kHz) Outcomes assessed not included in the review: 3. Laboratory tests: serum glucose, insulin, ionized calcium, phosphate, magnesium and albumin concentrations 4. Adverse events: headache, lethargy, abdominal pain, diarrhoea, cough, nausea; study reports events occurring in the first week (after both groups had received placebo) and the relative risk of symptoms worsening over time
Notes	Funding sources: "We thank… F. Hoffman La Roche & Co. for financial support"
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	"...allocation… was by a random number code generated by independent Fremantle Hospital Pharmacy staff"
Allocation concealment (selection bias)	Low risk	"...who kept the code strictly confidential until the last volunteer had completed the protocol"
Blinding of participants and personnel (performance bias) Adverse effects/events	Low risk	"Tablets were prepared in individually numbered but otherwise unlabelled containers... identical placebo tablets…"
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	"Allocation of active or placebo formulation was by a random number code generated by independent Freemantle hospital staff who kept the code strictly confidential" Comment: not mentioned whether outcomes assessors were blinded
Incomplete outcome data (attrition bias); efficacy	Unclear risk	N/A
Incomplete outcome data (attrition bias); safety	Low risk	"Of 106 randomised volunteers, 95 (90%) completed all study procedures... eight subjects withdrew after initial assessment and three after the second. Follow‐up of these individuals revealed no toxicity in those allocated mefloquine"
Selective reporting (reporting bias); efficacy	Unclear risk	N/A
Selective reporting (reporting bias); safety	High risk	Comment: not all symptoms were reported, only those occurring in > 10% of participants in both groups. Absolute numbers of participants experiencing each symptom after mefloquine/placebo commenced not provided, only relative risk of symptoms worsening over time
Other bias	High risk	"We thank… F. Hoffman La Roche & Co. for financial support"