Gehring 2012a

Methods	Randomised controlled trial, three arms, open‐label, stratification by tumour location.
Participants	Inclusion criteria: age > 18; KPS > 70; primary brain tumour; subjective complaint of cognitive decline or fatigue; being considered for stimulant therapy by their neuro‐oncologist; the ability to speak and understand English or Spanish. Exclusion criteria: current use of psychostimulants, monoamine oxidase inhibitors, anticoagulants, drugs similar to erythropoietin, or illicit drugs; history of hypersensitivity reaction to methylphenidate or modafinil; history of uncontrolled seizures, cardiac or pulmonary disease, or hypertension (systolic > 140 mm Hg, diastolic > 90 mm Hg, or not on a stable dose of anti‐hypertensive medication for the past month; severe headaches; current glaucoma, narcolepsy, Tourette’s syndrome, major psychiatric diagnosis, alcohol or drug abuse; current use of herbals/supplements for fatigue relief, e.g. Ginkgo biloba, ginseng, St John's Wort, dehydroepiandrosterone; unstable dose of antidepressants; comorbidities or medications that in the treating physician’s opinion could potentially interfere with safe administration of MPH or MOD. No. randomised: methylphenidate: 24; modafinil: 10. Follow‐up: 4 weeks. Setting: one cancer centre in the United States.
Interventions	Arm I: 10 mg of oral methylphenidate (immediate release) taken in divided doses for 4 weeks Arm II: 18 mg or oral methylphenidate (sustained release) taken in the morning for 4 weeks Arm III: 200 mg of oral modafinil taken in the morning for 4 weeks.
Outcomes	Cognitive function (WAIS‐III Digit span and Digit symbol, TMT, HVLT‐R, grooved pegboard, MAE COWA). Fatigue (BFI, POMS‐Fat, POMS‐Vig sub‐scales) Sleep disturbance (BSDS) Mood (POMS, BDI‐II, STAI) QoL (FACT general and brain modules, FIM)
Notes	Arm I and II combined for statistical analysis. Second drug used as control arm, rather than placebo. Groups also compared with normative data.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	"Patients were stratified by tumour location (i.e., right verses left hemisphere) and randomly assigned to one of three conditions." "a computer performed the randomization" (obtained via correspondence).
Allocation concealment (selection bias)	High risk	An open‐label design was used.
Blinding of participants and personnel (performance bias) All outcomes	High risk	An open‐label design was used.
Blinding of outcome assessment (detection bias) All outcomes	High risk	An open‐label design was used.
Incomplete outcome data (attrition bias) All outcomes	Low risk	Missing outcome data were balanced in numbers across treatment groups, and reasons for missing outcome data similar between treatment groups and unlikely to be related to outcomes measured.
Selective reporting (reporting bias)	Low risk	All outcomes reported.
Other bias	Unclear risk	Possible recruitment bias: "PBT patients were considered eligible for participation if… being considered for stimulant therapy by their neuro‐oncologist."