Ramtvedt 2013.

Methods	Multi‐center, randomized, placebo‐controlled, cross‐over trial Country: Norway Number of study sites: 4 Statistical methods: unclear
Participants	Sample size: 36* children/adolescents with an ADHD diagnosis according to DMS‐IV‐TR criteria Dropouts: 0 Psychiatric comorbid disorders: oppositional defiant disorder, anxiety/depression, learning disability Age range: 9 years to 14 years Mean age (SD): 11.4 (1.4) years Sex: 29 (81%) males ADHD subtype: 10 (28%) inattentive; 1 (3%) hyperactive ‐ impulsive; 25 (69%) combined
Interventions	Three interventions (all 36 children/adolescents participated in each of the three interventions): Dextroamphetamine (short acting), 5 mg twice a day in week 1; 10 mg twice a week in week 2, (n = 36) Methylphenidate (short acting): 10 mg three times a day in week 1; 15 mg twice a day, 10 mg once daily in week 2, (n = 36) Placebo (n = 36) Duration: 42 days (3 x 14‐day treatment periods)
Outcomes	Relevant outcomes: Improvement of ADHD symptoms, assessed with a 21‐item ADHD questionnaire developed for the study by parents and teachers** Adverse events Other outcomes: Inattention, assessed with Conners' Continuous Performance Test Motor activity, assessed with Conners' Continuous Performance Test
Notes	ClinicalTrials.gov identifier: NCT01220440 Authors' affiliation: university and hospital Study funding: public funds Outcomes were presented across two publications *Data only provided on those children/adolescents who completed the trial; no information provided regarding number of children/adolescents randomized **Data is presented as an aggregate score of parent and teacher ratings and therefore could not be incorporated into the meta‐analysis
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Method of sequence generation not described
Allocation concealment (selection bias)	Unclear risk	Method of allocation concealment not described
Incomplete outcome data (attrition bias) All outcomes	High risk	Primary analysis only included a subset of completers without any reason provided. No information on non‐completers provided
Selective reporting (reporting bias)	High risk	Although the registered protocol stated that they would collect adverse events using the Side‐Effects Rating Scale, this was not reported in the published manuscript
Other bias	Unclear risk	No information on the validity of the primary outcome measure provided
Blinding of participants and personnel (performance bias) All outcomes	High risk	Interventions and placebo were not identical
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Blinding of outcome assessment not described