Rapport 1985.

Study characteristics
Methods	Triple‐blind, randomised, cross‐over trial with 4 interventions: placebo 5 mg MPH 10 mg MPH 15 mg MPH Phases Week 1: baseline assessment Weeks 2‐5: cross‐over trial
Participants	Number of participants screened: 22 Number of participants included: 14 (12 boys, 2 girls) Number of participants followed up: 11 Number of withdrawals: 1 Number of participants excluded during the trial: 2 Diagnosis of ADHD: DSM‐III (subtype not stated) Age: 8.3/7.75 years (range 6‐10) IQ: > 100 MPH‐naive: not stated Ethnicity: white (86%), not stated (14%) Country: USA Setting: outpatient clinic Comorbidity: not stated Comedication: no Other sociodemographics: low to middle socioeconomic status Inclusion criteria DSM‐III diagnosis of ADHD by the child's paediatrician and the clinic's directing clinical psychologist Maternal report of a developmental history consistent with ADD‐H (Barkley 1981) Maternal rating of ≥ 2 SD above the mean for the child's age on the Werry‐Weiss‐Peters Activity Scale Teacher rating > 15 on the Abbreviated CTRS Children showing a favourable response to MPH (≥ 25% mean increase in classroom on‐task behaviour and decrease ≥ 2 SD on the Abbreviated CTRS (compared with baseline levels) during any of the active medication weeks) Performance on the Matching Familiar Figures Test characteristic of a "fast‐inaccurate or impulsive" responder (i.e. faster than average responses and higher than average error rates for the child's age) Exclusion criteria Any gross neurological, sensory or motor impairment Those currently taking medication Those classified as non‐responders, defined as neither improved (MPH‐induced facilitation ≥ 25%) on the Paired Associates Learning test nor declined in relation to baseline and placebo
Interventions	Participants were randomly assigned to 1 of 24 possible drug condition orders of fixed doses of MPH (5 mg, 10 mg, 15 mg) and placebo. As a result of the small sample size, not all drug orders were used. Post hoc comparison showed that doses were distributed approximately equally across different positions in the sequence Administration schedule: once daily, in the morning Duration of each medication condition: 7 days Washout before trial initiation: none, but all weekly dosage changes occurred on Saturdays to control for potential rebound effects Titration period: none Treatment compliance: both used and unused envelopes with capsules were returned to control for medication compliance. No results regarding compliance were reported
Outcomes	ADHD symptoms Abbreviated CTRS: teacher rated symptoms each Friday, in the morning, or at the end of each treatment condition Dosage changes occurred on Saturdays No useable data
Notes	Sample calculation: not stated Ethics approval: not stated Key conclusion of trial authors Results of the present investigation demonstrate a functional relationship between psychoactive medication and impulsivity, attention and behaviour of children with ADD in classroom settings Comment from review authors Although trial authors refer in 1 of the articles to the other 2 articles as a recent trial and a past investigation, it seems to us that these articles discuss the exact same trial Inclusion of MPH responders only/exclusion of MPH non‐responders/children who have previously experienced AEs while taking MPH before randomisation: no Any withdrawals due to AEs: no Funding source: not declared Email correspondence with trial authors: July 2013. Trial authors informed us that original records were shredded after 20 years, and that they could not provide the requested data (Ramstad 2013c [pers comm])
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Order of drug administration was determined by randomly assigning each child to 1 of 24 possible drug disorders
Allocation concealment (selection bias)	Low risk	MPH was packaged by the pharmacist in coloured gelatin capsules to avoid detection of dose and taste. Capsules were placed in individually dated envelopes to ensure accurate dose administration
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Triple‐blind design (children, teachers and observers were blinded)
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Teachers and observers were blind to when medication was administered and what specific doses were given
Incomplete outcome data (attrition bias) All outcomes	High risk	Selection bias (e.g. titration after randomisation → exclusion of MPH non‐responders or placebo responders): yes, data from 2 of 12 children in the present investigation were not included in the statistical analyses because they were classified as non‐responders according to results of the Paired Associates Learning test
Selective reporting (reporting bias)	Unclear risk	Not possible to receive a copy of the protocol