Gadow 1990.

Study characteristics
Methods	6‐week double‐blind, cross‐over trial in which participants received the following in random order: LD‐MPH MD‐MPH placebo Furthermore, a single case report from the trial is described
Participants	Number of participants screened: not stated Number of participants included: 11 (11 boys, 0 girls) Number of participants followed up: between 9 and 10, depending on ratings Number of withdrawals: 1‐2 Diagnosis of ADHD: DSM‐III (type not stated) Age: mean not reported (range 5.9‐11.9 years) Case report: 10 years of age IQ: > 75 Stimulant‐naive: 6 Ethnicity: not stated Country: USA Setting: outpatient clinic Comorbidity: disruptive behaviour disorder Comedication: not reported Other sociodemographics: children represented a full range of socioeconomic backgrounds   Inclusion criteria Boys 5‐12 years of age ADHD diagnosis according to DSM‐III Scored above research cut‐off (> 7) on the Aggression scale of IOWA CTRS, or were considered aggressive by their classroom teacher and were above cut‐off on ≥ 1 other conduct problem scale Scored ≥ 15 on the Abbreviated Teacher Rating Scale Above cut‐off on the Oppositional DIsorder or CD Index of the Parent or Teacher Version of the Stony Brook Child Psychiatric Checklist, Version 3 Exclusion criteria IQ < 70 Psychosis, pervasive developmental disorder, dangerous to self or others Seizure disorder, major organic brain dysfunction, medical illness, contraindication to medication treatment
Interventions	Participants were randomly assigned to 1 of 6 possible drug condition orders of LD‐MPH (0.3 mg/kg) and MD‐MPH (0.6 mg/kg) (upper limit 25 mg) and placebo Mean MPH dosage: not stated Administration schedule: twice/d, morning, noon, 3.5 h apart, 7 d/week Duration of each medication condition: 2 weeks Washout before trial initiation: yes Titration period: when MD was not preceded by LD condition, the child was gradually built up to MD. Data from these titration days were excluded from the analyses Treatment compliance: pill count, no further information
Outcomes	ADHD symptoms Conners' Abbreviated Teacher Rating Scale Conners' Abbreviated Parent Rating Scale Non‐serious AEs Stimulant Side Effects Checklist: rated systematically
Notes	Sample calculation: no information Ethics approval: no information Comment from trial authors Sample size was small, which decreased the probability of detecting clinically significant group differences between treatment conditions Key conclusion of trial authors Results of this trial indicate that MPH‐induced improvements in the classroom behaviour of aggressive hyperactive boys are associated with concomitant changes in the demeanour of classmates sitting in close proximity to the drug‐treated child 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: Ciba Pharmaceutical Company supplied MPH placebo Email correspondence with trial authors: July 2013. Emailed first trial author twice to get additional information (funding, ethics approval, etc.) and data from the trial. Trial authors not able to provide us with additional data
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Dose schedules were assigned on a random basis
Allocation concealment (selection bias)	Low risk	Medication and placebo pills were identical and were dispensed to parents and school nurses in dated, sealed envelopes
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Parents, teachers, observers, treating physicians and children were blinded to dose and order
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Parents, teachers, observers, treating physicians and children were blinded to dose and order
Incomplete outcome data (attrition bias) All outcomes	Low risk	For classroom analyses, data on 10 boys were analysed, and for lunch room analyses, data on 9 boys were analysed Selection bias (e.g. titration before randomisation → exclusion): no
Selective reporting (reporting bias)	Unclear risk	No protocol identified. Email sent to first trial author. No answer; therefore not able to get information