Barkley 1989b.

Study characteristics
Methods	Triple‐blind, randomised, cross‐over trial with 3 interventions placebo LD‐MPH (0.30 mg/kg) HD‐MPH (0.5 mg/kg)
Participants	Number of participants screened: not stated Number of participants included: 83 (71 boys, 12 girls) Number of participants followed up: 80 Number of withdrawals: 3 Diagnosis of ADHD: DSM‐III‐R (subtype distribution not stated) Age: mean 8.2 years (range 5‐13) IQ: mean 105.1 MPH‐naive: 85% Ethnicity: not stated Country: USA Setting: outpatient clinic Comorbidity: not stated Comedication: not stated Additional sociodemographics: mothers, married (n = 48), divorced (n = 13), unmarried or widowed (n = 13) Inclusion criteria 6‐13 years of age ADHD according to DSM‐III‐R Complaints from teacher, parents or both, of significant inattention, overactivity and impulsivity Appearance of symptoms before 7 years of age Symptoms for 12 months Above the 93rd percentile of the hyperactivity scale on parent or teacher report forms of the CBC) Simple language IQ score > 80 on the Peabody Picture Vocabulary Test ‐ Revised Exclusion criteria Gross sensory or motor deficits Tic disorders, Tourette's syndrome or an immediate family history of such Seizures Gross brain damage Autism Thought disturbance or schizoid, schizotypal or frank psychotic features Significant cardiac problems or high BP Excessive levels of anxiety or fear Levels of depression that exceed or equal the problems of ADHD Most participants (n = 74) were subdivided into children placed 2 SD above the normal mean on the Aggressive scale of the parent form of the CBCL (T score > 70) and those who did not have ≥ 2 SD above the normal mean to form aggressive (n = 37) and non‐aggressive subgroups (n = 37). Groups did not differ in age, years of education, maternal age or maternal years of education
Interventions	Participants were randomly assigned to 1 of 6 possible drug condition orders of 0.3 mg/kg MPH, 0.5 mg/kg MPH and placebo Mean MPH dosage: not stated Administration schedule: morning and noon Duration of each medication condition: 7‐10 days Washout before trial initiation: not stated Medication‐free period between interventions: not stated Titration period: none Treatment compliance: unused capsules returned to the clinic each week for adherence check No family was discontinued from the trial because of non‐compliance with the drug regimen, defined as more than 1 day of failure to take medication, or 2 missed capsules/week
Outcomes	ADHD symptoms CPRS ‐ Revised: parent‐rated at the end of each drug condition CTRS ‐ Revised: teacher‐rated at the end of each drug condition Child Attention Profile: teacher‐rated at the end of each drug condition General behaviour Home Situations Questionnaire: parent‐rated at the end of each drug condition (7 to 10 days) School Situations Questionnaire: teacher‐rated at the end of each drug condition (7 to 10 days) Adverse effects Barkley Side Effects Rating Scale: parent‐ and teacher‐rated at the end of each drug condition (7 to 10 days)
Notes	Sample calculation: not stated Ethics approval: trial was approved by the Human Subjects Committee at the medical centre Comments from trial authors Limitations Rates of side effects were based on a large sample that was screened before admission to the drug trial Low doses of medication to detect side effects Use of rating scale rather than direct behavioural observation Key conclusions of trial authors In their drug responding, aggressive and non‐aggressive participants were quite similar. The few exceptions involved measures of conduct, on which aggressive participants were initially rated as more extreme and subsequently showed a greater degree of improvement from medication than non‐aggressive participants With this dose range, stimulants result in few/mild side effects; systematic monitoring of side effects suggested before/during clinical trials of stimulants Comment from review authors The Barkley Side Effects Rating Scale was labelled generically as a “behaviour questionnaire” to disguise its intended use as a monitoring tool for potential side effects. The purpose was to prevent prejudice on the part of respondents, who might potentially distort their ratings if they knew these could be side effects of the medication Exclusion of MPH non‐responders/children who have previously experienced AEs while taking MPH before randomisation: no Any withdrawals due to AEs: yes; 3. One child discontinued because of nervous facial tics, headache and dizziness; a second as the result of excessive thinking and disjointed thinking (during HD‐MPH); and a third because of headache, dizziness and increased hyperactivity Funding source: trial was internally funded by the medical school Email correspondence with trial authors: July 2013. We obtained additional information regarding funding and ethics approval. Unfortunately, it was not possible to receive from the trial authors supplemental data on ADHD symptoms and general behaviour
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Trial used a completely counterbalanced design, with participants randomly assigned in relatively equal numbers to 1‐6 possible drug conditions
Allocation concealment (selection bias)	Unclear risk	No information
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Both medication and placebo were crushed and placed within orange opaque gelatin capsules to disguise distinctive differences in flavour between medication and placebo and dose differences across conditions. Children and their parents and teachers, as well as the research assistant evaluating the children, were blinded to medication conditions
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Children and their parents and teachers, as well as the research assistant evaluating the children, were blinded to medication conditions
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	No information on incomplete outcome data Selection bias (e.g. titration after randomisation → exclusion of MPH non‐responders or placebo responders): no
Selective reporting (reporting bias)	Low risk	No indication of selective reporting