Barkley 1991.

Study characteristics
Methods	Triple‐blind, placebo‐controlled, cross‐over design with participants randomly assigned to the following conditions 3 doses of MPH (5 mg, 10 mg and 15 mg twice/d) placebo Each intervention period lasted 1 week
Participants	Number of participants screened: not stated Number of participants included: 40 Number of participants followed up: 40 (36 boys, 4 girls) Number of withdrawals: 0 Diagnosis of ADD: DSM‐III‐R (with hyperactivity: 58%; without hyperactivity: 42%) Age: mean 8.6 years (range 6‐12) IQ: 103.5 MPH‐naive: not stated Ethnicity: not stated Country: USA Setting: outpatient clinic Comorbidity: borderline and low internalising symptoms. No others stated Comedication: not stated Other sociodemographics: 54.8 on Hollingheads Two Factor Index Participants were divided into different categories : Based on type of ADD ADD with hyperactivity (n = 2) ADD without hyperactivity (n = 17) Based on severity of internalising symptoms by maternal ratings on the Internalising scale of the CBCL High internalising group (> 70) (n = 12) Borderline internalising group (65‐70) (n = 17) Low internalising group (< 65) (n = 11) Inclusion criteria Diagnosis of ADD(H) according to DSM‐III‐R IQ estimate of ≥ 80 on a standardised IQ test given within the past year or on the WISC‐R, given at trial screening Was the biological child of both current parents or had been adopted by them shortly after birth (within the first year) No evidence of deafness, blindness, severe language delay, cerebral palsy, epilepsy, autism or psychosis, as established through medical history, parental interview and child play diagnostic interview Additional criteria for children with combined ADHD Teacher complaints of short attention span, impulsivity and overactivity as revealed by parent reports Duration of 6 months for these problems Age of onset of these problems before 7 years Score > 93rd percentile on the Inattention and Overactivity scales of the Child Attention Problems Rating Scale No history of treatment with stimulant drugs, or, if such a history, physician consent to stop taking medication for 48 h before evaluation in the trial Additional criteria for children with ADD‐H Same criteria as for children with combined ADHD, with the exception of the 4th criterion. Instead, score > 93rd percentile on the Inattention scale of the Child Attention Problems Rating Scale, but a score < 84th percentile on the Overactivity scale of the Child Attention Problems Rating Scale Differences regarding the 2 ADHD groups They differed significantly on child’s IQ score, with the ADD without hyperactivity group scoring significantly lower than the ADD with hyperactivity group. Child IQ was correlated with all of the dependent measures (measures that we used in this review were not affected by differences in IQ) Exclusion criteria History of tics or Tourette's syndrome, given the controversy over whether stimulants may create or exacerbate these conditions Those with a history of cardiac surgery, high BP or cerebral vascular accident, given the known cardiac pressor effects of stimulants Those with a history of adverse reactions to a stimulant
Interventions	Participants were randomly assigned to 1 of 3 possible drug condition orders of MPH (5 mg, 10 mg and 15 mg) and placebo Administration schedule: twice/d, morning and noon Duration of each medication condition: 1 week Washout before trial initiation: no Titration period: none, but highest dose was never given unless preceded by moderate dose Compliance: children were permitted to miss 1 day of medication over 7 days and still remain in the trial. No families were removed from this trial because of non‐compliance as defined in this way
Outcomes	ADHD symptoms Parent‐rated Attention Deficit Hyperactivity Disorder Rating Scale (ADHD‐RS): rated at the end of each drug condition order Home Situations Questionnaire: number of problem settings and mean severity of problems, rated at the end of each drug condition order Teacher‐rated (teachers completed questionnaires at the conclusion of each medication condition) Self Control Rating Scale Child Attention Problems School Situations Questionnaire: number of problem settings and mean severity of problems Non‐serious AEs Home Side Effects Rating Scale: parent‐rated at the end of each drug condition order School Side Effects Rating Scale: parent‐rated at the end of each drug condition order
Notes	Sample calculation: no Ethics approval: trial was approved by the Institutional Review Board at the medical centre Key conclusion of trial authors This trial indicates that ADHD, inattentive type, and ADHD, combined type, do not show dramatic differences in their manner of responding to MPH across 3 dose levels (5 mg, 10 mg and 15 mg), with both groups displaying generally positive drug responses. However, more children with ADHD, inattentive type, had minimal or no response or did best on the low dose of medication, whereas the vast majority of children with ADHD, combined type, showed a positive response, primarily to moderate to high doses of MPH Exclusion of MPH non‐responders/children who have previously experienced AEs while taking MPH before randomisation: yes; excluded those with history of adverse reactions, but included both naive and prior users of antipsychotics Any withdrawals due to AEs: no Funding source: NIMH Email correspondence with trial author: 18 January 2013. Dr. Barkley informed us that data from the trial on side effects, for example, are no longer available
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Randomly assigned to 1 of 6 possible drug conditions
Allocation concealment (selection bias)	Low risk	Hospital pharmacy prepared placebo (lactose powder) and MPH by crushing and placing them into 6 orange opaque gelatin capsules
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Triple‐blind design
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Triple‐blind design
Incomplete outcome data (attrition bias) All outcomes	Low risk	No incomplete outcome data Selection bias (e.g. titration after randomisation → exclusion of MPH non‐responders or placebo responders): no
Selective reporting (reporting bias)	Unclear risk	Protocol not available