Horn 1991.

Study characteristics
Methods	12‐week, double‐blind, randomised, parallel trial with 6 arms: placebo LD‐MPH HD‐MPH Placebo + behavioural parent training + child self control instruction LD‐MPH + behavioural parent training + child self control instruction HD‐MPH + behavioural parent training + child self control instruction 9‐month follow‐up
Participants	Number of participants screened: 117 Number of participants included: 107 (83 boys, 24 girls) Number of withdrawals after randomisation: 18 Number followed up: at end of trial (12 weeks) 78, 9 months after termination of trial 71 Diagnosis of ADHD: DSM‐III‐R (subtype not stated) Age: mean 8.27 years (range 7‐11) IQ: > 70 MPH‐naive: not stated Ethnicity: white (84.9%), African American (9.4%), Hispanic (3.8%), Asian American (1.9%) Country: USA Setting: outpatient clinic Comorbidity: CD (7.5%), disruptive oppositional disorder (15%), ODD (43%) Comedication: not stated Other sociodemographics: mean yearly family income USD 25,019. No significant differences in baseline demographics were noted between treatment groups or between treatment groups and treatment dropouts in any of the following parameters: child's age, grade, sex, IQ; parental marital status, annual family income and maternal education and age. However, a significantly greater number of non‐white children were included in the placebo‐alone condition than in remaining treatment conditions or among treatments dropouts Inclusion criteria 7‐11 years of age Exact agreement between a licensed clinical psychologist and a board certified paediatrician with respect to diagnosis of ADHD according to DSM‐III‐R criteria Score on Hyperkinesis Index of the CPRS or CTRS ≥ 2 SD above published means Current psychostimulant therapy required to be withdrawn for the course of the trial Exclusion criteria Comorbid anxiety, depressive disorder or both Gross physical impairments, intellectual deficits or psychosis in child or parents
Interventions	Participants were randomly assigned to 1 of 6 treatment conditions, including placebo, LD‐MPH (0.4 mg/kg) and HD‐MPH (0.8 mg/kg) No of participants assigned to each group: 16 to each arm Administration schedule: daily Duration of intervention: 12 weeks Washout before trial initiation: 2 weeks before initial diagnostic evaluation Titration period: no Treatment compliance: 87.39% of parents in the non‐placebo medication conditions group reported anonymously that their child took the medications almost every day, whereas the remainder of families reported usage an average of 3 to 4 days a week. Stimulant medication was withdrawn immediately after post‐test assessments
Outcomes	ADHD symptoms All dependent measures were administered at pre‐test, post‐test (within 1 week or 2 weeks of the end of the treatment phase of the trial) and follow‐up CBCL SNAP CTRS AEs Medication side effects, monitored by a board certified paediatrician
Notes	Sample calculation: no information Ethics approval: no information Exclusion of MPH non‐responders/children who have previously experienced AEs while taking MPH before randomisation: no Any withdrawals due to AEs: yes, 3 Funding: not declared Email correspondence with trial authors: August 2013. Not possible to receive supplemental information or data through personal email correspondence with trial authors. They do not recommend inclusion of the trial in this review because of problems with the design and methods used at the time the trial was carried out (Ramstad 2013a [pers comm])
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	2 of 12 families were randomly assigned to each of the 6 treatment conditions. This procedure was repeated 8 times, so that by the end of the trial, 16 families were randomly assigned to each of the 6 treatment conditions, for a total of 96 families
Allocation concealment (selection bias)	Unclear risk	No information
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Double‐blind
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Double‐blind. Note that assessors were blinded to treatment status at all times
Incomplete outcome data (attrition bias) All outcomes	Low risk	Analyses were conducted on 4 samples: (1) participants completing the entire trial and followed up at 9 months (n = 71), (2) participants completing the entire trial (n = 78), (3) participants completing the entire trial + participants completing ≥ 6 weeks (n = 90) and (4) all of the above + those who dropped out immediately after initiating treatment (n = 96). When no post‐test and/or follow‐up data were available for dropouts, pre‐test values or post‐test values, or both, were substituted. Given that analyses produced essentially the same results, only the analyses that include completers are reported Selection bias (e.g. titration after randomisation → exclusion): no
Selective reporting (reporting bias)	Unclear risk	No protocol identified