Garfinkel 1983.

Study characteristics
Methods	Double‐blind, randomised, placebo‐controlled, cross‐over experiment with 4 arms: MPH clomipramine desipramine placebo Trial lasted 20 weeks; first and last 2 weeks were baseline periods during which no medication was given. Different interventions lasted 3 weeks each (Monday to Friday) with a 7‐day washout between changes in drugs
Participants	Number of participants screened: not stated Number of participants included: 12 (all boys) Number of participants followed up: 12 Number of withdrawals: 0 Diagnosis: DSM‐III Age: mean 7.3 years (range 5.9‐11.6) IQ: > 70 MPH‐naive: 50% Ethnicity: not stated Country: USA Setting: outpatient clinic and patient ward. 8 participants were day hospital patients and 4 were inpatients Comorbidity: no Comedication: no comedication during trial period Other sociodemographics: children presented with remarkably similar clinical, family and educational histories. None of the children had localising neurological signs or met criteria for other psychiatric diagnoses Inclusion criteria ADD according to DSM‐III Exclusion criteria None mentioned directly
Interventions	Participants were randomly assigned to the different possible drug condition orders of MPH, clomipramine, desipramine and placebo Mean MPH dosage: 18 mg/d Administration schedule: twice/d, morning and lunch Duration of each medication condition: 3 weeks Washout before trial initiation: 2 weeks before trial entry and 7 days between drug conditions Titration period: first week of the drug condition after randomisation Treatment compliance: not stated
Outcomes	ADHD symptoms CTRS: rated daily by care workers (in settings away from the classroom) and teachers (in classroom) General behaviour Werry‐Weiss‐Peters Activity Rating Scale: rated daily by parents for day patients and by evening child care staff for fully hospitalised children Non‐serious AEs DBP and SBP, measured by the nurse while participants were sitting and standing at 09.00 am and 12.00 pm Apical pulse, morning and afternoon
Notes	Sample calculation: no information Ethics approval: no information Key conclusions of trial authors MPH was significantly better for improving classroom and behavioural manifestations of ADD as compared with placebo, desipramine and clomipramine Tricyclic antidepressants may have a significant therapeutic effect as indicated by mood elevation and amelioration of non‐classroom behaviour in the evenings Results encourage further clinical and pharmacological investigation of ADD using various alternative treatments Comment from review authors Only information and data extracted from the MPH group and the placebo group 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: Ontario Mental Health Foundation Email correspondence with trial authors. October 2013. We obtained supplemental information regarding number of dropouts and SD for CRS. We also wanted other data but were not able to obtain these, as the trial took place several years ago and the data are no longer available
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Children were randomly given MPH, desipramine, clomipramine, placebo. A Latin square was followed to control for the order of presentation of drugs
Allocation concealment (selection bias)	Low risk	Children were randomly given MPH, desipramine, clomipramine, placebo. A Latin square was followed to control for the order of presentation of drugs
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Participants, parents, attending physicians, teachers, nursing staff and child care workers did not know which medication the child received, ensuring the double‐blind procedure. Medication was added to lactose powder and was placed in identical‐appearing gelatin capsules
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Participants, parents, attending physicians, teachers, nursing staff and child care workers did not know which medication the child received, ensuring the double‐blind procedure. Medication was added to lactose powder and was placed in identical‐appearing gelatin capsules
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	No information 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 reporting bias