Sharp 1999.

Study characteristics
Methods	Cross‐over trial with 3 interventions: MPH dextroamphetamine placebo Phases 3‐week medical‐free baseline, randomised administration of MPH, dextroamphetamine and placebo followed by breakfast and lunch daily Stepwise increase in stimulant dose each week Each phase lasted 3 weeks Dosage range: MPH 10 mg/d‐70 mg/d, dextroamphetamine 5 mg/d‐30 mg/d
Participants	Number of participants screened: 150 (girls) Number of participants included: 42 girls and 56 comparison boys Number of participants followed up: 42 Number of withdrawals: 1 girl given placebo at each phase Diagnosis of ADHD: DSM‐IV (subtype not stated) Mean age: 9.1 years (range 6.0‐12.7) Mean IQ: boys 109.3, girls 105.2 MPH‐naive: Some of the participants had no previous experience with stimulants. No information on exact number. Ethnicity: girls: white (67%), African American (19%), Hispanic (14%); boys: white (73%), African American (21%), Hispanic (4%), Asian (2%) Country: USA Setting: outpatient clinic, laboratory classroom setting Comorbidity: 71% boys, 69% girls; ODD (33% boys, 50% girls), CD (7% boys, 2% girls), major depression (0% boys, 7% girls), separation anxiety (0% boys, 2% girls), specific phobias (0% boys, 7% girls), trichotillomania (2% boys, 0% girls), tic disorders not otherwise specified (13% boys, 2% girls), enuresis (18% boys, 12% girls), reading disorder (5% boys, 8% girls) Comedication: not stated Other sociodemographics: socioeconomic status mean score 48.0 ± 25.8 (girls), 52.4 ± 26.9 (boys) Inclusion criteria Girls with a history of severe hyperactivity, impulsivity and inattentiveness that interfered with home and school functioning Symptoms of ADHD present in ≥ 2 settings Conners' Hyperactivity factor scores from home teachers were ≥ 2 SD greater than age and sex norms Exclusion criteria Full‐scale IQ < 80 on WISC‐R Chronic medical or neurological disease, including Tourette’s disorder and chronic tic disorders
Interventions	Participants were randomly assigned to 1 of 6 possible drug condition orders of MPH, dextroamphetamine and placebo Sharp 1999 Mean MPH dosage: week 1 mean dose 0.45 mg/kg, week 2 mean dose 0.85 mg/kg and week 3 mean dose 1.28 mg/kg Administration schedule: breakfast and lunch ‐ 5 days per week administered by nurse; administered by parent on weekends Elia 1991 (secondary reference under Sharp 1999): Mean MPH dosage: week one 0.9 mg/kg, week two 1.5 mg/kg and week three 2.5 mg/kg Administration schedule: 9:00 am and 1:00 pm throughout the entire week (7 days) Duration of each medication condition: 3 weeks Washout before trial initiation: 3‐week medication‐free period before trial Medication‐free period between interventions: none Titration period: none Treatment compliance: not stated
Outcomes	ADHD symptoms Conners' Hyperactivity and Conduct factors score: teacher‐rated Wender Utah Rating Scale (WURS): parent‐rated Abbreviated CTRS: completed weekly CTRS: completed weekly Serious AEs Mean beneficial and adverse effects of dextroamphetamine and MPH were nearly identical for all ratings, including ratings of appetite problems. However, objectively verified significant decreases in body weight (drug main effect, F 10.27; P = 0.0002) were significantly greater for dextroamphetamine (mean change −1.1 ± 1.0 kg from baseline; P = 0.02) than for MPH (−0.4 ± 1.1 kg; not significant) Stimulant‐related AEs and body weight General behaviour CBCL: parent‐rated Woodstock‐Johnson Achivement Battery: observer (psychologist)‐rated Teachers Report Form (TRF): teacher‐rated Conners' Parent Questionnaire (CPQ): completed weekly by parents Children’s Psychiatric Rating Scale (CPRS): completed weekly by psychiatric staff Activity monitor (measuring truncal motor activity): from 9:00 am‐4:00 pm Non‐serious AEs Subject Treatment Emergent Symptom Scale (STESS): physicians and parents Some items on the CPRS, such as nervous mannerisms and obsessive thinking: rated as AEs (Elia 1991, secondary reference under Sharp 1999) 24‐h urine and blood studies: at baseline and at third week for each treatment
Notes	Sample calculation: no Ethics approval: no information available Key conclusion of trial authors Sharp 1999: "Our primary conclusion is that our sample of girls demonstrated very similar patterns of comorbidity and impairment and identical patterns of drug response" Comment from review authors We need a lot of information on the girls Inclusion of MPH responders only/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: not declared Email correspondence with trial authors: June 2014. We received supplemental information from Dr. Castellanos
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Order of medication (MPH, placebo or dextroamphetamine) was randomly assigned on the basis of a table administered by the research pharmacy. Dose escalation was fixed, with 2 ranges, depending on body weight (> or < 30 kg). 3 weeks (1 dose per week) in each phase. For most children, this was followed by random assignment to pemoline vs placebo after another washout period
Allocation concealment (selection bias)	Unclear risk	No information
Blinding of participants and personnel (performance bias) All outcomes	Low risk	MPH, dextroamphetamine and placebo were packaged in identical capsules by the NIH pharmacy and were administered by NH nurses in double‐blinded randomly assigned order
Blinding of outcome assessment (detection bias) All outcomes	Low risk	All ratings were performed blind to treatment assignment
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	Did not describe what they did with the missing data Selection bias (e.g. titration after randomisation → exclusion): no
Selective reporting (reporting bias)	Unclear risk	No protocol published.