Carlson 2007.

Study characteristics
Methods	6‐week, multi‐centre, double‐blind, parallel, RCT with 2 arms: atomoxetine + MPH atomoxetine + placebo RCT was preceded by a 4‐week, open‐label, atomoxetine and placebo phase, during which participants who had adequate response were removed
Participants	Number of participants screened: not stated Number of participants included: met inclusion criteria 25; phase 1 (4‐week open‐label atomoxetine and placebo phase) 24 (20 boys, 4 girls); phase 2 (6‐week, double‐blind RCT) 17 Number of participants followed up: MPH 8, placebo 7 Number of withdrawals: MPH 1, placebo 1 Diagnosis of ADHD: DSM‐IV (combined (79% in phase 1)) Age: phase 1 mean 9.6 years (range 6‐12) IQ: > 70 Stimulant‐naive: 4% Ethnicity: white (83%), other (17%) Country: USA Setting: outpatient clinic Comorbidity: ODD (50% in phase 1) Comedication: no Other sociodemographics: none Inclusion criteria 6‐12 years of age DSM‐IV diagnosis of ADHD, any type Rating on ADHD‐RS‐IV, Parent Version ‐ Investigator Administered and Scored Version of ≥ 1.5 SD above age and sex norms Severity rating of at least moderate on the CGI‐S History (preceding 12 months) of insufficient response to an adequate stimulant trial, which was defined as gradual titration of stimulant medication for ≥ 2 weeks at specified doses for each medication. Inadequate response was determined by the child's prescribing physician, who also documented his or her opinion that a change in treatment was needed Participants must be of normal intelligence, as assessed by the investigator (i.e. without a general impairment of intelligence, and likely, in the investigator's judgement, to achieve a score ≥ 70 on an IQ test) (Administration of a formal IQ test is not an entry requirement for this trial. Specific learning disabilities are not considered general impairments of intelligence) Participants must be able to swallow capsules Exclusion criteria Weighed < 22 kg or > 60 kg at trial entry Had any other Axis I diagnosis, including pervasive developmental disorder, mood or anxiety disorder (presence of comorbid ODD was not an exclusion criterion) Bipolar disorder, autism Any medical conditions that would contraindicate the use of atomoxetine or ER‐MPH History of any seizure disorder and/or Rolandic seizures (other than febrile seizures) or prior ECG abnormalities in the absence of seizures, or history of taking (or are currently taking) anticonvulsants for seizure control History of severe allergies to > 1 class of medication or multiple adverse drug reactions, including hypersensitivity to MPH History of intolerance or non‐response to atomoxetine Used any concomitant psychotropic or excluded medications Ingestion of any excluded medications 5 days before baseline ratings and randomisation
Interventions	Participants were randomly assigned to ER‐OROS‐MPH or placebo Co‐intervention: atomoxetine Number of participants randomly assigned: MPH 9, placebo 8 MPH mean dosage: 1.02 mg/kg Administration schedule: once/d Duration of intervention: 6 weeks Titration period for MPH: after randomisation to target dose of 1.08 mg/kg/d (max 1.2 mg/kg/d ) Treatment compliance: not stated
Outcomes	ADHD symptoms ADHD‐RS‐IV, Parent Version ‐ Investigator‐Administered and scores: parent‐rated CPRS Revised, Short Form: parent‐rated at baseline and at weeks 4, 5, 6 and 10 General behaviour Weekly parent ratings of evening and morning behaviour: parent‐rated at baseline and at weeks 4, 5, 6 and 10 Non‐serious AEs Vital signs, at each visit Weight, at each visit Spontaneous AE reports, at each visit
Notes	Sample calculation: included sample (25 participants) is too small in relation to the sample calculation (85 participants) Ethics approval: yes Comment from trial authors The present findings can be applied only to children with an inadequate stimulant response. Given the small trial sample, particularly in the combination treatment groups, the findings must be considered preliminary Key conclusion of trial authors Methylphenidate appears to be safely combined with atomoxetine, but conclusions are limited by small sample Comment from review authors For the review, we used data from trial phase 2, that is, the RCT with 2 arms (intervention: MPH + atomoxetine; control: placebo + atomoxetine) Exclusion of MPH non‐responders/children who have previously experienced AEs while taking MPH before randomisation: no, but exclusion of atomoxetine/placebo responders before trial phase 2 Withdrawals due to adverse events: yes, 2 Funding source: research was funded by Eli Lilly and Company, Indianapolis, Indiana Email correspondence with trial authors. June‐November 2013. We received from trial authors and sponsoring pharmaceutical company, supplemental information regarding IQ and blinding procedures, as well as data on weight, treatment‐emergent AEs and ECG
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Randomly assigned via an interactive voice response system to receive ER‐MPH or placebo
Allocation concealment (selection bias)	Low risk	Investigators and participants were blinded to the precise visit at which randomisation to blinded placebo or MPH occurred, as this visit is not identified in the investigator’s copy of the protocol, and as use of blinded trial drug begins at visit 2 and continues up to visit 8
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
Incomplete outcome data (attrition bias) All outcomes	High risk	Efficacy outcome measures were conducted on the ITT sample by using an LOCF method Selection bias (e.g. titration after randomisation → exclusion of MPH non‐responders or placebo responders): yes, 1 participant discontinued due to perceived lack of efficacy
Selective reporting (reporting bias)	Low risk	No indication of selective reporting