Heriot 2008.

Study characteristics
Methods	3‐month, parallel trial with 4 arms MPH and parent training programme MPH and supportive non‐training parent group Placebo and parent training Placebo and supportive non‐training parent group
Participants	Number of participants screened: 93 Number of participants included: 20‐26 (13 boys, 3 girls) (6 withdrew, but unclear whether before or after random assignment) Number of participants followed up: MPH + parent training 4, MPH + no training 4, placebo + training 4, placebo + no training 4 Number of withdrawals: not stated Diagnosis of ADHD: DSM‐IV (combined (25%), hyperactive‐impulsive (56%), inattentive (19%)) Age: mean 4.78 years (range 3‐6) IQ: mean 97 (range 80‐123) MPH‐naive: 100% Ethnicity: not stated Country: New Zealand Setting: outpatient clinic Comorbidity (type: ODD 5/31%) Comedication: no Other sociodemographics: No significant differences in baseline demographics were noted between the 4 intervention groups. Inclusion criteria Between 3.0‐5.9 years of age Resident with a primary caregiver for ≥ 6 months Meet diagnostic criteria for ADHD as defined in DSM‐IV Features of ADHD had to be present for ≥ 12 months and to a degree that was considered to be developmentally inappropriate and functionally inappropriate and functionally impairing across settings Above the 93rd percentile on the Global Index subscale of the CRS Exclusion criteria Currently in hospital Currently in another treatment trial Currently receiving treatment Full scale IQ < 80 Pervasive developmental disorder or psychosis Major neurological or medical illness that would interfere with participation or require medications incompatible with MPH Chronic serious tics or Tourette’s disorder History of child abuse Inability of parent to understand English
Interventions	Participants were randomly assigned to MPH + parent training, MPH + no parent training, placebo + parent training or placebo + no parent training Number randomised to each group: not stated Mean MPH dosage: 0.3 mg/kg Administration schedule: twice/d ‐ morning and lunchtime Duration of intervention: 3 months Titration period: dosage was built up over the 1st week Treatment compliance: not stated
Outcomes	ADHD symptoms ADHD‐RS‐IV Parent and Teacher versions: data not reported in an useable form ‐ not possible to obtain data from trial author CPRS ‐ Revised, Long Version: no data provided in the article CTRS ‐ Revised, Long Version: no data provided in the article Non‐serious AEs Stimulant drug, Barkley Side Effects Rating Scale
Notes	Sample calculation: no Ethics approval: yes; obtained from the University of Waikato, Department of Psychology Ethical Review Committee and the Waikato Ethics Committee for Waikato Hospital Comments from trial authors 7 of the 16 children no longer met DSM‐IV (1994) diagnostic criteria for ADHD at the end of the trial Ethnicity did not appear to affect outcomes, although it was unclear whether the content of the programme was equally appropriate for all ethnic groups Limitations Small number of participants Of those willing to participate, only 43% met eligibility criteria for the trial No follow‐up data Key conclusion of trial authors Children were more likely to improve when treatment involved ≥ 1 active component (medication or parent training). However, variability in individual parental and child participant responses to all treatment conditions was notable, indicating the importance of interaction between treatment variables and other factors. Findings are discussed within the framework of a transactional model, and inferences are drawn about limitations of the idea that a "best treatment" exists that is universally applicable. Although improvement among children receiving treatment with MPH was greater than for others, it is notable that all treatments were associated with improvement in some children. Exclusion of MPH non‐responders/children who have previously experienced AEs while taking MPH before randomisation: no Any withdrawals due to AEs: unclear Funding source: no funding to conduct the trial was received from any party Email correspondence with trial authors: January, February and May 2014. Emailed trial author but have not received a reply
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	High risk	Participants were allocated to conditions sequentially, using RAND function (SPSS) to generate the sequence. Each parent was randomly assigned to attend the training programme group or the support group.
Allocation concealment (selection bias)	Unclear risk	Not stated
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Throughout the period of data collection, participants and therapist were blinded to medication status
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Throughout the period of data collection, participants and therapist were blinded to medication status. All parents and teachers were blinded to the intervention
Incomplete outcome data (attrition bias) All outcomes	High risk	Only data for completing participants were reported Selection bias (e.g. titration after randomisation → exclusion): unclear. 4 were excluded after randomisation because of "treatment integrity problems"
Selective reporting (reporting bias)	Unclear risk	We were unable to identify a protocol