Optimising medication management in children and young people with ADHD using a computerised test (QbTest): a feasibility randomised controlled trial.

Background: Medication for attention deficit hyperactivity disorder (ADHD) should be closely monitored to ensure optimisation. There is growing interest in using computerised assessments of ADHD symptoms to support medication monitoring. The aim of this study was to assess the feasibility and acceptability of a randomised controlled trial (RCT) to evaluate the efficacy of one such computerised assessment, the Quantified Behavior (Qb) Test, as part of medication management for ADHD.

Methods: This feasibility multi-site RCT conducted in child and adolescent mental health and community paediatric settings recruited participants aged 6-15 years diagnosed with ADHD starting stimulant medication. Participants were randomised into one of two arms: experimental (QbTest protocol) where participants completed a QbTest at baseline and two follow-up QbTests on medication (2-4 weeks and 8-10 weeks later) and control where participants received treatment as usual, including at least two follow-up consultations. Measures of parent, teacher, and clinician-rated symptoms and global functioning were completed at each time point. Clinicians recorded treatment decision-making and health economic measures were obtained. Data were analysed using multi-level modelling and participants (children and parents) and clinicians were interviewed about their experiences, resulting data were thematically analysed.

Results: Forty-four children and young people were randomised. Completion of study outcome measures by care-givers and teachers ranged from 52 to 78% at baseline to 47-65% at follow-up. Participants reported the questionnaires to be useful to complete. SNAP-IV inattention scores showed greater reduction in the intervention than the control group (- 5.85, 95% CI - 10.33, - 1.36,). Engagement with the intervention ranged from 100% at baseline, to 78% follow-up 1 and 57% follow-up 2. However, only 37% of QbTests were conducted in the correct time period. Interview data highlighted that the objectivity of the QbTest was appreciated by clinicians and parents. Clinicians commented that the additional time and resources required meant that it is not feasible to use QbTest for all cases.

Conclusion: The trial design and protocol appear to be feasible and acceptable but could be improved by modifying QbTest time periods and the method of data collection. With these changes, the protocol may be appropriate for a full trial. Adding QbTest may improve symptom outcome as measured by SNAP-IV.

Trial Registration: ClinicalTrials.gov, NCT03368573 , prospectively registered, 11th December 2017, and ISRCTN, ISRCTN69461593 , retrospectively registered, 10th April 2018.