Current treatment guidelines for Attention-Deficit/Hyperactivity Disorder (ADHD) rely heavily on pharmacological treatments but do not incorporate individualised treatment plans. Treatments are applied in a ‘one size fits all’ manner, relying on a time-consuming trial-and-error process to find optimal treatment regimens. One third of individuals do not show an adequate response or experience adverse events but there is limited knowledge on the individual factors that might indicate a likelihood of poor response to ADHD medications.
PhD aims
– Identify key moderators and potential mediators from Individual Participant Data (IPD) trial meta-analysis examining treatment response to stimulant and non-stimulant medication in children and young people with ADHD.
– Adapt existing structured and natural language processing phenotyping to characterise these moderators/mediators from within the electronic health care records (EHR) of over 9000 young people with ADHD to inform potential prediction models.
– Develop a digital platform prototype using prediction models that could guide treatment selection in clinical practice and improve outcomes in ADHD.
– Examine how novel wearable data (activity and sleep data), integrated into the EHR, may enhance prediction.
– Provide the student with high quality training in quantitative methods, academic writing and presentation, and translation of findings to maximise impact.
Timeline
Year 1-2: Project 1- Secondary data analyses on data from recently collected repository of n>5000 individual child participant data level from pharmacological trials testing the effectiveness of ADHD medications; identifying patient-level social, emotional and cognitive moderators, and treatment-related mediators, of clinically meaningful response.
Year 2-3: Project 2 – Translate the mediation/moderation variables from Project 1 to data extraction protocols from SLaM records (n>8000 young people with ADHD). Develop transparent prediction models to determine effectiveness of treatment of specific stimulant and non-stimulant formulations in real world community ADHD treatment setting.
Year 3-4: Project 3 – Undertake workshops with young people, parents, and clinicians, and work with CAMHS Digital Lab experts, to examine the usability of risk calculators within existing dashboards available to clinicians, specifically to examine the potential utility of Project 2 to aid medication optimisation in young people with ADHD.
Year 4: Project 4 – Examine movement and sleep data pre-collected by CAMHS Digital Lab PACES+ actigraphy devices (n=80), using data nested in electronic healthcare records, to explore whether these objective measures improve the prediction of future ADHD treatment response trajectories for these young people. Thesis submission and research dissemination.