This project offers a rare opportunity to analyse millions of real-world data points from apps which have collectively been used by over two million patients and clinicians worldwide. Drawing on log-level data from Bright Therapeutics’ platform – Recovery Record (for eating disorders) and Recovery Path (for substance use disorders) – the candidate will apply advanced statistical, machine learning, and causal-inference approaches to uncover how digital interventions drive clinical outcomes.
The datasets include high-dimensional behavioural data points such as self-monitoring logs, feature usage, messaging between patients and clinicians, and symptom-change trajectories. These rich, longitudinal data open the door to:
– Mapping engagement archetypes and trajectories across diverse patient and clinician cohorts.
– Developing predictive models for treatment retention, remission, and dropout.
– Evaluating engagement features (e.g., check-ins, therapeutic tool use, clinician–patient exchanges) as mediators of outcomes.
– Translating insights into adaptive intervention strategies and product refinements for next-generation digital therapeutics.
By integrating cutting-edge analytics with clinically validated digital platforms, this project bridges behavioural, data, and computer science to improve real-world mental health outcomes. The work will generate high-impact publications and inform both scientific understanding and app innovation at Bright Therapeutics.