Only one-third of individuals treated for psychiatric disorders like depression benefit from first-line therapies, while the rest experience limited efficacy or significant side effects (Williams & Hack, Nat. Med. 2020). This inefficiency arises largely because psychiatric medications are developed and prescribed using a “one-size-fits-all” approach, despite clear evidence that clinical symptoms and treatment responses depend heavily on individual factors such as sex, ethnic background, age, genotype, comedications, comorbidities, disease progression, and environmental parameters.
This project aims to leverage the rich dataset from UK Biobank, which includes a wide array of variables with potential impacts on psychiatric treatment outcomes. By analyzing these demographic, clinical, and genetic data, the goal is to uncover how these factors influence drug efficacy and safety, determine their relative importance, and develop predictive models that inform personalized prescribing practices. These models will enable clinicians to select appropriate drugs, determine optimal dosing, and refine other prescription modalities, thereby reducing reliance on trial-and-error approaches.
Given the chronic nature of psychiatric disorders, the extended time required to find effective treatments places a heavy burden on patients and healthcare systems. By identifying and modeling patient-specific determinants of drug response, this project will advance precision psychiatry, significantly reducing delays in achieving symptom remission, improving treatment outcomes, and minimizing adverse effects. This work aligns with global trends in precision medicine and has the potential to transform psychiatric care, making it more personalized, efficient, and effective.