Computational pharmacoepidemiology
Computational pharmacoepidemiology
Led by Dr Stella Talic, Dr Harvey Koh and Professor Dragan Gasevic
Computational pharmacoepidemiology is an evolving field that integrates epidemiological methods, clinical pharmacology, and advanced digital technologies to study the use, safety, and effectiveness of medicines in real-world populations. It leverages large-scale health data, such as electronic health records (EHRs), claims databases, registries, and patient-generated data, to generate actionable insights that inform clinical practice, regulatory decisions, and public health policy.
Traditional clinical trials often lack the scale, diversity, and duration needed to detect rare or long-term adverse drug events. Digital pharmacoepidemiology fills this gap by enabling:
- Post-marketing surveillance of medication safety
- Real-world evidence generation for regulatory and clinical decision-making
- Rapid signal detection for emerging risks
- Personalized medicine through predictive analytics
Key technologies
- Artificial Intelligence (AI) & Machine Learning (ML): Used for pattern recognition, risk prediction, and natural language processing of clinical notes.
- Large Language Models (LLM): Using Mixture of Experts (MoE) models to grade drug-drug interaction evidence.
- Deep Learning (DL): Using DL to create digital twins for the
- Cloud-based platforms & Clean Rooms: Enable secure, privacy-preserving analysis of sensitive health data.
- Real-World Data (RWD): Includes EHRs, pharmacy records, genomic data, and social determinants of health.