Large Scale Infrastructure for Health Data Analytics

The opportunities for data analytics to inform the science and practice of health care are growing. For example, in the UK, 65 million citizens have lifelong e-health records that can be used to examine patterns of disease, treatment and outcomes. Similarly the real-world impact of interventions such as new drugs can be evaluated in these records. Such approaches need to solve issues around information governance, confidentiality, understanding data and provenance and developing methods for big data. Here we describe a large-scale service that addresses the opportunities brought by the availability of large-scale e-health records. The service has been used to support 50 research projects in the UK across a wide range of scientific areas and can be seen as an exemplar for the developing field of health data analytics.