16 June 2017
Research using de-identified health data, generated from electronic health records, prescription records and laboratory public health surveillance systems for example, has great potential to benefit society. In England, our wealth of routine health data sources can help us to understand causes of disease, monitor disease trends, assess the effectiveness of treatments and screening programmes and plan healthcare services.
The sheer quantity of routine health data and that it comes from real people living in real environments provides a depth and breadth of information that it is not possible to obtain using data from clinical trials. Secure linkage between health datasets offers a further opportunity to improve the richness of data available for research.
But are we making the most of these data sources for research? Despite much hype and huge investments in “big data”, we are far from realising the anticipated revolution in personalised healthcare. The PHG Foundation has previously focused on ways to support responsible data sharing for clinical genomics to improve the safety and quality of health services.
In Linking and sharing routine health data for research in England we consider the barriers and facilitators to accessing and using de-identified health datasets for research gleaned from interviews with data users. Interviewees were researchers from different organisations and different disease areas who ranged from post-doctoral level to professors and directors of organisations.
Interviewees pointed to substantial challenges in obtaining valid research results from routine health datasets – challenges which cross areas from organisational culture to governance, communication and the need for specialist technical expertise. These, together with detailed recommendations to facilitate research by improving systems, incentives and capacity to enable secure data sharing, are discussed in our forthcoming report, to be published shortly
In these uncertain political times, it is vital to maintain the momentum of UK initiatives to support effective research using health and biomedical data to improve population health - for example the multi-funded MRC research institute, Health Data Research UK (HDR-UK). This is all the more pressing as Brexit starts to exert its effects on our access to European datasets and Europe-wide plans for data harmonisation and sharing. The health of the public should not suffer because of political and other barriers to the access and use of routine health data for research.