Clinical Informatics and Dementia Research in the United Kingdom

Everyday in every clinical contact between doctors, nurses and other health workers and their patients data is collected, stored and used for decision making about healthcare. But until now using that information for research has been relatively limited. Today with the increasing use of electronic medical records that information can be used – to evaluate treatments, to investigate risk factors and, at least potentially, to help in the recruitment to investigational studies to test new therapies. However, in order to utilize this data – for dementia research and other indications - there are challenges, both technical and those of governance. 

I will describe a system that has met these challenges in the UK, the Case Records Interactive Search system, that enables access to substantial numbers of case records for research. With deep anonymisation of data, access to both encoded and narrative data and the use of natural language processing to enable data recovery from the narrative data together with a service user led governance structure, the CRIS system is not only delivering larger data sets than can be realistically generated from research but is increasingly being used to audit and improve health care delivery. The CRIS system is being linked to one of the worlds largest longitudinal cohort studies through the UK Dementias Platform.

In Europe more generally the IMI European Medical Information Framework has similar ambitions but includes aggregation of and access to huge data sets from epidemiological cohort studies, through cohort studies to clinical databases. Aggregating data through access to meta-data as well as individual-level data the EMIF programme together with the UK Dementias Platform will link to, indeed are both essential components of, the European Prevention of Alzheimer’s Disease platform; an ambitious plan to develop preventative therapies for Alzheimer’s.

Such ambitious plans for large scale multi-national trials can only be achieved at scale and at realizable cost through public-private partnership and with the utilization of existing clinical information currently all too often underutilized in parent cohorts and clinical datasets.