NHS bodies throughout the UK collect a mass of data at all stages of the patient journey. From GP practices, which may hold a complete record of a patient’s care, including all contacts with health professionals, prescription information, and health outcomes; to hospitals, which capture information on the procedure a patient has had, which consultants were involved, what medicines were used and what follow-up care is necessary.
The possibilities that exist for healthcare benefit with big data are what make it so exciting. However, before we start thinking about the NHS as a big data warehouse of vast amounts of data waiting to be harnessed, it’s important to note that big data is not just about quantity; a crucial component is how we link to it. Key to this is being able to use data to match the health needs of patients with the care they receive or the medicines they are prescribed. An example of this can be seen when we look at prescribing rates. We know that in some parts of the country a person can access a medicine, whereas just a few miles away, another person can't. These regional variations do not, on their own, provide hard evidence of differences in access, as what patients need may vary considerably from region to region. Big data however offers the potential to directly link prescribing within target groups to the needs of that same population. This highlights how making full use of big data could provide an unparalleled source of hard facts to improve health care throughout the UK.
Improving patient care also means being able to understand how medicines are being used by the NHS. It is currently very difficult to know whether medicines are being used in line with NICE recommendations and we need to better understand if medicines usage is linked to the right health problem and, in turn, the right patient. Initiatives, such as those in cancer, suggest this could be achieved using big data. The National Cancer Intelligence Network’s registries for cancer accurately estimate how many people are living with different types of cancers, which could potentially link to information about outcomes and treatments.
The sharing of electronic information between GPs, hospital consultants and other health professionals via an appropriate unified electronic patient record is arguably the key to big data initiatives, as it is the basic building block upon which all subsequent analyses are based. It is encouraging to see some development in this area. In 2013 NHS England announced the ‘care.data’ initiative as a move to capture and link data from primary and secondary care. Health informatics and the eHealth agenda have been rapidly progressing in Scotland, with the Information Service Division (ISD) launching the electronic Data Research and Innovation Service (eDRIS) earlier this year. This is an analytics and consulting service giving access to a wide range of linked NHS data warehouses across both community and secondary care accessible through a secure ‘National Safe Haven’ system.
There is no question that big data analytics have huge potential to better match health care provision with health need. Perhaps the biggest practical challenge is how to use big data for the benefit of the whole population. Data sets that are currently collected are housed in information systems originally designed for clinicians managing individual patients, or indeed for managers to run hospitals. It is essential that we take full advantage of ever more accessible and cost-effective technology, to modernise what we currently do. We look forward to the debate on how we can best do this at the ABPI and NIHR Big Data conference in November.
The ABPI and National Institute for Health Research (NIHR) conference ‘360° of Health Data’ is on 21 November 2013.
To register or find out more about the conference, visit: www.abpievents.org.uk
Username: R&DConference, Password: November2013