On the 6th July, the publication of two separate reports into data sharing practices in UK led to the closure of care.data, an initiative established by the government to provide a nationwide database linking health and social care information from different healthcare settings.
The opportunities and challenges of real-world data in the UK, particularly those associated with the implementation of the care.data programme, were outlined in our recent commentary in BMC Medicine. However, in the wake of this closure, the government has stated it remains committed to realising the benefits of improved data sharing, and therefore the challenges and opportunities we outline remain pertinent to the on-going conversations surrounding the future of real-world data in the UK.
The first report, completed by the Care Quality Commission (CQC), provided several recommendations to improve data security standards across the breadth of the NHS. Indeed, as discussed in our piece, privacy fears overburdened the care.data programme from the outset. Concerns centred on the risk of data breaches, including those that may arise from deductive re-identification of anonymised or pseudo-anonymised data.
The CQC’s recommendations for security provisions around NHS data sharing practices may go some way to allaying fears, although adequately addressing privacy concerns will continue to reflect a key obstacle for any future real-world national dataset in the UK.
In her report, the National Data Guardian, Fiona Caldicott also addressed this issue, arguing an open dialogue with patients and health care professionals must be maintained to adequately communicate risks and acknowledge ongoing fears.
This dialogue will also focus on the implementation of a new opt-out model, where patients will be given the opportunity to choose if their data is shared for purposes outside their direct care. In our piece, we underline the substantial benefits of maintaining an opt-out versus opt-in approach.
Increasing public engagement with models of consent will hopefully operate to reduce the number of opt-outs which, as we discuss, has the potential to substantially influence the value of any subsequent database.
Making a UK national medical records database work
Real-world data is becoming increasingly important for healthcare decision making. A national medical records database covering the entire pathway of care has the potential to provide tremendous insights into disease epidemiology, as well as for understanding treatment safety and efficacy questions. This in turn can guide more effective service commissioning throughout the NHS.
Ultimately, better communication of the benefits and risks associated with data sharing on the scale intended by the now defunct care.data programme seems key. We know that adequately framing arguments around privacy concerns can lead to more positive attitudes, and that the more people know about how these data are used, the more accepting they are of data sharing.
Indeed, Nordic countries continue to show that national mandatory health registries can be implemented with great success. As work using the Clinical Practice Research Datalink has shown, this type of real-word data has the potential to make important advancements in medical research.
Effective communication and patient engagement will therefore be essential for the mistakes of care.data not to be repeated, and for the NHS and its patient community to reap the rewards of wide-scale data sharing.
There seems to be a huge confusion between sharing relevant information for patient care for specific purposes (e.g. referral for a hip replacement), sharing information (or making entire records available) for emergency care (e.g.various Inegrated Care models – where it is important that the information is up to date, so needs a ‘look-up’ function) and the care.data model which was never intended to be either related to direct care or be anything other than gaining access to cradle-to-grave records going back to 1948 and possibly further.
George Freeman was clear in his acceptance of the Review: removal of the Type 1 opt-out meant that the Care Data program could be abolished: the brand was toxic, and as the Review supported the removal of the power of patients to object to the upload of their medical records to HSCIC, unnecessary.
Identifiable – and annonymised – data is and always has been needed for managing the NHS and for research: how does this relate to the data-rape of uploading all medical records in their entirety to HSCIC and accepting the assurances of The Powers That Be that your confidentiality won’t be at risk?
Patients and the public were reassured as recently as September 2013 (by the Secretary of State for Health, no less) that if they didn’ want their identifiable information uploaded to HSCIC, all they had to do was to tell their GP and their wishes would be respected.
A week may be a long time in politics, but medical records may extend over 100 years..
But Care.Data was about selling medical records to companies.
It was not about Research, that was the PR cover story.