This case study was originally posted in the OGP Global Report.

The National Health Service (NHS) remains an immensely popular part of the public welfare system in the UK, but most citizens also agree that it could be more effective and efficient. Citizens want to know their options for the best possible care and some budget-minded politicians seek to cut costs and identify a role for private providers.

Since OGP’s founding, the UK has put improving health outcomes and efficiencies at the center of its OGP efforts. Indeed, making health outcome data usable by the public has been a major initiative between 2011 and 2016 in UK OGP action plans.

The first and second UK OGP action plans aimed to address these issues by publishing open data and seeking feedback from the public. The efforts, in some respects, have been clear successes. In other respects, the NHS stumbled upon a much more complex set of issues around the limits of transparency.

The Reforms

Improving healthcare was a central part of the UK story in OGP. Indeed, at the initial launch of the Partnership, the UK touted the release and impact of clinical performance data in improving surgical outcomes. The first OGP action planAction plans are at the core of a government’s participation in OGP. They are the product of a co-creation process in which government and civil society jointly develop commitments to open governmen... set up ministry-level “transparencyAccording to OGP’s Articles of Governance, transparency occurs when “government-held information (including on activities and decisions) is open, comprehensive, timely, freely available to the pub... More councils” which had various stakeholders (providers, members of the public, and officials) to identify high-priority datasets to release. The second action plan, beginning in 2013, had two commitments that aimed to improve NHS performance. They covered:

• Publication of clinical and other performance indicators,
• Implementation of a patient recommendation tool, and
• A “Patient-Centred Outcome Measurement” (PCOM) tool on services available for rare and complex medical conditions.

The Results

Many of the outcomes from each of the initiatives were significant. Others provide insight into future reforms. They include:

• Clinical performance data: Twelve core clinical datasets and data from general practice settings, adult social care, and patient-centered outcome measurements were published. Overall, more than 7,000 new datasets on health and clinical and performance data are now available on an easily searchable data platform. In March 2015, the NHS reported that it had achieved coverage of 97% of practices and clinics. NHS England was working toward “better open data.”
• Social care data: Progress on the release of information about social care services was slower. The government “proposed to re-baseline the ambition to achieve 8,750 by April 2015 and 10,000 by April 2016,” through work with partners like home care providers.
• Public feedback: The Friends and Family Test, piloted in 2013, was rolled out across all care settings. The published aggregate data included over five million pieces of feedback. The NHS Insight Team is using this data to feed into service improvement. NHS England claims the changes have helped drive up healthcare standards leading to “many improvements, large and small, across the country,” but there was no clear evidence for how these changes fed into broader improved standards.

The Setback

Care.data was a cornerstone of this bundle of reforms. It aimed to centralize patient data through the General Practice Extraction Service which covers more than 9,000 primary care clinics in the UK. Patients would need to opt out of their data being shared, otherwise it would be shared with the public database.

Immediately, the program met with resistance, eventually leading to its closure. A broad range of concerns included:

• Medical professionals concerns: In 2014, 40% of general practices opted out of the scheme due to lack of confidence in the program’s ability to maintain their standard of “total confidentiality.”
• Civil society concerns: Within OGP, the involved civil society organizations refused to engage with the commitment, as they shared the privacy concerns of medical staff. Namely, members of the public were concerned about: (1) how researchers would access data; (2) whether data would or could stay anonymous; and (3) whether the public could properly consent to their data being used.
• Review and concern: In June 2015, the Major Projects Authority gave care.data the lowest possible rating, “red,” meaning the project seemed “unachievable.” This was, in part, due to administrative issues, but also due to the major outcry over privacy concerns.
• Report and closure: In 2016, Dame Fiona Caldicott, the national data guardian, issued a report on the patient privacy in the scheme (referred to as “Caldicott 3” after its 1998 and 2013 predecessors), which confirmed privacy concerns, and the program closed immediately. The report confirmed that there had been sale of private data for years.

Care.data lessons

For other countries looking into opening health data, the lessons of care.data are three-fold. It seems that, however controversial, there were adequate checks in place to ensure accountability in the end.

• Protections of privacy: There needed to be adequate protections ex-ante to ensure the protection of personal data, and assurances of appropriate re-use.
• Deliberation and trust: There was little involvement of the public or service providers in the discussion of how such data might be properly anonymized. The “Caldicott 3” report identified this as a major solution to restoring trust and seeing if this could go forward at all.
• Public watchdogs: The Major Projects Authority’s annual report as well as the National Data Guardian’s report both provided public opportunities for a reckoning with the considerable risks and costs of these major data initiatives.

Photo Credit: Colin & Linda McKie, Adobe Stock