NHS Digital has lofty ambitions when it comes to its data sharing roadmap
More “granular” and standardised patient information could pave the way for seamless data-sharing
Data sharing within the NHS has always been a contentious topic, with its long history of failed projects and scandals anchored on breaching patient privacy. Take the DeepMind controversy, where sensitive patient information was handed to Google's AI research division without their consent. And who, of course, can forget care.data.
NHS Digital's latest data sharing initiative hinges on standardising the data that is collected from patients, as well as vastly raising the level of detail in this data. Its vision, according to consultation papers, is to clean up the data that is collected, and open this up to commissioners, researchers, and public health planners. The effort will require collective input and collaboration not just across the wider NHS, but suppliers, regulators, and the Department for Health and Social Care (DHSC).
Currently, the Data Coordination Board oversees an extensive list of all information standards and data collection instances within the health service. This ranges from the 'healthy child programme' to the 'systemic anti-cancer therapy data set'. This is a monumental amount of data, much of which is segmented in different silos and not standardised in a meaningful way.
It's an eye-watering amount of data, as anybody can see, but these plans would still kick it up a notch. Clinicians and others will be expected to collect "granular" levels of data about their patients using the SNOMED CT standard, and link this with other NHS data to support direct care. This process comprises the first phase of the journey.
This SNOMED CT standard is a worldwide process for recording highly detailed clinical information about a patient that would allow data to be shared across the health and social care landscape. It was rolled out in a phased approach in primary care units from last year, with secondary care units expected to use the system by 1 April 2020. This would also pave the way for advanced analytics to improve the quality of healthcare outcomes.
Phases two and three, meanwhile, will involve first improving the insights that can be extracted from SNOMED CT coded patient data, and secondly to establish the workflows that can channel this data into a format accessible by external parties such as researchers.
But there are several causes for concern with these plans, specifically with regards to trust in the NHS for handling such sensitive and detailed patient information. NHS Digital maintains that it publicised the consultation widely, sharing this in its bulletins and with relevant stakeholders appropriately, although others disagree this was extensive enough. It opened in February, before shutting quietly last Wednesday.
Secondly, trust, or at least the perception of trust, is a major issue, as NHS Digital's CEO Sarah Wilkinson herself outlined recently, according to Digital Health. Speaking at an AI in healthcare event earlier this month, Wilkinson warned that a lack of clarity around the use of data was slowing down progress, and that a lack of trust in the system could have a dramatic impact on research. Ultimately, she reiterated the need to "directly address people's concerns by laying out our ethical approach to dealing with data and providing absolute clarity on how we intend to use health data".
For medConfidential's Phil Booth, meanwhile, the vision itself at the heart of the plans is "deeply concerning", with the campaign group's coordinator also suggesting there were enough loopholes for the data to make its way into the hands of commercial bodies. He also argued that if people knew about the extent of the NHS' data-sharing plans, the number of people using NHS Digital's opt-out tool could exceed 25% of patients. This would be enough to render the dataset useless.
"Let me be very clear, everything stored in your medical record is stored as a dated, coded event. What was very striking, was the clear intention for the coded data to be flowed not just between clinicians or for the use of someone's direct care, but into other uses up to right at the top; policymakers," he told IT Pro.
"This is precisely what the whole debate was around with care.data back in 2014, which is taking the rich, detailed data from GP records, and taking it and flowing it for secondary uses. This could include commissioning, as well as planning and audit.
There is, of course, sound reasoning for any perceived distrust between the wider public and the NHS when it comes to safeguarding patient data. Despite pushing a relatively optimistic outlook, the past has been blighted with issues. For example, there are a number of blunders that come to mind, including one case of 150,000 patients' data being shared against their expressed consent, not to mention the time researchers learned the NHS lost almost 10,000 patient records in 2017.
How NHS Digital's opt-out mechanism, which has been in force since last year, will affect its ambitions is also yet to be seen. The consultation document failed to reference whether this factors into consideration nor did it make any mention of patient consent in general.
"No new data will ever be collected or processed by NHS without going through rigorous, public processes to ensure that the collection has a clear legal basis and is proportionate in scope," an NHS Digital spokesperson said.
"Our Data Processing Services Programme is separate from any proposals around the type of data that we collect, but will improve the way that we collect, store and process data, in line with our statutory functions."
The roadmap to improving data-sharing standards within the NHS has been laid out in some detail, with a clear vision too. But there remains a sea of factors all parties involved must tackle if these plans stand any chance of success.
Security analytics for your multi-cloud deployments
IBM Security QRadar SIEM solution briefDownload now
Five reasons to move to the cloud
Join the enterprises moving their workloads to the cloudDownload now
Architecting hybrid IT and edge for digital advantage
Why business leaders should consider a hybrid IT strategyDownload now
Six reasons to accelerate remote asset monitoring with AI
How to optimise resources, increase productivity, and grow profit margins with AIDownload now