Single patient record data will be available to researchers by default
- 18 February 2025
- The development of a universal national patient record across health and care is set to be a central recommendation for the 10 year health plan, according to papers seen by Digital Health News
- A draft report from the digital and technology working group recommends the adoption of a single patient record to be accessed by health and care professionals, patients, policymakers and researchers
- It also calls for "a wholesale upgrade of technology foundations, progressive replacement of legacy systems, and transformational use of automation and AI to improve productivity and empower patients"
Exclusive: A universal patient record across health and care to be accessed by providers, patients, policymakers and researchers, will be a central recommendation in the NHS 10 year health plan, according to papers seen by Digital Health News.
A leaked draft report from the digital and technology working group, ahead of the plan which is expected to be published in spring 2025, outlines a vision for the NHS centred on digitally enhanced and inclusive care.
To achieve this, it calls for “a wholesale upgrade of technology foundations, progressive replacement of legacy systems, and transformational use of automation and AI to improve productivity and empower patients”.
The paper recommends the adoption of a “universal record used across all settings such that relevant personal information is accessible to involved workers with agreed levels of access”.
Patients, researchers and policymakers would also be able to access the record, with anonymised healthcare information “made available for research purposes by default, with opt out available by exception”.
The paper adds: “Providers will have access to a single patient record that includes all events and interactions that patients have with the health and care system and any other personal data captured by individuals that is relevant to their care.
“With all HCPs [health and care professionals] caring for a person working off the same set of facts, all members of a multidisciplinary team will be working from an agreed and continuously updated care plan, reducing duplication and miscommunication, and supporting coordinated care and improved clinical and social care outcomes.”
Wes Streeting, health secretary, first outlined plans for a single patient record, summarising patient health information, test results, and letters through the NHS App, in October 2024.
This was followed in January 2025 by proposals for a digital platform to allow medical information such as medication history, to be shared between NHS and care staff.
The draft paper extends this ambition, stating that the aim for a “universal record used across all settings such that relevant personal information is accessible to involved HCPs.”
This appears to resurrect the core centralising ambition for a national patient record under the failed NHS National Programme for IT (NPfIT), launched by the Labour government in 2002.
It is envisaged the single patient record would be able to connect to data from remote monitoring devices, such as blood pressure monitors, and patients would be able to self-record and submit their own health data from wearables.
The patient record would be fed data from a range of systems, particularly electronic patient records (EPRs), all of which “must be designed to record information using ambient technologies”.
It stresses that there is an urgent need to replace the large number of legacy systems, including EPRs, currently used across the health and care system and suggests that the single patient record could “provide a digital backbone and a means to replace many legacy core services” such as the personal demographics service and NHS spine.
The recommendation echoes recommendations made by the Tony Blair Institute for Global Change in August 2024 for the adoption of a digital health record.
Digital Health News contacted NHS England for comment.
13 Comments
It will be interesting to see what the plan is for a “single patient record”, is it anything more than those three words written down somewhere? In the meantime we spend millions on silo’d data in EPRs, more millions with shared care record providers, plus FDP, research data lakes, and hundreds of other stand alone systems. It may well take 10 years to achieve a single record, therefore no benefits delivered during the 10 year plan.
Alec, it looks as though the latest iteration of the “single patient record” is even more ambitious than all the previous versions!
To include a single record that is “actively used in real time by health & social care staff” and “Over time, this dataset should be broadened to include other data held by local or central government, including employment, education and safeguarding.”
What could possibly go wrong?
https://www.england.nhs.uk/long-read/neighbourhood-health-guidelines-2025-26/
17. Best practice also suggests systems should consider:
improving coordination, personalisation and continuity of care for people with complex needs, including increased agency in managing their own care, supported by:
a single electronic health and care record that is actively used in real-time by frontline health and social care staff
a care coordination function between the person or their carer and the wider multi-professional team supporting them if needed, working across organisational boundaries
A. Population health management
Ensure there is a person-level, longitudinal, linked dataset encompassing:
general practice and wider primary care
community health services
mental health
acute care
social care
public health
Over time, this dataset should be broadened to include other data held by local or central government, including employment, education and safeguarding. It should be supported by appropriate data sharing and processing agreements. This should enable analysis of population health outcomes, biopsychosocial risk drivers and health and care system resource use. NHS England will continue to work with the National Data Guardian to support integrated care boards (ICBs) to navigate the necessary information governance requirements, but partners should already share existing data wherever possible.
There seems to be an assumption that the difficult part of a single patient record is the technical digital platform, not the record itself – and a touching belief that any detailed record can be anonymised to the extent that it would be impossible to reidentify the individual to which it relates.
And that is without considering the fact that different organisations have different record needs and conventions, that conditions change and evolve, that errors are made and need correcting – and that curating a record requires a very great deal of highly skilled work with information about each individual record.
See the July 2020 BCS PHCSG report on Shared Records https://www.bcs.org/membership-and-registrations/member-communities/primary-health-care-specialist-group/specialist-interest-meetings/#
BTW, is this single record envisioned as a “read only” or a “read/write” record?
If the later, does the leaked report have anything on how clinical governance – including responsibility for acting on information in the record – will be designed?
After all, the purpose behind keeping medical records is to provide care to the patient: not data for researchers!
I couldn’t agree more.
Many Mental Health and Primary Care records contain a lot of free text including third party information which isn’t easily anonymised and cannot be safely shared.
e.g. “David’s homicidal ideation was reported to his psychiatrist by his wife Janet and, as a result, he was arrested and sectioned”.
People massively underestimate the volume of unstructured records held across the NHS and how fragmented it is. My own is spread across one general practice, three general hospitals, a community trust, a regional diabetic retinopathy service, a private hospital (which performs NHS funded cataract operations) and a cancer centre.
Good luck to anyone trying to integrate all that – even as a read only record – especially as one of those hospitals doesn’t yet have an EPR! As Amy Freeman famously put it “you can’t AI paper”.
Dave, thinking of your medical records adds another problem for the design of a single patient record (an ambition of IMG in 1992… still unrealised): duplication.
Take strokes as an example of a condition which can occur more than once & is likely to be recorded in all records in different locations: when amalgamenting them, is it one stroke recorded in 5 different record silos, 5 strokes or something inbetween?
I’ll be interested to see how the digital and technology working group plans to address the problem!
It will be interesting to see the proposed timescales for this worthy undertaking.
I still have the scars from working on NPfIT, I’d hate to see history repeated so here’s my short list of some things NPfIT underestimated:-
Public resistance to “the State” having access to their most intimate information – now multiplied by conspiracy theories about the Deep State and compounded by a distrust of a health service which now routinely provides patient addresses to the authorities to enable them to track down missing, otherwise undocumented, immigrants awaiting deportation.
The sheer complexity of legacy systems within a local health economy and the number of diagnostic and imaging systems which aren’t HL7/FIHR compliant – the numbers have reduced over the years but there’s still a stubborn remainder. A new complication is the fact that the maintenance of many Trust Integration Engines has been outsourced to third parties, sometimes to sub-contractors of the IT support provider. The bureaucracy involved in having simple changes (like opening a single port to an analyser or a middleware solution) can take weeks to arrange.
The power of the “not invented here” lobby, especially from poorly directed and misinformed Information Governance teams who insist on not accepting the outcome of collective or centralised due diligence exercises.
I’ll be watching progress with interest and detachment as I’m now retired and wish everyone involved Good Luck!
Surely Interweave is the perfect candidate for this infrastructure? With it already being an NHS owned product, with existing integrations to health and social care systems for shared care records and other boxes ticked such as connection to NHS app, remote monitoring device feeds, etc. Its the obvious choice! -lets not spend bazillions on a external platform and invest in our home grown tech.
Hi Gerry, interweave already connects to the NHSApp and provides wayfinder capability. It also works with several remote monitoring systems. As an NHS asset it would be a brilliant idea to get the value for NHS pound (which is already pressured). I message a keep repeating to NHS England.
Will patients be able to opt-out of this single patient record, or only the research access? Not all patients will think that this is suitable for their personal health data, and won’t want to participate.
Comments are closed.