The Eternal Return of a Single Medical Record
Given my interests, the last few months have been interesting. The new Labour government has made healthcare one of its key areas of concern with Wes Streeting, regarded by many in the party as a rising star, appointed health secretary. Streeting has made digitisation of medical records a key part of his own ministerial branding and has spent last week, in advance of the new government’s official budget, doing the media rounds talking up the opportunities and benefits of digital medical records. The key transformations, according to Streeting, are from ‘hospital to community’ and from ‘analogue to digital’, with the former powered by the latter.
This is all good fodder for my interests, such as they are, in the sociology of medical records. Medical records perform a role in clinical work, and they are, in my opinion fascinating and complex technologies. As specialised and dispersed as medicine now is, it requires a notebook to orientate the dizzying array of knowledges, questions, problems, and uncertainties around the central unit of the patient under investigation. To bring together the cardiologist, the GP, the physio, the radiologist, and all the other the ever-proliferating experts that speak about parts and processes of the human organism in increasing magnification and specificity, one needs a system of information transmission that centres on the person and their problems.
Seen as such I’ll admit to having a kind of romantic attachment to medical records. I see the clinician’s job as something akin to Philip Marlow’s gumshoe detective work. Each clinician is thrown into a complex mystery in which the truth is illusive, everyone else seems to have access to knowledge that you do not, and the ethics of each decision only becomes clear in retrospect. The medical record is one of few tools the clinician has at their disposal that actively tries to narrate the mystery presented by each case.
Given that, this move from ‘analogue to digital’, which echoes so many policy promises of the last two decades, pricks my ears. How will a new Labour government imagine the transformation of this medical record, how, who and in what ways do they see digitisation as helpful?
Given these questions, it is worth noting that digitisation of medical records is, in many ways, the keystone of Labour’s transformation of the healthcare service (without it, there is no precision medicine, no AI, no population health, and no preventative medicine). Streeting’s announcements come off the back of the three-hundred-plus page ‘Darzi Report’ into the state of the NHS in England, however, Labour’s promise to digitise predates Darzi. It was a key part of new Prime Minister Keir Starmer’s pitch to repair the health service in the run up to the general election (he promised a ‘totally digital NHS’). Digitisation is for Labour, seen as a way of resolving wider problems in the NHS, and so this focus on medical records represents something more than the shift from paper-to-digital. Analogue to Digital is seen as a change to the whole system, not just a change to the medical record.
In truth, all of this is not new at all. The past twenty years (since the National program for IT), has been awash with policy makers promising a 21st century NHS through information technology, paperless NHS, a technological revolution, all of which was described as a way of transforming the deployment of healing services. From 1998–to-2024, policy makers have said that NHS England will be modernised through a digitisation of medical records and each time it has been imagined that digitisation will create a single digital-medical-record which patients, clinicians, managers, and researchers can each interrogate at their own leisure according to their own needs.
In 1998 the then newly elected Labour government asked the NHS executive to produce an ‘information strategy’ for the NHS. That strategy called for a single patient record that might provide a ‘full picture’ of a patient’s’ health.
A primary objective of the new information strategy is to support improved coordination of care. Developing Electronic Patient Records will facilitate the shift from profession-specific and institutional (or place-related) records to integrated lifelong person-based records and provide the essential source of almost all the information necessary to deliver the strategy objectives fully. Constructing a network of locally-based EPRs over the lifetime of this strategy will create a first generation person-based record for use across the NHS – the “Electronic Health Record” (EHR).
(Burns & NHS Executive, 1998)
Twenty-six years later, in a press release (21st October 2024) the Department of Health and Social Care said:
In transforming the NHS from analogue to digital, the government will create a more modern NHS by bringing together a single patient record, summarising patient health information, test results, and letters in one place, through the NHS App. It will put patients in control of their own medical history, meaning they do not have to repeat it at every appointment, and that staff have the full picture of patients’ health. New laws are set to be introduced to make NHS patient health records available across all NHS trusts, GP surgeries and ambulance services in England - speeding up patient care, reducing repeat medical tests and minimising medication errors.
(Department of Health, 2024)
What was recognised in 1998, and has perhaps been lost to time, was contained in the last line in the first quotation above. That is, in 1998 there was a recognition that a ‘summary’ medical record would need to be made up of many smaller records held by professions and by organisations. In 2024, no such concession is made. The smaller records in 1998 recognised the divergent imperatives to record medical information: for audit, for financial records, for clinical purposes, for research purposes, etc. A summary record would pool the pertinent, and sharable information from these smaller records. In 2024, digitisation is seen as a method through which locally held records can be dissolved altogether, replaced by centralised databases in which all medical information is pooled. Such an inversion imagines that the medical record can sit in the centre of clinical work, and all the different imperatives can pull the information that they need at their leisure. Doing so requires a centralised mandate on what is recorded and how, rather than local systems that make decisions about what information should be recorded according to their need. In my experience, this transformation has been framed incorrectly as an engineering problem (“how do we get all constituents’ data to flow into the same place”), rather than a sociological one (“who holds authority and control over what is recorded, and how?”).
The announcements made by the new government in London appear to be in broad alignment with a consistent trend in healthcare tech policy around the world. Framed as an engineering problem, the development of a single (digital) medical record, can be resolved by the development of ‘interoperable’ systems. Interoperability assumes that information entered into medical records is inherently compatible with all other information, it is only the format of the information that has stopped it from being shared (as data).
Interoperability has not delivered on this promise because the different constituents of the medical record do not trade in ‘data’ but trade in communication (on narrative). Interoperability is premised upon an, as-yet-unresolved, agreement about how clinical work will be transformed into data. The medical record services divergent imperatives: clinical (patient and clinician), organisational (managers), and research oriented (scientists). In order to function at that scale agreement needs to be reached on how these various imperatives will coexist. Medical records function, as far as they do currently, because clinicians, technicians and administrators complete them with a mind to their intended audience. Different forms are filled out differently, information is stored in such a way that readers can ignore things that are not relevant to them, and most clinicians will write summaries, problem lists, and treatment plans in narrative letters to help orientate new readers to the case of each patient. This work of summarising, reiterating, and explaining requires the careful labour of an author who is considerate of their reader. The authors and readers of medical records are many.
What then, are we to make of 2024’s new round of calls for a single, digital, interoperable, Medical Record? I think we should remain cognisant of the fact that during the NHS’ nearly 75 years, the last 25 of those years (a third of the NHS’ life) have been accompanied by calls to digitise medical records in order that we might (a) have a single medical record, and (b) that such a digital medical record could revolutionise the ways that the healing sciences are deployed. In so doing, we should attempt to critique the taken-for-granted sense that the reason medical records have not been shared more widely, hitherto, is because of a limitation of medium (paper) and not of form (medical record). Though geography has played a role in the shape and style of the medical record, it is the epistemic and organisational imperatives that have bordered the medical record and prohibited wider circulation. Computers can improve the speed of transmission and alter the calculations regarding how much and in what ways information can be stored. However, it is not as though medical records could not be circulated differently as paper. In the UK, we have had multiple paper based ‘patient held records’. Most obviously, maternity books which are paper records kept and maintained by patients themselves during pregnancy. Maternity records are the primary source of clinical information during pregnancy, they are and exemplar of what Streeting terms, ‘Patient Passports’, and they tend to be made of paper, not wires and tin. There is no technical (engineering) reason why all patient records could not be managed in this same way. What stops patient held records are the informational imperatives of clinicians and organisations and scientists. The limitation to the patient holding their own record is not paper, but is the fact that the organisation, and the clinician needs their own record of events that they maintain authority over. The paper medical records used for the first 50 years of the NHS’ life represented a negotiated settlement between patients, clinicians, managers and scientists about what was recorded and how, as well as how it would be made available and when. The idea that technology can fundamentally transform this settlement without new agreement about these things is a convenient fairytale provided to and by Health Secretaries since the turn of the millennium.
Ultimately, this is why I find interoperability to be such a fascinating aspect of healthcare technology discourse. It dresses in the clothes of a software engineering problem. The great assurance given to digitisation of medical records is that we now have the technology to make medical records interoperable. SNOMED CT, APIs, sophisticated platform EHRs and EPRs that are ready to make the great leap to a system-agnostic database-medical-record. A medical record for everyone. However, interoperability is an ontological problem, not an engineering one. If medical records are going to be made interoperable, then clinical discourse must be made to speak in one fixed and codified language. That is, ultimately the transformation to a digital service is a sociological problem, in which clinicians, managers, patients, and scientists must all agree upon what reality is, and must always commit to speaking about it in the same way.
The difficulty encountered over the last 20 yers of digitisation is premised on this confusion over what constitutes engineering problems, and what does not. Furthermore, should it be possible to align these various stakeholders, and to create one 'ontology' for medicine, it would restrict the ability of patients to present as novel, unquantifiable, mysterious, and unique cases. To turn a medical record digital in the way currently proposed, one first needs to turn medicine digital. To do so medical work must be transformed from mystery to puzzle and in so doing (and as far as such a thing is possible), each of us becomes a piece to be placed by medicine, rather than a mystery to be navigated by investigators. Ultimately, digitising the medical record is not a question of technology, but a question of how we want healthcare systems to care for and about people.
Quotations from:
Burns, F., NHS Executive (1998) Information For Health: An Information Strategy For The Modern NHS: A National Strategy For Local Implementation. Leeds: NHS Executive
Department of Health and Social Care (2024) Government issues rallying cry to the nation to help fix NHS. Gov.uk. [Online]. Available at: https://www.gov.uk/government/news/government-issues-rallying-cry-to-the-nation-to-help-fix-nhs Accessed: 24th October 2024
(Image credit: "Paper files of medical records" by Newtown grafitti is licensed under CC BY 2.0.)
