Electronic patient records: why the NHS urgently needs a strategy to reap the benefits

Since the early 2000s, successive governments have tried to encourage the rollout of electronic patient records (EPRs) – computer-based software for documenting patient care – across secondary and community care. There are good reasons for doing so. Used effectively, EPRs offer opportunities for improving productivity and reshaping care, including benefits that could be fundamental to achieving the government’s three ‘shifts’ (from analogue to digital, sickness to prevention, and hospital to home). Evidence cited by NHS England suggests that digitally mature trusts operate with approximately 13% improved efficiency compared with their less digitally mature peers. 

As a result, the number of NHS trusts with an EPR has been steadily increasing: according to NHS England, as of February 2024, 189 trusts had a system in place. Although the NHS is now on the verge of reaching 100% EPR coverage, some trusts are still struggling to get one in place, and many more trusts are finding it difficult to realise the full benefits of EPRs.

To explore the challenges involved, in 2024 we conducted exploratory interviews with leaders in five NHS acute trusts in England, under conditions of anonymity. We also looked at the example set by a leading medical centre in New York. This analysis offers insights from both these sources, with illustrative quotes from the interviews cited throughout. We conclude by recommending next steps for unlocking the potential of EPRs.

Although this piece focuses on EPR use in acute trusts, some of the underlying lessons on realising benefits from this kind of technology will also be relevant for providers in other sectors, such as emergency, primary, community care and more. 

Although initial efforts to digitise care records in primary care began in the early 1980s, plans for secondary care started slightly later. Figure 1 sets out a timeline of attempts to deploy EPRs in the NHS over the past two decades.

In 2002 the National Programme for IT set out to create a centralised, national system where patient records could be accessed across different NHS organisations by 2010. Initially expected to cost around £7bn, in July 2011 the Committee of Public Accounts stated that the Department of Health and Social Care (DHSC) had ‘accepted it is unable to deliver its original vision of a standardised care records system with an electronic record for every NHS patient’, after spending more than £10bn.

Government aims and timescales for digitising patient records have changed several times, as successive administrations have found the challenge of rolling out EPRs and enabling secure data exchange more difficult than anticipated. Professor Robert Wachter’s 2016 report, ‘Making IT Work’, identified several reasons for these difficulties, including fragmented infrastructure and legacy systems, lack of IT expertise, cultural resistance to change and the shadow of previous programmes for IT.

The Frontline Digitisation programme was launched in 2021 with the aim of getting an EPR into every NHS trust by March 2025. In May 2024 NHS England was targeting 98% coverage by March 2026, but in January 2025 they said that all trusts without an EPR will ‘continue to work to procure and implement one as quickly as is safely possible’. The plan for the last 2% (which equates to seven trusts), and the reasons for this latest delay are currently being debated, but it is clear the NHS has been slow to come to terms with the range of barriers that some trusts have faced with implementing EPRs. In addition, over the next few years a significant number of trusts’ existing EPR contracts will expire, and these will need to be renewed or replaced.

While the achievement of 90% EPR coverage is substantial, scepticism remains among both industry and health care professionals as to the benefits currently being gained from these systems. Many feel that EPRs have, for the most part, been poorly implemented or are being used only for their most basic functionalities. While recent Health Foundation research found that EPRs were seen by clinical staff as one of the technologies with the greatest potential to save them time in their work within the next 5 years, it also surfaced significant concerns that EPRs are not yet being used effectively or to their full potential. For example, several interviewees highlighted longstanding challenges with interoperability and data sharing – describing the frustrations they encountered in being unable to access notes and test results from other providers, leading to time-consuming efforts to track down patient information. This kind of qualitative evidence is supported by the results from the first year of the Digital Maturity Assessment (DMA), which revealed that although 90% of trusts have an EPR in place, only 10–30% of these EPRs are using more advanced functions – such as integrated prescriptions or record sharing with other hospitals.

It is important to note that the introduction of EPRs has been a challenging process for many trusts – attempting to maintain business as usual while their processes of delivering and recording clinical care undergo substantial change. This has been recognised by NHS England, who published a contract notice worth £13.3m in October 2024 for ‘tiger teams’ to support the delivery of EPRs across England. These teams aim to provide additional resource, expertise and information to help with the rollout of EPRs.

The importance of effective implementation and use of EPRs is highlighted by experience to date, with evidence that for some trusts implementation has been associated with significant increases in patient safety incidents in some hospitals. More generally, a literature review recently commissioned by the Health Foundation suggested that EPRs are a type of technology where the benefits are particularly sensitive to implementation factors: 56% of relevant studies found a positive impact on staff time with 44% finding no or negative impact.

The patient perspective must be central to considering next steps for EPRs. These systems can help improve the experience of health service users by joining up information systems, helping to achieve the government’s goal of only having to tell your story once. Our research has found public support for the NHS making greater use of electronic records. While a detailed investigation of patient views was beyond the scope of this research, it will be key for a strategy on the next steps.

This section draws on analysis of our interviews with leaders in five NHS trusts in England. The overarching story is that acquiring EPRs is only the first step in a long journey, and that the resources (both financial and human) needed to reap the intended benefits of an EPR are often significantly underestimated. Different EPR systems are used across the trusts interviewed, and it was clear that providers of these systems vary in terms of cost, level of customisation available, and the extent to which they have been designed with the NHS (or other health systems) in mind. 

Lessons from trusts’ implementation journeys so far relate to vision and ambition; issues with data input and extraction; training and culture; and the challenge of maximising benefits. 

Vision and ambition

All trusts interviewed were driven by a belief that EPRs would improve productivity, as well as a general sense that this was a natural modernising step that needed to be done. The biggest difference between the trusts we spoke to was in the level of functionality in their systems, and the funding available to support EPR use. 

Specifically, most did not invest in operational research or quality improvement teams from the outset. The initial financial outlay was already so large that setting aside extra money for these so-called ‘luxury’ teams was often not possible, particularly with many organisations focused on ‘keeping the lights on’. Notably, the trusts that did invest in these areas had access to ‘non-core’ sources of funding from well-funded charitable arms or private partnerships with industry. For most, the focus was on simply setting up the system (‘digitising paper’), with the assumption (or hope) that operational benefits would naturally follow. The absence of a roadmap for how to use EPRs to reshape care stands in sharp contrast to the case study from the US (in the next section). 

Data input quality and extraction

Data within EPRs can be used not only for direct care but also for things like service planning, improvement or research, but this requires it to be structured (organised in a standardised and comprehensible format) and extracted. In many trusts this is only just beginning to happen in a meaningful way. We heard in our interviews that without effective data extraction EPRs ‘are essentially digital notebooks’, but this extraction has proved much more challenging than trusts anticipated, or suppliers had promised.

The quality of the insights that can be gained from EPRs is dependent on the quality of the data input. To be high quality, data must be both accurate and captured in a way that is accessible. Culturally, many clinicians are not used to recording their interactions in this formalised way (often viewing the approach as ‘American’). Many clinicians are uploading PDFs or typing notes in unstructured data fields because the correct structured fields are not available, because it isn’t straightforward to do so, or because they are reluctant to abandon the traditional narrative method of clinical data entry. 

For example, despite an EPR having a clear field for ‘smoker’ (which produces structured data), clinicians may instead write ‘smoker’ in the unstructured notes. This creates multiple places a clinician might look to find a single piece of information. It also makes identifying and extracting data on smokers challenging – and unreliable if extracted data doesn’t include free-text fields, which is often the case in service evaluation and research. Even the apparently simple task of finding all smokers in a dataset is now complex, with analysts having to spend time combing through multiple sources to find references to ‘smokers’. For routine reporting, many trusts have now managed to standardise datasets and produce daily business intelligence reports. However, each time a bespoke ‘data pull’ is needed for quality improvement or research, these challenges reappear, and staff are rarely available to tackle them.

Trusts also need ready access to their own data, but this is not always easy. Suppliers sometimes hold data in separate data warehouses, and we heard that getting such data out can be ‘a nightmare’. A more detailed understanding of the data access agreements between trusts and EPR providers is needed and will be key to ensuring EPRs can be used properly. The upcoming Data Use and Access Bill aims to make patients’ data easily transferable across the NHS by applying information standards to IT suppliers and strengthening enforcement powers. This Bill should provide an opportunity to address some of the issues our research surfaced on data extraction.

Training and culture

We heard from interviewees that training is a huge challenge (‘I’ve got 16,000 staff; how do I train them to use the system?’). Getting this right is crucial: if clinicians have not been trained on how to use an EPR effectively then it can feel not particularly useful and, at worst, like a burden. This can result in low enthusiasm and therefore low adoption and ineffective use (eg resorting to unstructured fields), creating a cycle in which systems are poorly used and become less useful over time, in turn lessening uptake. 

Various types of training are relevant to getting the full range of benefits from EPRs. 

For staff who are entering information into the EPR systems, the first is training on basic, day-to-day use of the system – for example, general functionality, how to use shortcuts and how to code medical conditions. This will help ensure that data are high quality.

However, it is also vital to ensure staff understand the importance of entering data accurately and comprehensively. Training in ‘why’ to do things, not just ‘how’ to do things, can help give staff more ownership and awareness of the benefits that EPRs can bring when used in the right way, both for individual patient care but also for the wider system. 

A further step would then be training on how to use more advanced functionalities to generate insights. Currently, most clinicians ‘don’t know the full functionality of the system and are using it at its most basic level’. Training on how to use more complex tools (such as for quality improvement) and interpret outputs is essential to get more out of these systems and engage clinicians in shaping care. 

Above all of this sits the question of culture. Building a culture where staff are ‘in dialogue with’ the EPR is the ultimate ambition. Ideally, clinicians should be querying and interacting with the system, continually bettering their understanding of how digital tools can help. Where the EPR software can be improved, clinicians and others should ideally work with their EPR provider to improve the functionality.

Top-down directives to clinicians to begin using a new technology will not be sufficient to create this culture. One trust we spoke to has introduced EPR projects into its junior doctor training, encouraging clinicians to use the system and ‘play around with it’ right away as part of mandatory quality improvement training. We heard that allowing clinicians to become familiar with the system and its benefits from the outset has already reaped positive results, with one project identifying incorrectly coded allergies and taking action to address this. 

Maximising benefits 

Most resources dedicated to EPRs focus on day-to-day upkeep. One trust interviewed employs 20 full-time staff just to manage annual upgrades to the system. This is essential activity but does not touch on the work of gathering insights or intelligence from the EPR. To better utilise EPRs trusts require ‘specific financial investment and staff whose roles are not just related to the daily maintenance of the system’. Data scientists, data engineers and research staff are needed to support the extraction and use of EPR data, but it is rare for trusts to have such teams. Those that do, often rely on ‘non-core’ sources of funding from charitable arms or partnerships with industry. But most trusts do not have access to this kind of funding, and are already struggling under current financial pressures, making it near impossible for them to invest in maximising the potential of their EPR. One interviewee described how spending on EPRs is ‘competing with the roof falling in’. 

Having people and resources specifically dedicated to developing new ways of working, through using the higher order functionalities of EPRs, would enable trusts to tap into much greater benefits. This could help staff to use tools that are often available but underutilised, such as appointment management, automatic triaging of patients who don’t need to be seen in person, or synthesising patient appointments in one portal. Focusing additional resource on these areas could therefore allow the NHS to draw out additional benefits from EPRs relatively quickly. One of our interviewees stated bluntly, ‘we’re triaging in the EPR before we even see a patient…this could take out 30% of our referrals’. 

It’s important to note the importance of feedback loops as trusts progress through their EPR journeys. The stages of developing an EPR are set out in Figure 2.

There are health systems that are considerably more advanced in their use of EPRs than most NHS providers. One such example is NYU Langone Health (NYULH) in New York. Although the UK and US health care systems differ fundamentally, understanding what it took to achieve successful use of EPRs, both financially and strategically, can provide useful intelligence for NHS policymakers and providers in the UK.

NYU Langone Health is a large academic medical centre in New York City and one of the most digitally sophisticated hospital systems in the US. It has five major hospitals, a community hospital and a huge network of outpatient and primary care clinics, serving around 8 million individual patients per year, with revenue of around $14bn per year. It is recognised as being one of the top health providers in the US for quality of care and patient safety. We spoke with a senior leader at NYULH to better understand what it takes to get to this stage of digital maturity, and the role played by their EPR. 

NYULH is known by patients and its EPR provider for the quality of their digital products and capability. They have gone well beyond using the EPR just for patient records, and have established it as a clinical decision support tool. NYULH is now routinely building and embedding AI and other predictive models into their care, running statistical tests, and using a sophisticated set of system-wide performance and operational dashboards. 

Having introduced its EPR in 2008, it took 15 years to reach this position. 

The process was initially spurred by a new dean (the equivalent of a trust CEO) who wanted to improve access to information to drive change and improve quality – as he originally remarked, ‘I’m driving a 747 without a cockpit or windows’. At first, different service areas were gradually brought onto the EPR. Then, in 2012, Hurricane Sandy caused a complete shutdown of NYULH, creating 'a flooded mess that turned out to be a medical gem'. This crisis turbocharged adoption of the EPR: as each system and office prepared to reopen (a process that took 4 months), they were introduced to the EPR. Initial implementation alone is estimated to have cost around $500m, a figure echoed by other US providers to have recently adopted EPRs. 

Wider US policy also supported this digitisation drive. The passing of the Affordable Care Act in 2010 promoted the use of electronic patient records, building upon the previous Health Information Technology for Economic and Clinical Health Act. By 2014, 97% of hospitals had adopted a certified EPR, up from just 9% in 2008 – a huge increase over a relatively short period.

Notably, the Act also included incentive payments to increase the ‘meaningful use’ of EPRs. This was defined as using EPRs to:

• improve quality, safety, efficiency, and reduce health disparities
• engage patients and families in their health care
• improve care coordination
• improve population and public health.

Some key factors enabling the effective use of EPRs at NYULH are staffing, organisational structure and culture. 

Staffing 

Health IT is a key enabler of NYULH’s drive to deliver high-quality, safe care. NYULH’s IT department employs approximately 900 US-based staff plus large offshore teams. The bulk of this staffing is dedicated to the routine upkeep of the EPR. The team that maintains the EPR and works on integration with new apps or technologies consists of around 250 people in the US, plus a team in India. Large user-experience teams are responsible for making the ‘front end’ make sense for staff and patients.

Clinical informatics, the division responsible for the higher order uses of the EPR (such as improving clinical care and research), is staffed by 20 clinical informaticists as well as associated IT staff. This team has access to the medical school’s teams of academic researchers and AI developers, including teams like the predictive analytics unit who build bespoke AI models. 

Organisational structure

At NYULH, the IT departments are built into the organisation’s decision-making and governance structures. In this relatively unique model, new projects feed into the existing operational and senior leadership committees, and IT is seen as a core component of how the health system runs. There is also a bi-weekly project portfolio meeting with senior leaders, reviewing up to 100 projects to decide whether to approve and fund them. 

Culture

Digital integration is a core part of the culture of the organisation, and a way in which they distinguish themselves in the highly competitive New York City health care market. The implementation and expansion of the EPR was described by our interviewee as ‘ruthless’. When a new hospital or clinic is brought on, the organisation is immediately held to account for use of the EPR. They are monitored and supported by meetings every 12 hours for the first few weeks, and there is no EPR ‘down time’ scheduled.

Since re-introducing the EPR in 2012, the focus at NYULH has been on whole-hospital change. The EPR shuts down four times a year for 2 hours, allowing for upgrades – which clinicians accept as a necessary annoyance to keep the system fresh and functional. Within this culture, EPR performance indicators are seen as part of core clinical and operational metrics. 

It is perhaps not surprising that reaching maturity of EPR use is proving such a challenge for the NHS. Large amounts of time and money are required just for upkeep, let alone to convert EPRs into intelligent tools that can transform the way care is delivered. As a result, most of the hoped-for productivity and quality gains have yet to materialise. 

A small (and gradually diminishing) number of trusts still need to implement EPRs, and a larger number will need to upgrade or replace their current systems soon. As well as continuing to support these organisations, it is now critical that attention shifts to encouraging the meaningful use of EPRs to reap the benefits for patients, staff, providers and the health system as a whole.

Putting a strategy in place

With 100% EPR coverage in sight, now is the right time for the government to put a strategy in place for the next stage of the EPR journey. This strategy should cover both how to get the basics right and how to develop more advanced uses of EPRs. As part of this, individual trusts should be asked to develop their own plans for the next stages of EPR usage. 

The rollout of EPRs in the US, guided by legislation, included a ‘meaningful use’ programme which established three stages of EPR adoption, each with specific requirements and timeframes: basic adoption and data gathering; advanced clinical processes; and improved outcomes. While fundamental differences between US and UK health care systems rightly mean that lessons should be applied with caution, this example highlights the importance of focusing on the implementation and optimisation of EPRs beyond initial procurement. There must be a detailed strategy for moving from basic functionalities to use of advanced capabilities that can significantly improve care. 

So, what should a strategy for the next steps for EPRs in the NHS consider? 

Immediate next steps must focus on getting current EPRs working properly, with particular attention to data quality and staff training. Inputting high-quality, correctly coded data is crucial for trusts to be able to extract useful data that can be analysed and bring wider benefits to the system. Undertaking robust staff training and engagement is essential, including having a senior figure able to champion the benefits of an EPR. Staff initially need to be trained in both the day-to-day use of the EPR and how to use it for quality improvement. Working closely with staff will be important to understand how to make EPRs as useful as possible and embed them into workflows. Here, it will be key to learn from the mistakes of previous tech-related change programmes where the role of technology was overestimated, and the role of people underestimated.  

EPR suppliers increasingly provide additional tools within their systems (albeit often at an additional cost) that can help improve productivity and patient care. Working collaboratively with industry to get the most out of these more established tools, such as for appointment management, may yield more benefits in the short term than more cutting-edge or eye-catching risk prediction models or AI solutions. In any case, it won’t be possible to move to some of the more advanced functionalities of EPRs without getting the basics right (‘you can’t run an AI on paper’). As Professor Robert Wachter noted in his 2023 lecture for the Health Foundation, ‘for some parts of the NHS, investing in shiny new AI tools now would be a bit like starting to decorate your bedroom before your house is fully built’.

Once trusts have the basics in place, the strategy can then set out how trusts should be expected to develop their EPRs to enable use of higher order functionalities – such as automatic triaging and real-time decision support – where considerable benefits could be found. As our US case study shows, providers themselves bear much of the responsibility for good EPR implementation and usage. As such, direction at a national level will need to be coupled with trust-level strategies that can take into account their own context and where they are on their implementation journey, and consider the detailed requirements involved.

Supporting this strategy with resources

It’s important to realise that reaching maturity on EPRs will need to be supported by funding and coupled with mechanisms to ensure accountability for progress. This will need to include three different types of funding: for the systems themselves, for implementation support, and for ongoing maintenance, improvement and optimisation. As discussed, some trusts have been able to forge further ahead because of their access to ‘non-core’ funding to support infrastructure and change management, but many trusts do not have access to these kinds of resources. Although funding alone is not sufficient, and while it will of course be a challenge to find resources for this in the current constrained financial environment, trusts without access to such support will struggle to develop and use the ‘higher order’ functionalities of EPRs, where many of the wider benefits to productivity and quality of care are found.

Learning from those that are further ahead

Any strategy developed must work for all trusts, not just those leading the way. Indeed, there will be important opportunities to learn from providers who are further ahead in their EPR journey – both in the UK and elsewhere. Despite differences between the UK and US health systems, the US experience can offer valuable learning here – on the pitfalls as well as the positives. A strategy for improving EPR use must find appropriate routes to capture and share these learnings.

EPR systems are a chance to transform the way care is delivered. Simply digitising current information and processes will leave the NHS in much the same position it is now. But ensuring trusts can use EPRs effectively will allow the NHS to capitalise on their transformative potential for improving and integrating care, streamlining pathways and increasing productivity.

As the NHS approaches the milestone of 100% EPR coverage, it is right to recognise this achievement, despite the length of time it has taken. However, there is no time to waste in improving the use and capabilities of these systems. The government need not wait for the few remaining trusts to implement EPRs to move forward with a national strategy.

The longer EPRs are in place without delivering benefits for staff and patients, the harder it might become to bring people on board. The NHS urgently needs a route map that sets out first, what trusts must do to ensure their EPR’s basic functionalities are being used as intended, and second, how to go beyond this to reap the ever-growing list of potential benefits that such systems can offer.