EPR adoption in NHS trusts: what year one reveals

The gap between go-live and adoption

NHS trusts spend between £13 million and £200 million on Electronic Patient Record programmes. Go-live is the milestone everyone plans for. But the evidence from three years of Frontline Digitisation deployments tells a consistent story: going live is the beginning, not the finish line.

NHS England’s 2024/25 Digital Maturity Assessment found that 93% of providers now use Electronic Patient Records. But only 30% of trusts with EPRs have fully integrated bi-directional data flows. Coverage is nearly universal. Adoption is not.

The 2024 EPR Usability Survey, the largest study of EPR experience in NHS history (23,622 clinical staff), puts numbers to the gap. Only 38% of clinicians and 42% of nurses consider their EPR easy to learn. Just 34% say it makes them more efficient. And 47% say it actually reduces their efficiency.

These are not the numbers of a technology that has failed. They are the numbers of a technology that has been deployed but not yet adopted.

What year one looks like

Across trusts deploying Epic, Oracle Health, MEDITECH, Nervecentre, and Alcidion between 2023 and 2025, the same patterns repeat.

Training drops off when staff need it most. The Usability Survey found that 44% of clinical staff receive no ongoing EPR training after go-live, while 60% of clinicians and 70% of nurses say they want more. At Royal Surrey NHS Foundation Trust, where roughly half of staff were fully trained ahead of an Oracle Health go-live, 927 Datix incidents were reported in the first 16 months.

Support demand spikes and stays elevated. At Barking, Havering and Redbridge University Hospitals, the trust board reported 1,518 incidents in the go-live month, 63 of them directly EPR-related. Nottingham University Hospitals declared a critical incident within days of deploying Nervecentre. Early-stage incidents are expected. The structural problem is that support gaps persist long after floorwalkers leave.

Features go unused. 48% of clinicians and 43% of nurses spend three or more hours per week on “unproductive charting”. Staff skip features they find confusing and build workarounds. The HFMA documented a case where porters at one trust refused to move patients without paper notes, months after those notes had been digitised, because nobody had updated their checklist.

Attrition signals appear early. The Usability Survey found that 8.6% of clinical staff intended to leave their role due to burnout, 26% planned to leave their trust within two years, and 18% planned to leave healthcare entirely. The Royal College of Physicians found that the single thing most likely to improve the working lives of consultant physicians is working IT systems.

Why legacy methods are not enough

Most trusts rely on classroom training before go-live, floorwalkers for the first weeks, super-users for peer support, and tip sheets for reference. Despite this support infrastructure, year-one adoption patterns remain largely unaddressed

• Classroom training covers system mechanics but cannot address real-patient workflows, edge cases, or role-specific tasks encountered in the live system.
• Floorwalkers are expensive and temporary.
• Super-users have their own clinical responsibilities.
• Tip sheets are static and quickly outdated.
• Help desk tickets are the fallback when everything else fails.

The NHS Providers Digital Transformation Survey 2024 found that operational pressures impacting training and adoption ranked as the second-highest barrier to progress, cited by 50% of trust leaders. These methods were designed for deployment, not sustained adoption. They front-load support into the go-live window and assume usage will improve naturally. The evidence shows it does not.

What to measure instead

One of the gaps between deployment and adoption is a measurement problem. Trusts have limited visibility into what happens after go-live. Usage data sits in vendor dashboards. Training records sit in spreadsheets. Support tickets sit in service management tools. Nobody looks across all three.

Meaningful year-one measurement means tracking adoption by role and department (not just login counts), feature utilisation (not just system access), task completion and error rates on compliance-critical workflows, support ticket patterns over time, and staff experience at regular intervals. The EPR Usability Survey has established a national baseline. Trusts that build on it locally will close the optimisation gap faster.

The year-two question

Year one after an EPR go-live is defined by stabilisation: getting the system running, managing incidents, keeping clinical operations safe. Year two is where the investment case is either proved or left unrealised.

The NHS has reached a turning point. Coverage is nearly universal. The national conversation has shifted from deployment to optimisation and benefits realisation. But the tools and methods most trusts use to support adoption were designed for go-live, not for the 12 months that follow.

The trusts that will realise the full value of their EPR investment are the ones that treat adoption as a continuous discipline, not a go-live event. That means measuring what staff actually do in the system, not just whether they log in. It means providing help in context, at the moment staff need it, not in a classroom weeks before go-live. And it means building the evidence base that proves the investment is working, in language that finance directors and trust boards understand.

The evidence from three years of Frontline Digitisation deployments is clear. Going live is the start. What happens in year one determines whether the programme delivers on its promise.