NHS trusts hit 93% EPR coverage. Only 30% can share the data inside it.

By its own measure, the NHS has almost finished digitising hospital records. NHS England’s 2025 Digital Maturity Assessment found that 93% of providers now run an electronic patient record, and industry analysts expect acute trust coverage to pass 97% by March 2026. That figure deserves to be taken seriously, not waved off as a vanity statistic. In 2016 the Wachter review scrapped the government’s “paperless by 2020” target as unrealistic and proposed 2023 instead; the NHS missed that deadline too. A 2019 National Audit Office review called investment “inadequate” and warned that funding plans rested on limited cost data. Getting from a minority of trusts with any electronic record to near-universal coverage inside a decade, through a repeatedly underfunded and repeatedly missed programme, is a genuine result.

But coverage was never the hard part. The same 2025 assessment, completed by 205 trusts and 42 integrated care boards, found that among the 93% running an EPR, only 30% have fully integrated bi-directional data flows: the ability for information to move continuously between systems, rather than sit wherever it was typed in. NHS England’s own report calls this “limiting system-wide productivity.” That is not a fresh discovery. The same assessment’s first results, in December 2023, found the identical pattern in different words: only 10 to 30% of trusts with an EPR had functions like integrated prescribing or record-sharing with other hospitals. Two years and several extra points of coverage later, the integration figure sits exactly where it started.

Why the gap hasn’t closed on its own

The obvious explanation is training and culture, and there’s genuine evidence for it. A Health Foundation survey of 1,725 NHS staff, run between July and October 2025, found only 46% had received basic training on their trust’s EPR and 49% role-specific training; 37% said the system was not currently working well in their organisation. Those gaps are real and worth closing. But they can’t be the whole story: training investment is a local, trust-by-trust decision that should vary widely across 205 different organisations. An integration figure stuck at 10 to 30% for three straight years, across all of them, points to something structural, not local habits.

The structural cause sits in procurement. Trusts bought EPRs independently, against a national target defined as adoption, not portability. At least seven vendors hold meaningful NHS market share, led by Oracle Health with the largest single share, and System C, Epic, MEDITECH, Nervecentre, Dedalus and Altera dividing most of the rest (this specific vendor breakdown traces to one trade-press source, not yet independently corroborated). Neighbouring trusts inside the same integrated care system routinely run different platforms built on different data models, and a vendor whose commercial position rests partly on switching cost has little reason to make it cheap to move data into a competitor’s system. The same pattern turns up outside EPRs: electronic bed management now reaches 70% of integrated care systems, but fewer than 40% report a measurable benefit. Coverage and benefit are being treated as one metric when the data shows they are two, capable of moving in opposite directions at once.

The fix already admits the diagnosis

NHS England’s response on the patient-facing side gives away that this is understood, even if it isn’t fixed clinically yet. For over a decade, trusts separately bought patient portal products: DrDoctor, Patients Know Best and Induction Zesty among them. NHS England now wants the NHS App to be patients’ sole “digital front door”. But look at how it’s being built: rather than replace those products, NHS England and the vendors are stitching them together behind a single app shell, Patients Know Best supplying the record engine, DrDoctor the booking engine, the join invisible to the patient. NHS England hasn’t solved the multi-vendor problem. It has hidden it behind a friendlier front end, a reasonable short-term fix for patients and no fix for the underlying data architecture.

What this means for the next commitment

The government’s 2025 10 Year Health Plan commits to a Single Patient Record, viewable through the NHS App, live across all providers by 2028, backed by a Health Bill introduced in May 2026 that places a legal duty on providers to share data. That duty answers a question the Digital Maturity Assessment data doesn’t raise. Trusts aren’t withholding data because sharing is optional: 90% already report a central data repository, and almost all say staff can reach a shared care record. The unresolved problem is technical and commercial, not legal: incompatible data models behind seven-plus vendor contracts, several approaching renewal. A legal duty to share doesn’t lower the cost of building a pipe between systems never designed to connect. The number worth watching over the next two years isn’t how many trusts sign up to the Single Patient Record. It’s whether the contracts renewed in that window require open data standards before go-live, or leave interoperability to be negotiated afterwards, the way EPR procurement did the first time.

The lesson generalises past the NHS to any large technology change programme. A rollout percentage is a procurement outcome: it measures whether something was bought and switched on. Integration is a design outcome: it measures whether the thing bought does what it was bought to do. Different clauses in different contracts decide each one, and a programme can hit 93% on the first while sitting flat on the second for three straight years, because nothing forces the two to move together unless someone writes that requirement into the deal before signature, not after the systems are already live.

Sources

• NHS England, Digital Maturity Assessment report 2024 and 2025 results, key findings
• NHS England, “Digitising the frontline”
• Digital Health News, “Only 30% of trusts with EPRs have integrated data flows, finds DMA” (March 2026)
• Digital Health News, “DMA results show only 10-30% of trusts with an EPR have key functionality” (December 2023)
• Digital Health News, “94% of NHS acute trusts have an EPR in place, finds report” (May 2025)
• The Health Foundation, “Electronic patient record systems in England: what do NHS staff think?” (March 2026)
• The Health Foundation, “Electronic patient records: why the NHS urgently needs a strategy to reap the benefits” (April 2025)
• National Audit Office, “Digital transformation in the NHS” (May 2019)
• Computer Weekly, “NHS England hits 90% target for electronic patient records”
• Computer Weekly, “NHS Modernisation Bill promises single patient record by 2028”
• House of Commons Library, “The 10 Year Health Plan for England”
• The King’s Fund, “Wachter Watch: Where Next For The NHS’s Digital Strategy?”