The contract runs for exactly twelve months. It began on 2 November 2026 and ends on 1 November 2027, and within that window NHS England has committed to a date by which clinicians should be able to see medical histories for at least 80 per cent of patients. The £47.7 million agreement with CGI IT UK Ltd, the British subsidiary of the Canadian technology group, covers building the core platform for a single patient record, supporting it and managing its eventual transition. It sits within the Transforming and Connecting Care programme, and it gives a long-discussed ambition a supplier and a calendar for the first time.

The sum is modest against the scale of NHS technology spending. The word that deserves more attention is "interim". A contract that spans build, support and transition suggests the long-term arrangements for running the platform remain unsettled, and that the programme has chosen to commit before resolving them. That may be pragmatic. A twelve-month term also preserves the centre's bargaining position, which matters given how earlier national programmes fared. The national programme for IT, wound up in 2011, failed for several reasons, among them contract structures that left the service with little influence over its supplier and a clinical workforce asked to adopt systems it had not shaped. An interim arrangement protects against the first problem only if someone is clearly accountable for what follows it.

The timetable is where the risk sits. By March 2027, four months into the contract, the 80 per cent visibility target is due. Meeting it will depend largely on regional shared care records that already exist, many built by local teams with the support of local clinicians. Yet the three-year plan is to retire those records in favour of direct system-to-system integration between providers. The programme therefore relies on assets it intends to dismantle. If the new integration layer is switched on before it matches the reliability of what it replaces, frontline staff will notice first, and their verdict will shape adoption more than any ministerial announcement.

Patient access follows in 2028, when core records begin to appear in the NHS App. The programme starts with maternity and frailty, a sensible choice. Both involve many handovers between services, and a missing detail in either can cause real harm. Mental health and oncology come later, and they carry the most sensitive data and the sharpest questions about consent. The public has seen what happens when data sharing runs ahead of explanation, and the care.data episode of a decade ago still shapes how many patients and clinicians view national data projects. Patients who can read their own records will also find errors, and the queries that follow will land on general practice, which has little spare capacity to absorb them.

The operational case is nonetheless strong. Waiting lists reflect limits on throughput as well as growth in demand, and a good deal of clinical time goes on repeating histories, chasing results and ordering tests already done elsewhere. A record that opens reliably returns minutes to clinicians across thousands of consultations. Those gains will not appear on a spreadsheet at the point of procurement, and they will not appear at all if clinicians conclude the record is incomplete. Staff who distrust a system reproduce its contents on paper and in private notes, which defeats the purpose.

The implications differ by audience. NHS leaders should resist pressure to decommission local shared care records until the replacement has demonstrated equivalent performance in their own systems, and they should insist on clear data on where the 80 per cent figure is and is not being met. Policymakers need to fund the integration work that trusts and general practices must carry out, since a national platform imposes local costs that rarely appear in the headline figure. They also need to decide, before 2027 ends, who will operate the platform when the interim contract does. Suppliers of electronic patient records and other health technology will be expected to connect to the platform, which makes interoperability standards a commercial matter as much as a technical one. Patients are entitled to clear answers about who can see what, delivered before the app goes live and not after the first complaint.

March 2027 will tell the service more than the contract value does. If the target is met and local records remain available as a safeguard, the centre will have earned credibility for the harder phases that follow. If the target is declared met on a narrow definition of visibility, the gap will surface later, at greater cost and under closer scrutiny. Every later stage of the programme depends on whether clinicians find the record complete and accurate when they open it.