There is something almost unavoidable happening across the NHS right now. Trusts are spending — rightly — significant sums on large-scale EPR deployments. Epic is live or in flight at several major acute trusts. Oracle Health (formerly Cerner) has a long-established presence across the country. Homegrown UK platforms such as NerveCentre have carved out strong positions in specific domains. The ambition behind all of this is sound: unified records, reduced duplication, safer care.
It is worth noting that the dominant EPR vendors are American businesses, built primarily for the US healthcare market — a world of insurance billing, CPT codes, and a regulatory environment that differs substantially from the NHS. Adapting those systems for NHS workflows, NHS data standards, and NHS organisational structures has required considerable work, and the gap between what the platform can do in its home market and what a UK trust actually needs is often larger than procurement teams appreciate until they are already committed.
And yet, in ward offices and outpatient clinics across the country, the paper hasn't gone away. The spreadsheets are still open. The WhatsApp groups between community nurses and specialist teams are still the fastest way to share a result. The EPR has arrived — and for the sickle cell team, the IBD nurses, the rare metabolic disease clinic, the day-to-day reality hasn't changed as much as anyone hoped.
That is not a criticism of EPR vendors. It is a structural reality that too few people are willing to say out loud.
The Economics of EPR Development
Large EPR platforms are built on a simple, rational logic: return on investment is driven by volume. The more patients a module serves, the more it justifies the development cost. An outpatient scheduling module used by every department in every trust is worth building to a high degree of polish. A disease-specific surveillance workflow for a condition affecting 20,000 people nationally — not so much.
This is not negligence; it is arithmetic. Rare disease services represent a fraction of a percent of the patient activity that an EPR vendor needs to justify building bespoke functionality. The haemoglobinopathy team's need for structured transcranial Doppler surveillance scheduling, or the requirement to track iron chelation adherence against ferritin trends over years, will always sit low on a roadmap that is also managing demand from cardiology, orthopaedics, and the emergency department.
The trusts that have lived through previous cycles of large PAS and EPR rollouts will recognise this pattern. The vendor promises a comprehensive system. The clinical teams are told their workflows will be accommodated. Go-live happens. Months later — sometimes years later — those same clinical teams are logging data into the EPR for compliance purposes and maintaining a parallel spreadsheet because the system doesn't actually model their pathway in a way that supports clinical decision-making or safe follow-up.
Where the Gaps Appear
In practice, the limitations of a generic EPR for specialist services tend to cluster around a few recurring themes.
Condition-specific surveillance and tracking. Rare and chronic conditions typically require structured, longitudinal monitoring — specific tests at specific intervals, with the results informing clinical decisions over years. Generic EPR modules handle episodic care well. They handle programmatic surveillance poorly. The data is often there, buried in results, but the tooling to surface it — to flag a patient who is overdue their annual MRI, or whose HbA1c trajectory suggests deterioration — requires configuration depth that most trusts never receive.
Patient-reported outcomes and community engagement. The conditions that most benefit from digital pathway tools are often the ones that spend the most time outside hospital. A patient with sickle cell disease, haemochromatosis, or a rare inflammatory condition has a care journey that extends into their home, their GP surgery, their community pharmacy. EPRs are built around the acute episode. They are not designed to collect structured PRO data between appointments, engage patients in self-monitoring, or bridge the gap between the specialist team and the community nurse who sees the patient most often.
Multi-organisational pathways. Rare and chronic conditions do not respect organisational boundaries. A patient with a complex haematological condition may be managed by a tertiary centre, receive transfusions at their local district general hospital (DGH), and attend a community clinic run by a different trust or an independent provider — each of which may be on a different EPR system, or no EPR at all. The specialist team cannot see what happened at the DGH. The community nurse cannot access the tertiary record. The patient, navigating this system, experiences fragmentation as the default.
Clinical workflow nuance. The workflows that specialist teams depend on — care planning, protocol adherence tracking, multi-disciplinary team coordination, transition from paediatric to adult services — require a depth of configuration that generic modules rarely deliver. The EPR may support a care plan. It rarely supports your care plan, structured the way your team actually works.
The "Wait for the Roadmap" Problem
One of the most significant risks for NHS trusts currently mid-EPR rollout is the suppression of practical problem-solving in favour of waiting. Clinical teams with real, immediate needs are told that their requirements are on the EPR roadmap. That may be true. The roadmap, however, is typically measured in years, and the conditions it prioritises will not be rare disease services.
In the meantime, manual processes continue. Data quality degrades. Patients fall through the gaps in follow-up. The clinical team continues to carry the cognitive load of maintaining a coherent view of their patients despite inadequate tooling — and the risk that creates is invisible to the board, because it is absorbed silently by clinical staff working harder than they should have to.
I have seen this cycle play out more than once over more than thirty years of building clinical systems for the NHS. I founded InfoFlex in the mid-1990s to do exactly this — fill the gaps that the PAS systems of the day could not. It is now deployed in over 130 NHS trusts and continues to do that job well. ProtoFlex is the next step: the same philosophy, rebuilt for a cloud-first world where the goal is not just to fill gaps within a single trust but to model pathways that span organisations, systems, and sectors — layered on top of whatever EPR infrastructure is already in place, integrating with it rather than competing with it. The gap between what a large administrative system can do and what a specialist clinical service needs has always existed. The scale and ambition of modern EPR programmes does not close it; if anything, the complexity of modern multi-site care makes it wider.
Complementary, Not Competing
The answer is not to resist EPR programmes or to position specialist tooling as an alternative to them. Trusts are right to invest in EPR infrastructure. The administrative, medication safety, and basic clinical documentation capabilities that a mature EPR delivers are genuinely transformative, and the interoperability foundations they lay — HL7 FHIR APIs, patient demographics integration, results feeds — make it easier, not harder, to connect specialist systems.
The honest framing is this: EPRs are the right system for what they were designed to do. They were not designed to model the specific, longitudinal, multi-stakeholder pathway of a patient with a rare or complex condition. That is not a gap the EPR vendors will close in the near term — the economics do not justify it, and the configuration complexity required would make the system unmanageable at scale.
What specialist services need is tooling that can be configured around their actual pathway — rapidly, affordably, and in a way that integrates with whatever EPR infrastructure already exists. Not a replacement for the EPR. A complement to it, deployed at the points where the EPR falls short.
That is precisely what ProtoFlex is designed to be. We model the full pathway for a condition — the surveillance schedule, the patient engagement touchpoints, the community-facing forms, the multi-organisational data flows — and we make that model available to every participant in that pathway, integrated with the EPR where data sharing is possible and filling the gaps where it is not.
A rare disease team that has just gone live on a new EPR and still finds their nurses maintaining spreadsheets to track overdue follow-up appointments hasn't failed to use their EPR properly. They have run into the structural limit of what a general-purpose system can do for a specialist service. That limit is real, it is predictable, and it is addressable — without waiting another three years for the next release cycle.
If that resonates with where your service is, we'd like to talk. Contact us to arrange a demonstration.
Marc Warburton is Managing Director of ProtoFlex Software. He founded InfoFlex in the 1990s — now deployed across 130+ NHS trusts — before establishing ProtoFlex to take that work further.