What is the most important problem we could solve without replacing everything first?
A recent Delphi study commissioned by the Spanish Ministry of Health offers an interesting external perspective on this question. The study brought together 152 experts to consider what the future health information environment in Spain should look like. Remarkably, the experts reached consensus on all 45 questions addressed in the study.
The study points towards longitudinal health information, open standards, including openEHR for structured clinical information and FHIR for exchange, alongside an architecture where clinical information retains its meaning as systems evolve. For us at Tieto, these are not separate principles. They are central to how we see healthcare architecture developing.
Start with the problem, not the system
Healthcare organizations cannot put improvements on hold while waiting for the perfect future architecture. Neither should they have to replace an entire EHR environment to improve one part of care.
The Spanish study points to the same idea: new clinical information models should be able to coexist with existing systems while new functionality is introduced. Rather than treating transformation simply as the replacement of existing systems, its recommendations describe a standardised longitudinal EHR architecture that can evolve progressively while preserving existing investments and supporting new capabilities. That sounds technical, but the consequence is very practical: Modernisation can start with a problem.
Perhaps professionals need a better view of information across care settings. Perhaps an organisation wants to improve a specific workflow. Perhaps the priority is getting data into a form where it can support new applications or AI.
Solve the problem in front of you, but without creating another dependency that makes the next change harder.
The patient has one journey. Why doesn't the information?
But there is more at stake here than how we modernise technology.
A patient may move between primary care, specialised care, social care, different providers and even different regions. A patient sees one journey. Yet the information often does not follow that same logic.
The Spanish study reinforces a simple principle: continuity of care requires continuity of information. The study identifies fragmented, vendor-specific systems as one of the barriers to a continuous view of the patient journey. Its recommendations point towards a longitudinal health record where information retains its meaning across organisations and systems. Interoperability is part of the answer, but simply moving information is not enough. The clinical meaning and context need to travel with it.
This is where open standards and structured clinical information become important. Not because healthcare needs more standards discussions, but because information should remain useful beyond the application where it was originally created.
AI needs more than access to data
There is plenty of discussion about what the next generation of healthcare AI will be able to do. I am equally interested in what it will have to work with. AI cannot compensate for clinical information that is fragmented or has lost its context.
While the Delphi study does not focus on specific AI applications, its recommendations highlight an important foundation for them: structured, semantically consistent information that remains useful beyond the system where it was created.
That is why AI and data architecture should be seen as part of the same conversation. We cannot predict every application healthcare organization will want five or ten years from now. We can build an information foundation that gives them more freedom to adopt those applications.
Futureproofing is not about predicting the future. It is about leaving room for it.
The next step may be smaller than we think
The Spanish study does not give Europe one blueprint to follow. Healthcare systems, priorities and starting points are too different for that. But the direction deserves attention: Across the recommendations, a common picture emerges: healthcare built around longitudinal information, open standards and an architecture that can change progressively rather than all at once.
Healthcare organisations do not necessarily have to wait for the next major replacement programme to start building towards a more open future.
The next step in healthcare modernisation may be smaller than we think. But if the architecture behind it is right, it can take us much further. The direction may be European, but the starting points are local. What this shift means in practice will look different from one healthcare system to another.
That is where the next part of the conversation begins.

