Data and decision support for complex care scenarios

At DMEA 2026, ID will be going on the offensive with a new clinical data repository. What exactly is that?

A clinical data repository, or CDR, is a digital platform for hospitals that brings together data from a wide variety of IT systems and devices. Without such an integrated data platform, modern hospital care would be almost impossible to organise. ID has been a market leader in the standardisation of both administrative and clinical hospital data for many years. To make even better use of this, we are now launching our own CDR. This can be seen for the first time at DMEA 2026.

What exactly does this CDR from ID look like?

Our FHIR native CDR, as we call it, combines all the essential components in one platform: a very powerful database that provides medical data in the FHIR standard, a terminology server that helps convert data into different formats, a query tool that assists in defining cohorts, dashboards and a basis for medical expert systems. Our cooperation partners are Firemetrics for the FHIR database and DMI for data management, data security and operations. The terminology server comes from us, and queries run via our standard query portal DaWiMed. This allows us to integrate our clinical decision support systems (CDSS) directly into the CDR, which is a major added value compared to many other CDRs. Our CDR can therefore be used directly in patient care, where it can provide a wide range of benefits during treatment. We will also integrate an NLP pipeline that allows free text to be made available to the expert systems in near real time.

Keyword expert systems: How much modern artificial intelligence and how many large language models (LLMs) are now included in your solutions? And how much is classic rule-based decision support?

I believe we need both. When it comes to CDSS, our focus remains on rule-based tools that deliver consistent, traceable quality. In our opinion, automation in hospitals cannot function without quality and traceability. Nevertheless, we also have AI adapters that allow machine learning and LLM to be docked where it makes sense. And we do that too.

You don't sound entirely convinced yet...

We are following this topic closely. The question of how LLM-based AI will develop in hospitals is one of the most exciting questions of the coming years. Like many others, we have now integrated an LLM-based chatbot into our solutions. This can also be seen live at DMEA. It allows users to virtually talk to the patient file while documenting. For example, they could ask: ‘Did the patient show signs of acute kidney failure and is there a creatinine history?’ This no longer needs to be looked up manually. I do believe that such tools will soon become standard. The question then is: how exactly do you operate them?

What options are available?

The obvious choice is to run something like this in the cloud, but then you become dependent on the cloud operator. The other option is to run it locally in your own data centre, but this requires a large number of rather expensive graphics cards. With an estimated 1,200 hospitals in the future, this will not be possible everywhere. Another question is: how much AI do I really need? For example, do coders really need a chatbot? After all, they are professionals who can code. Automation is much more helpful in this area. I am curious to see how this develops. We are currently discussing this intensively with many of our customers.

In addition to expert systems and AI, outpatient care is another key topic for ID at DMEA 2026. Why outpatient care?

In terms of health policy, this is one of the most important issues for hospitals at the moment. We can also see this in the webinars on outpatient care organised by our partner MEDIQON GmbH, some of which attract 600 participants. As experts in billing and coding, we can and must support our customers in a variety of ways – including with AI. In principle, our strategy is to expand all our solutions so that they can also be used in the outpatient sector. Our ID DIACOS® billing software will be expanded to include outpatient billing according to EBM and GOÄ. Controlling tools such as ID EFIX® will also be adapted to outpatient requirements. We have already done this in Switzerland, it is in the works in Germany – and we will be presenting it at DMEA. We will be supplementing the whole thing with a new product called ID Portalmanager. This will enable hospitals to specifically control the billing method for elective admissions.

How exactly does it work?

Ultimately, the portal simulates the different billing methods, i.e. purely inpatient DRG billing, hybrid DRG billing and hybrid outpatient billing. It then shows which method promises the best revenues. We will also develop a corresponding tool for emergency rooms, where the degree of freedom is naturally somewhat smaller. Speaking of portals: what we will also be showing at DMEA are several dashboards that have emerged from our strategic partnership with MEDIQON GmbH. We first presented this partnership at DMEA in 2025. These dashboards deal with issues such as physician financing costs, but also the analysis of regional care structures. This is becoming increasingly important for hospitals, as hospital reform aims to increase the formation of regional networks. This is also important in the context of outpatient care: there will be regions where the potential for outpatient care is not as high as some people think, for logistical and demographic reasons. It helps to know this in advance.