The evidence is clear: There is a tendency in healthcare to use more and more data coming from electronic health records and from registries that were not typically accessible in the past. And the use of this data is reaching across borders and across different institutions.
“In addition, there is a growing demand for patient-generated data,” says Andreas Kremer co-founder and managing director of Luxembourg-based start-up ITTM (Information Technology for Translational Medicine). “We are also moving into the direction of not only taking episode data, from when a patient is treated, but using longitudinal studies. So we analyse what is the natural history of a disease, what happens between visits to the doctor. Using that information to better understand how a condition is developing and use it for prevention and prediction.”
The trend that Kremer describes is underlined by the launch of the European Health Data Space, which the European Commission says will not only give patients greater control over their health data but will also save the EU some €11 billion over ten years.
Open-source tools
The challenge ITTM faces, Kremer explains, is that typically data sources are sitting in very diverse databases, or different models. The solution is to use open-source software or standards, ISO standards if available, to bridge between these formats. “At ITTM we strongly believe in the open-source community and in having standards instead of proprietary lock-ins.” So, the company uses the OMOP (Observational Medical Outcomes Partnership) Common Data Model as well as others like i2b2 software. Indeed, in early 2019 ITTM was one of the first companies in Europe to be certified for the OMOP CDM. “These are open-source possibilities, which are used across Europe and world.-wide, not only in Luxembourg. This ensures that we really make sure that data sets in Luxembourg, in Germany, France or Belgium for example, can be analysed together in a federated way regardless of the different organisations or different languages. The model also has the benefit that GDPR aspects are taken into consideration because we are not transporting data across borders but keep it in place and secure it.”
Using the expertise and security of a local player like the EBRC (European Business Reliance Centre) is an advantage and was especially useful in allowing ITTM to continue its work during the pandemic, Kremer explains.
“To have transparency, we must have clear documentation of the information and what it means. And we must be certain that the data we are gathering are treated FAIR (Findable, Accessible, Interoperable and Reusable)..”
When it comes to quality, Kremer says that having data that is fit for purpose is paramount. “Do I have the right information available in the right depth in the right format? Do I have the right meta information? Is it transparent?” As the data is not freely available, to ensure that it is actionable ITTM adheres to the FAIR data principles, meaning that data must be findable, accessible, interoperable and reusable. “In reality, that means making sure that when talking to a registry in Switzerland, or Germany or in Belgium, that they understand the code in the right way. To have transparency, we must have clear documentation of the information and what it means. That additional data service is a reason why we are in the market, and where we see the benefit to the clients.”
Prestigious partners
Indeed, leveraging its ability to help clients with the OMOPing process, has allowed ITTM to work with prestigious partners like the Charité university hospital in Berlin, the AKH general hospital in Vienna and the National Cancer Registry in Luxembourg.
Another benefit of using the OMOP Common Data Model is that it is not only used in Europe, but also in the United States and Asia. “So you really have a worldwide community to which you can send your research question and collect the results. You’re not only focusing on a small core at the end of the day.”
Kremer says that AI will also play a role in the future of healthcare data, though he admits that the “jury is still out” on precisely how it will be used and regulated. But ITTM is already involved in two oncology-related innovative healthcare initiatives – one OPTIMA, described as “Europe’s first interoperable, large scale oncology data and evidence generation platform” and the other one which is just starting and is focussing on Integration of heterogeneous Data and Evidence towards Regulatory and HTA Acceptance. “In both projects, the combination of real-world data with clinical guidelines and AI will be help improve results.”
ITTM has a very diverse and young team with gender parity, which Kremer says is rare in the data engineering field. “We have people with a bioscience, bioinformatics or medical informatics background, some data engineers, some more IT savvy people. They have worked in pharma or in a medical setting, but others come from a completely different angle. It’s important to have this diversity because the questions we are asked are also very diverse.”
Luxembourg ecosystem
The company has also been taking advantage of EU grants – some €6 million in the last five years – from the likes of the Innovative Medicines Initiative and the Innovative Health Initiative. “This helps us gain visibility and partner on interesting projects as well as be exposed to big pharma companies like Sanofi and Johnson & Johnson or Pfizer. By showing what we can do and gaining a good reputation, we can build and nurture a network,” says Kremer.
As for the future, the managing director says he wants ITTM to continue to grow organically. “We do have ideas where we could grow faster with an investment, but the ambition is really to grow to be a preferred partner in Europe for data management, for real-world data aspects. With partners like EBRC, having secure data center and networks, we want to be part of the Luxembourg ecosystem building a cloud-based data-driven health economy.”