Why Data Integration Centers are more than just technology | 5 Questions for… Data Integration Center Director Dr. Irina Lutz
Medical informatics is the intersection between medicine and IT. Different professional fields overlap and complement each other. Even less obvious specialties are represented and relevant here. The Data Integration Centers of the Medical Informatics Initiative embody this intersection. The Data Integration Centers bring together routine healthcare data centrally, anonymizing and harmonizing it to make it available for medical research. Behind the centers is a complex infrastructure that needs to be well thought out. Employees should have interdisciplinary knowledge. Dr. Irina Lutz heads one of the seven Data Integration Centerss established in SMITH so far. With a PhD in engineering, she combines technical expertise with computer science and basic clinical knowledge. In her current position at RWTH Aachen University Hospital, Dr. Lutz is responsible for the strategic planning of tasks and processes in the Data Integration Center.
She reveals to us in an interview what challenges are involved and where she sees the future of Data Integration Centers.
You come from an engineering background. How did you come to your current job and how can you contribute your skills as an engineer to the Data Integration Center?
I had connected with an employee of the University Hospital Aachen via Xing. That’s how I found out that a large project was being advertised as part of the Medical Informatics Initiative, and then I applied on my own initiative as a project manager. I was already familiar with the processes involved in scientific work, projects and the acquisition of third-party funding through my work as a project manager at the Technical University of Ilmenau. After that, I worked as an engineer in industry, at that time as a product manager. Therefore, I knew both the side of industry and the side of science. So I was a very good fit for the position as project manager and later as head of the Data Integration Center.
In addition, as an engineer, I bring plenty of experience in organization, planning, infrastructure setup, requirements gathering, process establishment, and a structured way of working for administrative-organizational work as well as the introduction of new ideas. All of these processes are very similar in IT and engineering. Product managers also do nothing but organize and evaluate all the processes involved in establishing products. You also get in touch with manufacturing companies and customers, which at my previous employer included hospitals or healthcare facilities. I have taken these experiences into my current job.
So as an engineer in MII, you don’t design and plan technical components, but an entire project.
Exactly, the processes are similar, only before it was the construction of a hardware in cooperation with companies, here it is the construction of an infrastructure. You also have to think ahead: What exactly do you want to build, what is needed in the process, what steps have to be taken? Are our resources sufficient? How can I strategically deploy employees according to their skills?
What challenges did you overcome during the commissioning of the Data Integration Center in Aachen?
I would say the biggest challenge was integrating Data Integration Center processes into processes that had been established for years at the University Hospital in Aachen. All of a sudden, clinic employees have to additionally record Broad Consent, the Board of Directors and other people who are already well occupied have to sit down and approve applications. The clinic’s data protection officer also has to find time for us. The Data Integration Center presents the UKA with additional tasks, and we as the Data Integration Center team have to somehow adapt to existing processes. In addition, the processes in the hospital are already tightly scheduled.
To what extent do the Data Integration Centers in the consortia differ from one another?
In the course of the project, it turned out that the primary systems used differ between the university hospitals. This means that the data is stored in different structures and is available in different formats. In one hospital, for example, only the HL7 format is used, while in the other the primary systems can already issue FHIR. It became apparent that it is not always possible to follow the same path when it comes to infrastructure implementation. This is also the difference that shows up in the infrastructure in the end. However, through these differences, clinics can learn from and support each other.
The last projectathon at the end of the current funding phase found that all SMITH Data Integration Centers are able to deliver data. How will the work in the Data Integration Centers develop in the future?
By and large, the infrastructure is in place. In the near future, we must concentrate on ensuring that the data quality is adequate. Then we must increasingly participate in data use projects and offer support within the hospital. Each Data Integration Center should establish itself within its own university hospital and provide more support to medical staff. A more intensive cooperation in projects with researching physicians from the own or other hospitals would be conceivable. The cross-clinic projects are also very exciting for the Data Integration Centers and important for further development.
Of course, data must also continue to be submitted. So far, we have only focused on data from the use cases in the SMITH project and have only integrated these from the primary systems into Data Integration Centers for the time being. However, there is much more data in the primary systems that we can bring into the Data Integration Centers. The goal is to have as much data available as possible so that when researchers want to query data, all the necessary data is there and can be made available directly.
Finish the following sentence: What excites me most about the field of medical informatics is…
… the idea of building interoperable databases in university hospitals that medical researchers can access. In the end, the infrastructure should primarily benefit citizens by improving medical treatments and possibly even curing diseases completely. I came to the Data Integration Center and the project with the motivation to participate in the development of methods to cure diseases.