On Monday September 30th and Tuesday October 1st, EHDEN invited a first group of five SMEs at the Erasmus Medical Centre in Rotterdam (The Netherlands). During these 2 days, these SME were certified by EHDEN for their knowledge on standardising health data to the OMOP common data model (CDM) and the installation of the technical infrastructure. By having set up a rigorous process to select, train and certify SMEs throughout Europe, the EHDEN project concluded another major milestone in the 1st year of the project and took a huge step towards training the European real-world data community.
This process started back in April of 2019, when EHDEN held its first open call for SMEs. From a total of 28 eligible applications, a group of 11 SMEs were initially selected. Over summer, these SMEs have followed the pre-defined curriculum in the EHDEN academy, an online learning environment which was developed de novo by EHDEN for the purpose of educating the community. Development of the courses in the academy has been a collaborative effort between OHDSI and EHDEN and these courses cover a wide range of topics spanning the whole journey from raw data to reliable evidence.
The first day of the meeting started with Peter Rijnbeek (Associate professor, Erasmus MC (NL)) and Patrick Ryan (Assistant professor, Columbia University Medical Centre (US) and Vice President of Observational Health Data Analytics at Janssen Research and Development) reiterating that EHDEN aims to build a community of both certified SMEs and data partners in Europe, to generate reliable evidence for the benefit of patients. It was very clear that SMEs are considered a cornerstone of this EU community, as their expertise will be an enabling factor in real world evidence generation at scale. A lot of the discussion focussed on the creation of an ecosystem in Europe where SMEs help each other for the benefit of science and thus patients.
To stimulate collaboration between SMEs multiple exercises were performed in groups, for example Maxim Moinat (The Hyve (NL)) assigned all participants randomly to one of 5 groups and provided them with scan reports of a dataset called “Frankenstein”. This fictional dataset was full of examples of potential problems one might encounter when defining an ETL (Extraction Transform Load) procedure. As an exercise, they were tasked to identify these problems.
On the 2nd day, the process of defining and implementing an ETL for a data source was emulated. First, the participants received an overview of the Integrated Primary Care Information (IPCI) database by the IPCI data specialist, Marcel de Wilde (Erasmus MC (NL)). Each SME had access to a simulated version of the IPCI database in a Virtual Machine in the Amazon Cloud. The SME then had to define the mapping logic and a process on how they would approach the implementation of the ETL for IPCI. During the reporting back, each SME had to present and discuss their proposed ETL implementation process and the lessons they learned throughout the day.
At the end of the second day, each SME received their EHDEN certificate. The EHDEN project will provide further support to these SMEs in future mapping exercises for Data Partners that received financial support through our open calls.