gms | German Medical Science

Introduction: The Portal of Medical Data Models (MDM Portal) [1] is a recognized German [2] and European [3] research infrastructure for medical research. To our knowledge, it constitutes Europe’s largest open-access platform for medical metadata stored in documentation forms. Currently (March 2021), it contains >24,360 data models and 591,585 items in CDISC ODM format [4]. 90.6% of the items are annotated via UMLS codes [5].

According to the German Science Council, a research infrastructure of national relevance should be involved in the training of young scientists and applied in university teaching [6]. Since 2019, the MDM Portal has been integrated as teaching module of the HiGHmed consortium within the Medical Informatics Initiative (module name: “Semantic analyses of medical data models”) [7]. Over the past five years, several MDM workshops have been organized together with relevant German networks such as TMF, DVMD and GMDS. International congress workshops have also been carried out. Several hundred stakeholders of medical research and data management were trained. Due to impossible face-to-face events, an online format was developed for our TMF workshop in November 2020.

Methods: Eight video tutorials [8] with a total duration of about 45 minutes and exercises have been created concerning the following topics:

1.

Search of data models

2.

Information and functions

3.

Download of data models

4.

Search of items in the MDR [9]

5.

Development of data models with the ODMEdit tool [10]

6.

Semantic annotation with UMLS-Codes

7.

Upload of data models

8.

Semantic comparison by using the CDEGenerator [11]

The material was sent to the 79 participants of our workshop with the request to solve the exercises by using the tutorials. During the online workshop, the content and functions of the portal were explained and questions and exercises were discussed in small groups. Solutions for standardized data management and exchange within two German research networks were presented. Feedback was collected immediately after the workshop via an online-questionnaire consisting of 13 closed-ended and 2 open questions. 3 questions assessed the workshop, the rest related to demographics, research background, use and acceptance of the MDM Portal. Answers were analyzed with descriptive statistics.

Results: 37 participants successfully completed the exercises by using the prepared material. 40 participants took part in the online-survey (response rate = 50.6%). 70.0% of the respondents (n=28) stated that the workshop’s content was extremely or very helpful. 27 respondents thought that the video tutorials and exercises were extremely or very helpful (responses 4 and 5 in 5-point Likert scale were grouped), 7 found them to some extent helpful, 6 gave no evaluation.

Discussion and conclusion: Due to the high level of acceptance of our online MDM workshop, which was based on self-determined, individualized learning by using video tutorials and exercises, we are planning further cooperation with interested institutions and universities. The prepared material and the recording of the workshop are going to be used for further training in the field of medical data science addressing challenges and solutions for standardization in medical research.

The authors declare that they have no competing interests.

The authors declare that an ethics committee vote is not required.