Skip to main content
2025 ISPE Annual Meeting Main banner

Informatics/Digital Epidemiology

Poster Session C

(C-118) Standardization of CLARION study-specific data using a common data model: Experience from multiple sclerosis registries and data sources

Tuesday, August 26, 2025
12:00 PM - 1:30 PM ET

    Co-Author(s)

  • TZ

    Tjalf Ziemssen

    Professor
    Center of Clinical Neuroscience, Department of Neurology, Dresden University of Technology, Dresden, Germany, Germany

  • HB

    Helmut Butzkueven

    Professor
    Department of Neuroscience, Central Clinical School, Monash University, Melbourne, Victoria, Australia, Australia

  • MM

    Melinda Magyari

    Danish Multiple Sclerosis Registry, Copenhagen University Hospital, Rigshospitalet, Denmark, Denmark

  • AG

    Anna Glaser

    Department of Clinical Neuroscience, Karolinska Institute, Sweden, Sweden

  • MS

    Merja Soilu-Hänninen

    Docent
    Turku University Hospital Neurocenter and Division of Clinical Neurosciences, University of Turku, Turku, Finland, Finland

  • MT

    Maria Trojano

    Professor
    Department of Translational Biomedicine and Neurosciences – DiBraiN, University of Bari “Aldo Moro”, Bari, Italy, Italy

  • PI

    Pietro Iaffaldano

    Associate Professor
    Department of Translational Biomedicine and Neurosciences – DiBraiN, University of Bari “Aldo Moro”, Bari, Italy, Italy

  • JK

    Jens Kuhle

    Professor
    Multiple Sclerosis Center and Research Center for Clinical Neuroimmunology and Neuroscience (RC2NB), Departments of Biomedicine and Clinical Research, University Hospital and University of Basel, Basel, Switzerland, Switzerland

  • NM

    Nicholas Moore

    Professor
    Bordeaux PharmacoEpi (BPE), Université de Bordeaux, Bordeaux, France, France

  • NS

    Nick Sicignano

    Senior Epidemiologist
    Health ResearchTx LLC, Trevose, PA, USA, United States

  • LP

    Luigi Pontieri

    Research Statistician
    Danish Multiple Sclerosis Registry and the Danish Multiple Sclerosis Center, Department of Neurology, Copenhagen University Hospital, Rigshospitalet Glostrup, Glostrup, Denmark, Denmark

  • JA

    Jan Harald Aarseth

    Statistician
    Norwegian MS Registry and Biobank, Department of Neurology, Haukeland University Hospital, Bergen, Norway, Norway

  • LF

    Lars Eric Forsberg

    Research Specialist
    Department of Clinical Neuroscience, Karolinska Institute, Stockholm, Sweden, Sweden

  • CZ

    Chao Zhu

    Research Fellow
    Department of Neuroscience, Central Clinical School, Monash University, Melbourne, Victoria, Australia, Australia

  • VM

    Vasili Mushnikov

    Statistical Programmer
    IQVIA, Helsinki, Finland, Netherlands

  • IB

    Irene Bezemer

    Senior Director of Global Epidemiology
    IQVIA, Amsterdam, The Netherlands, Netherlands

  • YK

    Yogesh Kumar Khunte

    Global Safety Lead, Global Patient Safety
    Merck Specialties Pvt. Ltd., Bangalore, India, an affiliate of the healthcare business of Merck KGaA, Darmstadt, Germany, India

    • Presenting Author(s)

  • MS

    Meritxell Sabidó

    Associate Director Epidemiology
    The healthcare business of Merck KGaA, Darmstadt, Germany, Germany

Background: CLARION (EUPAS24484) is a long-term safety study of people with multiple sclerosis (MS) newly initiating either cladribine tablets or fingolimod. In total, CLARION had a target population of 8000 people (4000 per cohort), who were identified from pre-existing data sources, except in Germany where primary data collection was performed. The heterogeneity of data sources warranted the development of a common data model (CDM) with study-specific data to standardize data structures, format, and meaning.

Objectives: To report the process of transforming data from data sources participating in CLARION to CDM, and discuss challenges and benefits.

Methods: A CDM was adopted for CLARION and applied to data sources: MS registries, Nordic national registers, German electronic case report forms, USA medical claims health data, and electronic health records. Each source received a common protocol and statistical analysis plan, standard vocabulary specification, drug codes and mapping tables, and ready-to-use standardized statistical analysis software for data conversion. The CDM terminology dictionary included International Classification of Diseases, Anatomical Therapeutic Chemical, Medical Dictionary for Regulatory Activities, and Healthcare Common Procedure Coding System codes. Source specific rules were used to transform original data that had been converted to a CDM format into the CDM extract programs. Aggregated results generated by a common programming code that uses the CDM as data source were created locally. A collection of datasets were transferred and analyzed centrally producing country-specific and combined results outputs.

Results: Overall, data from 8659 people from nine MS registries and data sources were successfully converted into the CDM, enabling a standardized analysis across data sources. Centrally developed programs removed the redundancy in programming and analyzing data by each partner, facilitated alignment of concepts, and removed study design misunderstandings. CDM conversion and harmonization of coding systems were laborious, yet they enhanced the reproducibility of the results and sped up the analysis process with high involvement of data source partners. The single analytical program also facilitated revision of programming.

Conclusions: The CDM allowed us to harmonize heterogeneous data sources to enable combined statistical analyses across sources with diverse data collection systems while enhancing data source autonomy and scientific independence. This will increase the traceability of results and confidence in evidence generated by CLARION.