gms | German Medical Science

GMDS 2012: 57. Jahrestagung der Deutschen Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie e. V. (GMDS)

Deutsche Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie

16. - 20.09.2012, Braunschweig

Integration of web-based Randomization- and EDC-Systems

Meeting Abstract

Search Medline for

  • Daniel Schrimpf - DKFZ, Biostatistik C060, Heidelberg, Deutschland
  • Lothar Pilz - University of Heidelberg, Medical Faculty Mannheim, Heidelberg, Deutschland

GMDS 2012. 57. Jahrestagung der Deutschen Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie e.V. (GMDS). Braunschweig, 16.-20.09.2012. Düsseldorf: German Medical Science GMS Publishing House; 2012. Doc12gmds098

DOI: 10.3205/12gmds098, URN: urn:nbn:de:0183-12gmds0988

Published: September 13, 2012

© 2012 Schrimpf et al.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License ( You are free: to Share – to copy, distribute and transmit the work, provided the original author and source are credited.



Introduction: Randomization – allocation of patients by chance – is used in clinical trials to reduce unknown influencing effects and to prevent selection bias, essential to guarantee the comparability between treatments. A further development of classical randomization methods are adaptive procedures, using influencing factors and therapy results to decide the primary study endpoint with fewer patients. In clinical trials documentation is carried out more often with “Electronic Data Capture”-systems (EDC) making study data directly available. In case of adaptive randomization methods the electronic usage of these data is highly preferable. A concept of integration of a randomization system (RS) and an EDC-system is shown.

Material and Methods: In the integration process different aspects have to be considered: (i) different types of interfaces between systems (e.g. web-interfaces for online systems); (ii) data formats for clinical trials like standards from the “Clinical Data Interchange Standards Consortium” (CDISC), e.g. the CDISC-ODM (Operational Data Model) standard. The development of software in clinical trials demands also the consideration of legal requirements, like the FDA 21 CRF 11, GCP-V and EudraLex Annex 11. EDC-systems use mostly a central database and interfaces to external systems, like laboratory- or PAC-systems and also have a web-interface giving access to enter and view the study data.

Results: Analysis of the required interfaces between the RS and the EDC-system shows, that the EDC-system must offer the possibility to provide study metadata (duration, amount of patients, captured patient characteristics, etc.) in a suitable manner, such that the RS can import this information without double data entry by the user. The metadata are needed to configure the randomization method in the system. Indispensable is another interface for the EDC-system to provide a selection of patient data, such that the randomization process can be adapted. This communication is operated via standardised interfaces and data formats. In the example of web-applications these are – among others – SOAP- and REST-interfaces yielding CDISC-ODM messages. As a consequence of this concept the RS has to offer the possibility to request the required data and to interpret the data format. Furthermore, the RS has to be easily adaptable on varying study situations and new randomization methods. Since the interfaces can be accessed directly by the two systems no user interaction – after configuration – for the randomization process is necessary. Thus, the data entry can take place exclusively in the EDC-system, and afterwards the assignment of the patient is completed by the RS.

Discussion: Integration of RS and EDC-systems allows, particularly for adaptive randomization methods, the usage of the study data in a unit manner for the randomization. Only one system is employed for the data entry and the further process takes place automatically between the systems. Thereby, adaptive procedures become practically feasible, the data quality increases, and the training effort decreases. As the RS is implemented independently, a high flexibility is achieved, e.g. for the adoption of randomization methods and the development of new procedures. Currently this concept is implemented by the aid of freely available software.