Sepsis Data Unit
Principal Investigator: Prof. Dr. Markus Löffler
Team: Dr. Frank Meineke, Dr. Christoph Engel, Florian Rißner, M. Löbe, S. Stäubert, S. Kropf, C. Porzelius, Lars Voitel
Duration: 01.08.2010 - 31.07.2015
The Problem
The core facility provides the basic IT-infrastructure and the basic support for designing and initiating patient related prospective interventional, diagnostic, or prognostic clinical trials. This includes setup and operation of a web-based clinical data management system, tools for a stringent ID-Management and a comprehensive research database able to integrate data from the Jena hospital information system. Regarding clinical trial design, we provide basic support at all levels in designing and initiating trials.
Results so far
Based on a detailed analysis of the necessary infrastructure, we built up a detailed overall integration model based on processes, components and interfaces. A comprehensive data privacy and data security concept guarantees that capture, storage and retrieval will be performed according to the legal requirements for data security. We developed a comprehensive and flexible web-based registry tool for identity-management for PID-registration, SIC-generation and CDMS subject creation. We decided to use OpenClinica as Clinical Data Management Systems (CDMS). OpenClinica features CDISC ODM file format, web-services and an Excel based electronic Case Report Form (eCRF) specification. We developed tools and methods, e.g. to generate annotated CRFs and to incorporate validation rules. The eCRFs for the studies Medusa, Alerts, Eidecs and the Sepsis Registry have been created. Training courses have been held and data entry has started. NeuroSOS-Nerve/NeuroSOS-CIP and NeuroPain started in early June 2012. Action, Osarst and Smooth are in late release phase.
The SDU biometrician supported the investigators in the process of designing and initiating the trials including biometric design, the documentation items and in endpoint discussions and performed intermediate analyses, e.g. Medusa. The server-infrastructure was set up in Leipzig within the audited data center of the IMISE/ZKSL. All components (CDMS, IFB-Registry, ID-Management, Research Datawarehouse i2b2, backend Databases) run on virtual machines and will be transferred stepwise to Jena by 2013. i2b2 allows overarching data queries without deeper knowledge of data structures or query languages. An additional datamart system supports a relational view on all data aggregated by the SDU. Trial data and samples of Jena HIS data can be loaded into datamart and datawarehouse via ETL-tools. Thus making combined analysis, e.g. Alerts with corresponding SAP risk indicators, possible.
Contact
Universität Leipzig
Institut für Medizinische Informatik, Statistik und Epidemiologie (IMISE)
Härtelstraße 16-18
04107 Leipzig