About the project

RAPID (Registry of Adult and Pediatric Intensive Care Data) is a decentralised, federated registry for routine intensive care data from adult and paediatric intensive care units in Germany. Data is collected automatically via digital interfaces directly from the intensive care information systems of the participating sites.

In contrast to traditional, centralised registers, the data in NUM RAPID remains local to the sites. Research enquiries are distributed to the local sites in a federated manner and automatically analysed there. In this way, only the aggregated results required for a specific question are passed on. This approach enables highly granular data usage while at the same time complying with data protection and data minimisation requirements.

NUM RAPID thus creates the basis for multi-centre clinical research in intensive care medicine, healthcare research, quality comparisons and improved pandemic and crisis prevention.

The most important facts at a glance

  • Development of a decentralised, federated intensive care register
  • Automated, comprehensive collection of routine intensive care data from intensive care information systems
  • Enabling multi-centre clinical, epidemiological and health system-related research
  • Investigating the relationship between quality indicators and patient outcomes
  • Establishment of cross-site benchmarking to support quality improvement processes
  • Provision of data for pandemic and crisis preparedness
  • High data protection requirements for federated evaluation
  • Highly heterogeneous manufacturer solutions for intensive care information systems at the sites
  • Ensuring standardisation, comparability and clinical interpretability of data

NUM RAPID builds on existing NUM infrastructures, in particular the AKTIN platform for acute, intensive care and emergency medicine, and is closely interlinked with the NUM Data Integration Centres.

The implementation will take place in several steps:

  • Clinically-driven definition of a standardised intensive care dataset in close collaboration between clinicians and data specialists
  • Formalisation of the data elements and development of corresponding FHIR profiles
  • Initial focus on a minimum viable product to capture the DIVI quality indicators and relevant outcome variables
  • Federated analysis of the data via distributed queries to the local RAPID databases

With this approach, NUM RAPID makes a key contribution to the sustainable utilisation of routine intensive care data for research, care and crisis management.