The fourth dimension: combining epidemiologic and typing data for the transition from data to knowledge for infectious diseases

摘要

Epidemiologic data (time, place, and person/animal) in combination with informatics and molecular laboratory techniques (type, i. e. the fourth dimension) makes fast and accurate early warning outbreak detections for infectious diseases on different geographic levels possible.We have shown that rapid MRSA outbreak detection, based on epidemiological and spa typing data, is a suitable alternative of classical approaches (ICP) and can assist in the identification of potential sources of infection in a hospital (Mellmann et al., 2006). We have also shown in an international multicenter study that high inter-laboratory reproducibility of DNA sequence-based typing of bacteria can be achieved due to the unambiguous nature of sequence data. By using dedicated client/server software, Ridom StaphType (Harmsen et al., 2003), a worldwide uniform terminology ("molecular Esperanto") can be ensured. Only high-quality sequence data are automatically accepted by the server (www.SpaServer.ridom.de) and, therefore, no curator is needed for administration of the database. Furthermore, we are currently busy in a concerted action of European laboratories to build capacities and to harmonize technology for sequence-based typing of microorganisms (www.SeqNet.org).The German National Reference Centre for Meningococci (NRZM, Widrzburg) stores information on analyzed meningococcal samples in a central database. The recorded information includes high-resolution typing data, obtained by serogrouping and epitope sequence typing of porA and fetA. We have assembled a server that receives an anonymized subset of the NRZM data. A PostgreSQL database stores the epidemiological information as well as additional static data, such as geographical borders or population figures for counties and federal states. Our custom developed software combines and controls the database and additional open source software components (UNM MapServer and OpenLaszlo) to build an epidemiological geographical information system (GIS). The user accesses the automatically generated maps via the Internet, using a Flash-based application (www.EpiScanGis.org). The server utilizes SaTScan (developed by M. Kulldorff et al.; www.satscan.org) to detect significant spatio-temporal clusters, taking the typing-, epidemiologic-data and population-at-risk into account. The SaTScan output is finally visualized within the GIS by depicting significant cluster of cases within the maps (Reinhardt et al., 2008).As a last example, we are currently busy in the framework of the BMBF funded network Food-Borne Zoonotic Infections of Humans (FBI-Zoo; www.fbi-zoo.de) to establish a client-server based data warehouse for the interdisciplinary consortium that will lay the foundation for more advanced analytical options in the future.Thus, the application of interactive, Internet-based tools can help achieving better quality control and faster cluster detection and allows for turning surveillance data into knowledge. Community building in an environment of mutual trust and sustainability of such services is crucial for long-term success.