Epidemiological Modelling: Pandemic Intervention
Key area 1
Coordination: The University Medical Center Göttingen
The current pandemic demonstrates the critical importance of dynamic epidemiological surveillance of the spread of infection. At the same time, rapid epidemiological validation of population-based intervention measures, verification of their effectiveness, and development of digital surveillance and reporting systems are in the foreground. In these areas, Göttingen has pioneered work to evaluate the effectiveness of infection epidemiological interventions.
For example, at the beginning of the SARS-CoV-2 pandemic a number of epidemiological modelling initiatives related to SARS-CoV-2 infections and COVID-19 have emerged. These benefit in particular from the Göttingen Campus's focus on data science, e.g., establishment of a number of tenure-track professorships in data science with funding from the federal program (young scientists pact). Moreover, this young and dynamic field is well organized at the Göttingen Campus, on the one hand through the interdisciplinary, cross-faculty Centre for Computational Sciences (ZAI) and Centre for Statistics (ZfS), and, on the other hand, through new structures such as the Sino-German Institute of Social Computing and the Campus Institute Data Science (CIDAS).
Research aim
In the past months, a number of initiatives were already launched to save epidemiological and clinical datasets related to SARS-CoV-2 infections and COVID-19 and to make them accessible to science according to the FAIR principles. The so-called data sciences, including computer science, mathematics, and statistics, are expected to contribute through advanced and new modelling approaches to enable subjects such as medicine and epidemiology as well as social and economic sciences to gain new insights from already existing data. The research approaches cover the following areas.
Systematic reviews and meta-analyses of epidemiological models
A number of models for predicting SARS-CoV-2 infection have been proposed. The models differ not only in their predictions but also in the parameters that need to be specified and the data on which they are based. Nonetheless, systematic collection and assessment are currently lacking. Here, from the user’s point of view, both prerequisites, the need to specify a large number of model parameters and access to relevant data need to be evaluated. Moreover, the statistical properties of the prediction (e.g., bias, uncertainty) are determined. Due to the reasons that the field is dynamic, timely and continuous collection, systematisation, and evaluation are necessary.
Targeted epidemiological surveys
As described above, data are now being collected systematically in many areas. However, gaps remain with regard to specific population groups characterised, for example, by particular risk factors, their living circumstances (e.g., nursing home residents), or their professional activities (as highlighted in the context of outbreaks in the meat processing industry). In addition, more attention is paid to developments in so-called hotspots.
Adaptive infection control
Due to the current pandemic evolution toward highly localised outbreak foci that can develop high dynamics on very short-time scales, epidemiological models of infection dynamics in different settings will be linked to computer-assisted decision support for adaptive surveillance of these areas given current regional prevalence and local incidence as well as the specific risk situation of those potentially affected. The aim is the efficient and effective infection control. Based on the surveillance data, a further step will also provide support for decision-making on measures such as resource allocation for COVID-19 cases or quarantine measures on the basis of a computer-assisted expert system.
Projects with an exemplary character
The following projects are examples of the initiatives for epidemiological modelling at the Göttingen site:
- At the Max Planck Institute for Dynamics and Self-Organization a working group headed by Dr. Viola Priesemann has developed a model for the spread of the SARS-CoV-2 pandemic in Germany, which aims in particular to record the effects of the measures introduced on the spread of the epidemic in Germany.
- At the interdisciplinary Centre for Statistics at the University of Göttingen several applications and draft research study proposals have been formulated for the epidemiological investigation of COVID-19. These initiatives have benefited in particular from the expertise of Professor Thomas Kneib’s working group in the area of spatio-temporal modelling of stochastic processes.
- At the Department of Medical Statistics of the University Medical Center Göttingen (Director: Professor Tim Friede) in mid-March 2020 the "R Shiny" app was developed in a short period of time for local prediction of SARS-CoV-2 infections as well as associated hospitalisations, particularly those involving intensive care. In addition, the Department of Medical Statistics has made several contributions in the area of clinical epidemiology related to COVID-19. Here, for example, a cohort to study renal involvement in COVID-19 (coordinated by Professor Oliver Gross, MD) and some methodological contributions to clinical trials in COVID-19 should be mentioned here.
- The Department of Infection Control and Infectious Diseases and the Department of Medical Informatics have developed a model-based system for the detection of nosocomial infections in cooperation with the Institute for Medical Informatics of the Hannover Medical School (MHH) and the Helmholtz Centre for Infection Research (HZI) and extended this to cover COVID-19 Infections.
Mobile digital monitoring tool called SORMAS
Surveillance, Outbreak Response Management, and Analysis System
The HZI has made a decisive contribution to the monitoring and management of COVID-19 by providing an innovative mobile digital monitoring tool called SORMAS (Surveillance, Outbreak Response Management, and Analysis System), which was developed by Professor Gérard Krause and colleagues in the course of the Ebola pandemic and has been used internationally to combat COVID-19.
In addition to its use in Africa (Ghana, Nigeria), SORMAS was already implemented in several European countries. For national use of SORMAS, the software was specifically adapted to the needs of public health authorities in order to make infection surveillance and contact tracing more efficient.
Coordination HZI
In addition, the HZI is coordinating an international project to collect clinical, epidemiological, and immunological data on which to base further decisions in the COVID-19 crisis.
Together with partners from science and industry, the HZI was able to obtain further important clues for managing the pandemic through mathematical modelling. Here, HZI scientists (Professor Michael Meyer-Hermann) described a method to recalculate the reproduction number at state level on a daily basis in order to provide clues for the further development of the spread of infection. A high-profilestudy together with the ifo Institute clarified the health and economic effects of the restriction measures through mathematical modelling.
Contact
contact information
- telephone: +49 551 3965821
- e-mail address: jwienan(at)uni-goettingen.de
secretariat
- telephone: +49 551 3965812
- e-mail address: anika.schindler(at)med.uni-goettingen.de
contact information
- telephone: +49 551 3965601
- fax: +49 551 3965605
- e-mail address: tim.friede(at)med.uni-goettingen.de
- location: Humboldtallee 32, EG, 124