Together with cooperation partners of the Institute for Biometrics and Epidemiology at the German Diabetes Center in Düsseldorf, the Biostatistics Working Group is researching both statistical-methodological projects and in the field of diabetes epidemiology. We develop statistical methods for meta-analyses, especially of diagnostic accuracy studies. These serve to evaluate new diagnostic procedures and to determine their optimal threshold values. Furthermore, we participate in research projects in the field of survival time analysis, which are concerned with the development of new statistical methods for the evaluation of medical studies.
In the field of diabetes epidemiology, we participate in modeling studies to estimate important epidemiological indicators such as prevalence, incidence and mortality. The basis for this is a differential equation that relates prevalence to incidence and the corresponding mortality rates. Based on this equation, it is also possible to determine future case numbers under assumptions about trends in incidence and mortality rates. These represent an important indicator of the future utilization of the health care system. In addition, projections of the costs incurred are also possible.
Together with cooperation partners from the Division of Diabetes Translation of the Centers for Disease Control and Prevention (CDC), the Biostatistics Working Group is researching specific problems of diabetes epidemiology in the US. Of particular interest is the impact of changes in diagnostic thresholds on the incidence of diabetes and its late complications such as loss of kidney function. Furthermore, the modeling of future type 2 diabetes cases among US adolescents is part of our collaborative projects.
The Working Group Biostatistics is involved in the national diabetes surveillance established by the Robert-Koch-Institute (RKI). Together with colleagues of the RKI we are dealing with the question of the influence of certain risk factors on future diabetes case numbers. Special attention is paid to the extent to which changes in risk factors are reflected in future case numbers. Our modeling studies provide a basis for the evaluation of prevention measures. Furthermore, we deal with the life years lost through diabetes and their future development.
Together with colleagues from the Department of Statistics at the LMU, the Working Group Biostatistics participates in Covid-19 research. The fields of application of biostatistics and its use in epidemiology are manifold. We explicitly deal with the estimation of the number of unreported cases („dark figure“) of infection, as well as with modelling the various reported epidemiological indicators (e.g. incidence, effective reproduction number, absolute case numbers). We could show that the effective reproduction number is the least susceptible to the underlying test behavior ("bias by incomplete case-detection"). Therefore, instead of using the cumulative number of cases, rate or absolute number of positive tests, we recommend that this key figure should preferably be used for communicating and comparing the course of the pandemic with other countries or other points in time.