Self-Referential Multi-Scale Modelling and Simulation of Severe Infectious Diseases

During the COVID-19 pandemic, modelling has been utilized extensively worldwide and in Germany. Nonetheless, the lack of validity and credibility was often criticized. We hypothesise that public discussion of forecasts leads to behavioural changes in the population. Individual behaviour, public communication and contagion dynamics form a self-referential system that is not sufficiently considered in current forecasts approaches. The project SEMSAI explores how model-based predictions can be adjusted to mirror real life more accurately by integrating the feedback mechanisms into modelling. Therefore, we ask: (1) What factors predict to what extent preventive behaviour during different stages of the COVID-19 pandemic in Germany? (2) How can public behaviour and the influential factors be integrated into modelling? Based on a literature review, we hypothesise that perceived severity, susceptibility, self-efficacy, response efficacy, attitudes, social norms, trust and knowledge have an impact on preventive behaviour and vaccination (intention). Using selected variables of the COSMO study and calculating multiple linear regression, we will identify relevant predictors and their impact on public behaviour and explore the possibility of directly incorporate the calculated estimates into the models. We also plan to utilize the COSMO raw data to examine whether clustering based on demographic data is possible. unknown other