EARLY WARNING INDICATORS OF EPIDEMICS ON A COUPLED BEHAVIOUR-DISEASE MODEL WITH VACCINE HESITANCE AND INCOMPLETE DATA

Abstract

The ubiquity of hesitance sometimes defies social pressure, especially when individuals are required to make health-care decisions they deem momentous. In epidemiology, such intervention hesitance can both initiate and prolong infectious disease outbreaks, especially when paired with vaccine denial. Previous literature has yielded effective early warning signals (EWS) of disease outbreak and vaccine crisis for coupled behaviour-infection systems; these EWS arise from characteristic phenomena undergone by model dynamics during critical transition(s). In this study, we investigate the resilience of these EWS to the incorporation of a destructive delay to vaccination decisions. Simulations were conducted on a static small world network, using a model coupling an SIRVp infection model with a social dynamic resembling a voting game with abstention. We find that some of the EWS tested retain their efficacy despite fundamental differences in model behaviour. We also find that these EWS (both pair- and cluster-based) can be reliably used while observing as little as 60% of the total network with relatively small loss of accuracy. These findings not only show the resilience of these EWS, but also allow for potential expansion of use cases and reduction in computational resource requirements.