[arXiv, 1 August 2023]

During epidemics, people will often modify their behaviour patterns over time in response to changes in their perceived risk of spreading or contracting the disease. This can substantially impact the trajectory of the epidemic. However, most infectious disease models assume stable population behaviour due to the challenges of modelling these changes. We present a flexible new class of models, called behavioural change individual-level models (BC-ILMs), that incorporate both individual-level covariate information and a data-driven behavioural change effect. Focusing on spatial BC-ILMs, we consider four "alarm" functions to model the effect of behavioural change as a function of infection prevalence over time. We show how these models can be estimated in a simulation setting. We investigate the impact of misspecifying the alarm function when fitting a BC-ILM, and find that if behavioural change is present in a population, using an incorrect alarm function will still result in an improvement in posterior predictive performance over a model that assumes stable population behaviour. We also find that using spike and slab priors on alarm function parameters is a simple and effective method to determine whether a behavioural change effect is present in a population. Finally, we show results from fitting spatial BC-ILMs to data from the 2001 U.K. foot and mouth disease epidemic.