Social dynamics are an integral part of the spread of disease affecting contact rates as well as the adoption of pharmaceutical and non-pharmaceutical interventions (NPIs). This talk will present behavioural-epidemiological models that feature tipping-point dynamics for the uptake of vaccines or adoption of NPIs that combine the risk of infection, perceived cost of the vaccine/NPI, and social payoffs for deviating from the decision making of others. The social payoffs are derived from a social norm of conformity, and they create a collective action problem. A key finding driven by this dilemma is that use of protective behaviours and infections can cycle due to inefficient and delayed behavioural changes. This results in a nonlinear response of the infection load to key epidemiological parameters. e.g., an intermediate transmission rate can result in greater prevalence of disease relative to more or less transmissible diseases. Further, global information about the prevalence of the disease and behaviour of others can increase the infection load and peak relative to information restricted to individuals’ contact networks. Thus, decisions driven by local information can mitigate the collective action problem across the population. Finally, the optimal public policy program to promote disease mitigating behaviours is shown to be one that focuses on overcoming social inertia at the start of an epidemic.