Expanded Game Theory Predictions of Adherence to Multiple Nonpharmaceutical Interventions

Abstract

Individual decision-making has increasingly been recognized as a critical component of epidemiological models. The benefits of nonpharmaceutical interventions (NPIs) such as mask-wearing and social distancing are particularly vulnerable to individual decision-making because of the ease with which people may switch between adherence and nonadherence. This study builds upon a known hypothetical two-strategy game of adherence and nonadherence to NPIs during a disease outbreak. It adds a third strategy such that players can choose between complete nonadherence, adherence to a weak and cheap NPI, and adherence to a strong and costly NPI. This study also introduces varying timescales of societal and individual dynamics and related algebraic analysis. The central question asks how individuals make decisions when presented with multiple NPIs, and how those decisions impact society. This study determines that when disease levels are fixed, people may choose a weaker strategy at the expense of average population fitness. When NPI decisions are fixed, either a disease-free or disease-endemic equilibrium will be stable, but never both. When neither the disease levels nor the NPI decisions are fixed, this study computationally determines that up to three strategies can exist at a stable equilibrium. This equilibrium forms through different trends regarding disease transmission, NPI efficacy, cost of disease, and cost of NPIs. Notably, in three-strategy equilibria, initial societal conditions determine the final frequency of strategies but not infectives. This research has implications for public health efforts that aim to change NPI behavior and/or disease dynamics in complex social systems.