Forecasting the dynamics of flu epidemics could be vital for policy-making concerning the allocation of public health resources. Reliable predictions about disease transmission networks also help fix the benchmark for reconciling diverse aspects to the decision-makers while selecting and implementing a suitable health intervention. To this aim, we propose an $SIR/VM$ epidemic game model to reveal the dynamic evolution of intervention policies, entangling social feedbacks, behavioral responses, and viral transmission on several network topologies into a single framework. Besides vaccination, this study introduces intermediate defense measures (IDM) as an alternative provision to restraint the epidemic resurgence in structured and unstructured populations. Although heterogeneity is a commonly observed phenomenon in human populations, many antecedent studies typically preferred homogenous networks. Here, we investigate the disparities found in epidemic diffusion within homogeneous and heterogeneous networks, employing a mean-field approximation and multi-agent simulation, respectively. Additionally, we explore both network types simultaneously and justify their potential impacts on control provisions’ success. As a general tendency, vaccination and IDM complement each other within the entire parametric regions. Our study elucidates the coexistence of multiple policies as well as the abrupt emergence of stain points adopting several network topologies. A careful investigation on stain points reveals that hub agents solely rely on free-riding brings the endemic state of an epidemic, triggered by a sudden extinction of vaccinators and self-protectors. The emergence of too many self-interested people spoils the herd immunity state and initiates the outbreaks, heavily observed in well-mixed and scale-free networks. On the other hand, the coexistence of policies occurs mostly in the networks mentioned above but rarely seen in the lattice network. The robustness of the proposed model has been tested by adding mean-field theoretical results for a nonspatial population and agent-based simulated outcomes for spatial populations under a wide variety of parametric conditions. Model outcomes confirm that the game-payoff regulates the epidemic dynamics, while the epidemic propagation governs individuals’ health status. Moreover, we expose a complex interplay between cost and efficacy of control provisions and justify keeping the provisional costs reasonably lower would be an ultimate challenge to maintain disease attenuation. The central theme of this paper is thus to portray the holistic summary of epidemic prevalence and the relative contribution of each intervention to epidemic remission.

预测流感流行的动态可能对有关公共卫生资源分配的决策至关重要。关于疾病传播网络的可靠预测还有助于确定基准，以便在选择和实施适当的健康干预措施时协调决策者的各个方面。为此，我们建议$\mathrm{小号}\mathrm{一世}\mathrm{[R}/V\mathrm{中号}$流行病博弈模型，以揭示干预政策的动态演变，将社会反馈，行为反应和病毒传播在几种网络拓扑结构上整合到一个框架中。除接种疫苗外，本研究还引入了中间防御措施（IDM）作为限制结构性和非结构性人群流行病复发的替代措施。尽管异质性是人类人群中普遍观察到的现象，但是许多先前的研究通常首选同质网络。在这里，我们调查均质场近似和多主体仿真分别在同质网络和异构网络内的流行扩散中发现的差异。此外，我们同时探索了两种网络类型，并论证了它们对控制规定成功的潜在影响。作为一般趋势，疫苗接种和IDM在整个参数区域内相互补充。我们的研究阐明了多种策略的共存以及采用几种网络拓扑结构的染色点的突然出现。仔细研究染色点后发现，集线器仅依靠搭便车导致流行病流行，这是由疫苗和自我保护剂的突然灭绝引起的。太多有私心的人的出现破坏了畜群的免疫状态，并引发了暴发，这是在混合良好且无规模的网络中大量观察到的。另一方面，策略的共存主要发生在上述网络中，而很少出现在网格网络中。通过添加非空间人口的均值理论结果和空间参数在多种参数条件下基于智能体的模拟结果，测试了所提出模型的鲁棒性。模型结果证实，博弈的收益调节着流行的动态，而流行的传播支配着个人的健康状况。而且，我们暴露了控制措施的成本和功效之间的复杂相互作用，并且证明保持合理的临时成本合理是维持疾病恶化的最终挑战。因此，本文的中心主题是描绘流行病流行的整体摘要以及每种干预措施对流行病缓解的相对贡献。