Mutual Feedback Between Epidemic Spreading and Information Diffusion

The impact that information diffusion has on epidemic spreading has recently attracted much attention. As a disease begins to spread in the population, information about the disease is transmitted to others, which in turn has an effect on the spread of disease. In this paper, using empirical results of the propagation of H7N9 and information about the disease, we clearly show that the spreading dynamics of the two-types of processes influence each other. We build a mathematical model in which both types of spreading dynamics are described using the SIS process in order to illustrate the influence of information diffusion on epidemic spreading. Both the simulation results and the pairwise analysis reveal that information diffusion can increase the threshold of an epidemic outbreak, decrease the final fraction of infected individuals and significantly decrease the rate at which the epidemic propagates. Additionally, we find that the multi-outbreak phenomena of epidemic spreading, along with the impact of information diffusion, is consistent with the empirical results. These findings highlight the requirement to maintain social awareness of diseases even when the epidemics seem to be under control in order to prevent a subsequent outbreak. These results may shed light on the in-depth understanding of the interplay between the dynamics of epidemic spreading and information diffusion.

💡 Research Summary

The paper investigates the bidirectional interaction between epidemic spreading and disease‑related information diffusion. Using real‑world data from the 2013‑2014 H7N9 avian influenza outbreak in China and corresponding micro‑blog (Weibo) posts, the authors first demonstrate that the temporal patterns of reported cases and online discussions are closely coupled: peaks in infection coincide with spikes in information, and a reduction in information precedes a larger second wave of infections.

To capture this interplay, the authors construct a network model in which each individual can be in one of four combined states: susceptible‑unaware (S⁻), susceptible‑aware (S⁺), infected‑unaware (I⁻), and infected‑aware (I⁺). Both disease transmission and information spreading follow SIS dynamics. Disease spreads with transmission probability β and recovery probability γ; information spreads with probability α, while aware individuals may forget the information with rates λ (for susceptibles) and δ (for infected). Awareness reduces infection risk (σ_S for S⁺, σ_I for I⁺) and increases recovery (factor ε>1).

Analytical treatment proceeds in two steps. A classical mean‑field approach yields differential equations for the overall infected fraction I and informed fraction Info, but it neglects network correlations. Therefore, the authors also develop a pairwise approximation that tracks the evolution of node‑pair states, providing a more accurate description of the system’s dynamics.

Simulation experiments are performed on random graphs (N=10 000, average degree 15) across a range of α (0–0.6) and β (0.1–0.5). Results show that higher α leads to a larger informed population, which in turn raises the epidemic threshold β_c dramatically (e.g., β_c≈0.012 when α>0.01, compared with β_c≈0.006 when α=0). Consequently, the final infected proportion and the speed of spread are markedly reduced. Conversely, when α is low and β high, the model reproduces a multi‑outbreak pattern: after an initial wave, information decays, the population becomes less vigilant, and a second, larger outbreak occurs—mirroring the empirical H7N9 observations.

The study highlights several key insights: (1) information diffusion acts as a non‑pharmaceutical intervention by inducing protective behavior; (2) the effectiveness of this intervention depends on both the diffusion rate α and the persistence parameters λ, δ; (3) maintaining a high level of public awareness even after an epidemic appears under control is crucial to prevent resurgence.

Limitations include the use of a static random network, homogeneous transmission parameters, and the omission of factors such as network heterogeneity, temporal contact changes, and the quality or emotional tone of information. The authors suggest future work on multiplex or adaptive networks, dynamic contact patterns, and incorporating trust or misinformation dynamics.

Overall, the paper provides a rigorous quantitative framework linking disease spread and information flow, supported by real data, and offers actionable guidance for public‑health policy: sustained information campaigns can substantially raise the epidemic threshold and mitigate both the size and speed of outbreaks.