Epidemic variability in hierarchical geographical networks with human activity patterns

Area of Science:

• Epidemiology
• Network Science
• Computational Social Science

Background:

• Human behavior statistics, like non-Poissonian patterns, are linked to hierarchical geographical network structures.
• Understanding epidemic dynamics in these complex networks is crucial for public health.

Purpose of the Study:

• To investigate the impact of homogeneous time delay (HOTD) and heterogeneous time delay (HETD) on epidemic spreading speed and outbreak variability.
• To analyze how network hierarchy and geographical distance influence disease transmission dynamics.

Main Methods:

• Simulations of epidemic spreading on hierarchical geographical networks.
• Comparison of two distinct contact patterns: HOTD and HETD.
• Analysis of spreading speed, outbreak variability, and the effect of initial seed location.

Main Results:

• HETD significantly slows epidemic spread compared to HOTD and null models, inducing multi-modal outbreak patterns.
• HETD leads to lower overall outbreak variability but sustained peaks, complicating long-term predictions.
• High-layer seeds accelerate spreading; HETD shows increased variability with high-layer seeds, unlike HOTD.

Conclusions:

• Contact patterns, particularly HETD, critically influence epidemic spreading dynamics in hierarchical networks.
• The interplay between time delay, network hierarchy, and seed location introduces significant unpredictability in epidemic forecasting.
• Findings highlight the need for nuanced models that account for geographical and temporal complexities in disease spread.