Estimating contact network properties by integrating multiple data sources associated with infectious diseases.

Area of Science:

• Epidemiology
• Network Science
• Computational Biology

Background:

• Contact network structure significantly impacts communicable disease transmission and control program effectiveness.
• Accurate measurement of contact networks is crucial for efficient resource allocation in public health.
• Existing methods for network estimation face challenges in precision and accuracy.

Purpose of the Study:

• To develop and assess a Bayesian approach for more precise and accurate estimation of contact network properties.
• To integrate multiple data sources, including epidemiological, viral genetic, and risk behavior survey data.
• To evaluate the performance of the proposed method using simulations and real-world data.

Main Methods:

• A Bayesian framework integrating multiple data sources for contact network estimation.
• Utilizing congruence class models for network representation.
• Conducting simulation studies with pathogens similar to SARS-CoV-2 and HIV.
• Applying the approach to HIV transmission data from a clinical consortium.

Main Results:

• Integration of epidemiological, viral genetic, and risk behavior survey data significantly reduces mean squared error (MSE) in contact network estimates.
• The proposed method demonstrates improved accuracy even when risk behavior surveys contain measurement error.
• Simulation studies identified specific scenarios where the integrated approach may not yield substantial improvements in MSE.

Conclusions:

• The developed Bayesian approach offers a powerful tool for enhancing the accuracy of contact network estimation by integrating diverse data streams.
• This improved network understanding can lead to more effective and resource-efficient strategies for mitigating communicable disease spread.
• Further research is warranted to explore the limitations and optimal application settings of this integrated data approach.