Bayesian Inference for Sexual Contact Networks Using Longitudinal Survey Data

2025-09-26

AI Summary

This study addresses key challenges in dynamic contact network modeling—including insufficient uncertainty quantification of critical parameters (e.g., relationship formation/dissolution timescales, differential incidental contact rates between single and partnered individuals) and poor identifiability of mobility patterns. We propose a longitudinal inference framework integrating a discrete-time mechanistic model with Approximate Bayesian Computation (ABC). Innovatively, we replace labor-intensive retrospective interviews with simplified binary survey data, improving the accuracy of contact frequency estimation. By jointly analyzing cross-sectional and multi-wave panel data, we substantially mitigate mobility identifiability issues. Applied to longitudinal data from 403 men who have sex with men in Stockholm, our approach estimates median durations for stable relationship formation and dissolution at 25 and 42 weeks, respectively, and incidental contact frequencies of once every 1.8 weeks for single individuals versus once every 4.5 weeks for partnered individuals. These results significantly enhance interpretability and rigorous uncertainty quantification of dynamic processes in network epidemiology.

Technology Category

• Reasoning under Uncertainty: Relational Probabilistic Models
• Planning, Routing, and Scheduling: Scheduling under Uncertainty
• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• User Modeling, Personalization and Recommendation: Studies of user behavior, including longitudinal effects of personalized systems
• Social Networks and Social Media: Social media analysis through the lenses of networks

Abstract

Characterizing sexual contact networks is essential for understanding sexually transmitted infections, but principled parameter inference for mechanistic network models remains challenging. We develop a discrete-time simulation framework that enables parameter estimation using approximate Bayesian computation. The interpretable model incorporates relationship formation, dissolution, concurrency, casual contacts, and population turnover. Applying our framework to survey data from 403 men who have sex with men in Stockholm, we provide principled uncertainty quantification for key network dynamics. Our analysis estimates the timescale for seeking a new steady relationship at 25 weeks and for relationship dissolution at 42 weeks. Casual contacts occur more frequently for single individuals (every 1.8 weeks) than for partnered individuals (every 4.5 weeks). However, while cross-sectional data constrains these parameters, migration rates remain poorly identified. We demonstrate that simple longitudinal data can resolve this issue. Tracking participant retention between survey waves directly informs migration rates, though survey dropout is a potential confounder. Furthermore, simple binary survey questions can outperform complex timeline follow-back methods for estimating contact frequencies. This framework provides a foundation for uncertainty quantification in network epidemiology and offers practical strategies to improve inference from surveys, the primary data source for studying sexual behavior.

Problem

• Estimating sexual contact network parameters using Bayesian inference
• Quantifying uncertainty in relationship formation and dissolution dynamics
• Improving network inference from longitudinal survey data collection

Innovation

• Discrete-time simulation framework for parameter estimation
• Approximate Bayesian computation enables uncertainty quantification
• Longitudinal survey data resolves migration rate identification