Chronic Disease Emergence in Coevolving Social Networks via Behavioral Feedback and Demographic Turnover

Chronic diseases emerge within social systems where individuals reorganize their ties, adapt their behaviors, and turn over through demographic processes; yet, how these coupled dynamics shape long-run disease trajectories remains poorly understood. We developed an agent-based model of how the network structure, behavioral adaptation, and population renewal jointly generate cardiovascular disease incidence in an evolving population. The model integrated age-dependent intrinsic risk from the Framingham Risk Score framework, behavioral transmission through social ties, homophily-driven network formation, and demographic turnover across simulated populations of up to 420 agents and horizons of 18-240 months. Highly connected agents showed lower simulated incidence than weakly connected agents (adjusted hazard ratio 0.43; 95% confidence interval 0.33, 0.57; top-tertile degree indicator) after adjustment for age, sex, smoking, blood pressure, body-mass index, physical activity, and diet. Individual disease probability rose with ego-network disease prevalence (Pearsons r = 0.703, p < 0.001). Global sensitivity analysis identified baseline incidence, activity level, and behavioral transmission as dominant drivers (78% of variance). Bayesian calibration concentrated within the empirical acceptance region, with posterior standard deviation of 12% of prior. These results point toward a systems-level understanding of non-communicable disease emergence.