In the context of achieving Universal Health Coverage, the allocation of healthcare resources faces complex decision-making challenges. A more scientifically sound and effective strategy is needed to balance the dynamic interplay between medical insurance cost control, clinical drug efficacy, and patient expenditure. Comorbid psychiatric disorders, as refractory diseases with high treatment costs, impose a heavy burden on both society and patients’ families. This study proposes a Bi-Objective Optimization Model for the Burden of Depression and Anxiety (BDAA), aiming to optimize disease burden at both societal and household levels. To address the high-dimensional and non-convex model, a tailored solution framework named PSO for Burden of Depression and Anxiety (BDAAPSO) which is based on swarm intelligence algorithms is designed. The framework employs a tripartite population strategy for division of labor and collaboration, incorporating a particle adjustment mechanism to enhance both convergence speed and solution diversity in Pareto front exploration. The proposed burden optimization approach provides a scientifically grounded and practically viable decision-making tool for mental health resource allocation, contributing to the dynamic equilibrium between medical equity and fund sustainability under the Sustainable Development Goal 3.