To comprehensively understand integrated disaster risk, international organizations such as the United Nations Office for Disaster Risk Reduction (UNDRR) and the United Nations Development Programme (UNDP) have incorporated disaster risk reduction measures into national planning and decision-making processes based on metric frameworks (23). The discourse on Disaster Risk Reduction (DRR) is undergoing a paradigm shift toward vulnerability-oriented approaches, with vulnerability emerging as a common evaluative characteristic in numerous risk assessments, providing practical information for accurate disaster prevention and mitigation (24).

Scholars have conducted in-depth research on integrated risks of infectious disease disasters. For instance: Mete et al. employed three risk factors from the INFORM COVID-19 Risk Index—hazard and exposure, lack of coping capacity, and vulnerability—to reassess national disaster risks in two phases (25); Pang et al. developed a disaster loss index model based on vector vulnerability, disaster-prone environmental instability, hazard intensity, disaster prevention capacity, and emergency response capability to study pandemic transmission’s environmental risks and socioeconomic impacts (26); Pluchino et al. established a risk index framework incorporating disease hazard (H), regional exposure (E), and population vulnerability (V) to assess epidemiological risks across geographical areas and identify high-risk zones (27); Kanga et al. created an integrated risk assessment framework combining hazard and vulnerability, defining infectious disease risk as C=H × V, followed by risk assessment and mapping (28). In summary, risk index evaluations primarily focus on disease risk, hazards, and vulnerability. By comprehensively considering multiple risk factors and their impacts, more effective risk assessment and management can be achieved. When constructing the comprehensive risk assessment index system for infectious disease disasters in this study, it becomes necessary to redefine these two subsystems—hazard and vulnerability.

The transmission intensity of infectious diseases determines both the likelihood of disease occurrence and the extent of its spread, necessitating the selection of indicators that can characterize disease transmission patterns as hazard factors. Analysis of viral epidemiological characteristics (29, 30) reveals that transmission routes primarily include aerosol transmission, airborne transmission, and direct contact transmission, with influencing factors being highly complex (31). The emergence and spread of infectious diseases are associated with several determinants, encompassing both anthropogenic factors (e.g., population density, travel and trade patterns, susceptibility across different demographic groups) and ecological factors (32). Therefore, this study extracts hazard-related influencing factors from the following dimensions: population characteristics (33), population aggregation (34), demographic dynamics (35), and environmental factors (36). These elements collectively form the framework for constructing the indicator system.

Epidemic disasters differ from natural disasters in that they primarily affect human health through interpersonal transmission and lead to lasting socioeconomic consequences. As a result, pandemic risk assessment studies tend to focus more on the vulnerability of populations and socioeconomic systems while often neglecting spatial considerations. However, many drivers of pandemic vulnerability are inherently linked to global connectivity and urbanization levels, arising from the complex interplay of spatial structural imbalances, uneven economic development, and insufficient governance capacity. Any deterioration in these factors may increase a city’s vulnerability and risk (37). In this study, we define vulnerability as the sensitivity of urban systems to external disturbances and their lack of coping capacity, which makes their structure and function prone to change.

To construct an urban vulnerability indicator system for public health emergencies, we conducted a comprehensive review of relevant literature, including the Population Vulnerability Index widely used in public health and medical fields (38), the Social Vulnerability Index (SVI) (39), urban vulnerability assessments (UVA) that incorporate both social and physical factors in local planning (40), and the Pandemic Vulnerability Index (PVI) (41). We extracted key influencing factors on urban vulnerability from socioeconomic (42), spatial-environmental (43), and infrastructural dimensions (44) to build our indicator system.

The establishment of a risk coupling assessment model for infectious disease hazards and urban vulnerability can effectively measure the threat level of infectious diseases and the degree of urban vulnerability, identify risk-influencing factors, and subsequently formulate targeted epidemic prevention and urban planning strategies. This provides crucial scientific support for disaster prevention and mitigation as well as resilient city development. To explore potential variables influencing infectious disease hazards and urban vulnerability, this study referenced variables included in previous research. Based on principles of data relevance, availability, and reliability, we screened and categorized key indicators to construct a comprehensive integrated risk assessment framework (Table 3).

In a multi-indicator evaluation system, different indicators may have varying units of measurement. Therefore, data standardization is required during the evaluation process. There are two types of evaluation indicators: positive and negative. For positive indicators, higher values indicate greater risk and vulnerability; for negative indicators, higher values indicate lower risk and vulnerability. Consequently, this study employs the extremum method to conduct positive transformation of all original indicators (Equations 1–2).

Positive indicators: (1) X ij = x ij − x min x max − x min ( i = 1 , 2 … , m ; j = 1 , 2 … , n ) Negative indicators: (2) X ij = x max − x ij x max − x min ( i = 1 , 2 … , m ; j = 1 , 2 … , n )

Where x_ij is the original data of the evaluation index; x_max and x_min are the maximum and minimum values of the evaluation index; X_ij is the indicator value after standardised processing. Here, i refers to the prefecture-level and above cities in the study, totaling m = 18. j represents the various indicators.

The entropy method objectively determines indicator weights by measuring information entropy to quantify data variability, effectively eliminating biases inherent in subjective weighting approaches (45). In information theory, entropy serves as a metric for system disorder and the amount of useful information contained within datasets. When evaluation objects demonstrate significant disparities in specific indicators, lower entropy values indicate greater informational utility, warranting higher weight assignments (46). The methodological procedure involves: standardizing raw data, calculating information entropy for each indicator, and deriving weight coefficients based on entropy values (47). This process rigorously accounts for relative importance among indicators, ensuring scientifically robust weight allocation. For public health risk assessment, the entropy method proves particularly effective in handling multi-source heterogeneous data, precisely capturing each risk factor’s actual contribution to support comprehensive evaluations (Equations 3–7).

The feature proportion of the i-th city under the j-th indicator can be defined as follows: (3) P ij = X ij ∑ i = 1 m X ij

Where m represents the total number of prefecture-level and above cities (here, m = 18), and the calculation constant k is given by: (4) k = 1 ln ( m )

The Information entropy of the j-th indicator can be defined as follows: (5) e j = − k ∑ i = 1 m P ij ln ( P ij )

Calculate the divergence coefficient gj for the j-th indicator: (6) g j = 1 − e j

Calculate the weight of the j-th indicator: (7) w j = g j ∑ j = 1 n g j

This study calculates the disaster hazard index by combining standardized indicator values with their respective weights, reflecting both the hazard intensity levels and spatial distribution patterns across the study areas. The computational formula is expressed as follows (Equation 8): (8) H i = ∑ j = 1 n W j × X ij

Where Hᵢ denotes the disaster hazard index for the i-th city, while higher values indicate greater hazard intensity; Wⱼ represents the weight of the j-th indicator derived from the entropy method, Xᵢⱼ corresponds to the standardized value of the indicator.

To quantify regional vulnerability, the same methodology was employed to calculate the urban vulnerability index, thereby enabling quantitative analysis of both the magnitude and spatial distribution of vulnerability across the study areas (Equation 9). (9) V i = ∑ j = 1 n W j × X ij

Where Vᵢ denotes the urban vulnerability index for the i-th city, while higher values indicate greater vulnerability degree; Wⱼ represents the weight of the j-th indicator derived from the entropy method, Xᵢⱼ corresponds to the standardized value of the indicator.

Risk analysis should concurrently consider both infectious disease hazard and urban vulnerability, as risk is a function of hazard and vulnerability. The computational formula can be expressed as (48, 49) (Equation 10). (10) R i = H i × V i

The above calculation demonstrates that regional disaster risk escalates with increasing hazard intensity and vulnerability levels.

The Coupling Coordination Degree Model (CCDM), based on coupling theory, effectively evaluates interaction effects and coordinated development levels between different systems. It has been widely applied to examine relationships among social, economic, and ecological systems (50, 51), including: production-living-ecological spaces (52, 53), economy-ecology interplay (54, 55), Urbanization-ecological environment dynamics (56), Cultural landscape conservation vs. socioeconomic development (57). Recently, CCDM has transitioned from social-economic-ecological studies to disaster risk research, enabling in-depth analyses of: spatial coupling relationship between multidimensional poverty and the risk of geological disaster (58), the coupling relationship between flood risk and population vulnerability (59), integrated effects and multidimensional impacts of “Hazard-Exposure-Vulnerability” on urban flood risks (60). These studies demonstrate applicability of CCDM in disaster risk assessment frameworks. However, existing research lacks spatial coupling perspectives to unravel interaction mechanisms between acute public health hazards and urban vulnerability.

“Coupling” refers to the process of interaction and mutual influence between two or more elements (61). This study employs CCDM to analysis the interdependent or mutually constraining relationships between disaster hazards and urban vulnerability. Within CCDM research, most scholars adopt the conventional model structure, calculated as follows (Equations 11–13): (11) C = 2 × H i × V i ( H i + V i ) 2 (12) T = α H i + β V i (13) D = C × T

Given the dimensional differences between disaster hazards and urban vulnerability, normalized ordinal values were employed to calculate their synchronization and overall coordination degree (Equations 14–16). Based on the final coupling coordination degree (D) values, and referencing the classification framework from Xiang et al.’s study (58), the coordinated development status between disaster hazards and urban vulnerability was categorized into 4 major classes. These were further subdivided into 6 subtypes according to the proportional relationship between the two systems (Table 4). (14) C = f ( x ) k × g ( x ) k ( αf ( x ) + βg ( x ) ) 2 k (15) T = αf ( x ) × βg ( x ) (16) D = C × T

Where C is the coupling degree, T is the coordination index between disaster hazards and urban vulnerability, D is the coupling coordination degree. f(x) is the normalised value of disaster hazards ranking; g(x) is the normalised value of urban vulnerability ranking. k is an adjustment coefficient (typically 2 ≤ k ≤ 5). To enhance discriminative capacity, this study sets k = 3 following Su et al. (62). Considering that f(x) is as important as g(x) (i.e., α + β = 1, with α = β = 0.5). The higher the value of D is, the better the coordination degree between disaster hazards and urban vulnerability is.

The standardized indicators were objectively weighted using the entropy method, yielding the respective indicator weights for infectious disease hazards and urban vulnerability (Figures 2, 3) as well as dimensional indices (Table 3). This enabled quantitative measurement of infectious disease hazards, urban vulnerability, comprehensive risk, and coupling coordination degree, with regional distribution patterns visualized through spatial mapping techniques. Furthermore, the study conducted qualitative analysis by incorporating regional development characteristics specific to Sichuan Province.

Based on the aforementioned assessment results of infectious disease hazards and urban vulnerability, we calculated the comprehensive risk index. Sichuan Province generally exhibits a low overall risk level. The provincial capital Chengdu showed the highest risk index (0.291), while Guangyuan and Panzhihua demonstrated the lowest values.

Using the Natural Breaks classification method in ArcGIS (63, 64), the study area was divided into five risk levels: low, relatively low, moderate, relatively high, and high risk zones, as illustrated in Figure 8. Spatially, the risk distribution across Sichuan follows a “high in central regions, low in peripheral areas” pattern. The Chengdu Plain and southern Sichuan regions show significantly higher risk levels compared to Panxi and northeastern Sichuan areas.

The Chengdu Plain and southern Sichuan concentrate most of the province’s large and medium-sized cities. While these urban areas benefit from advanced economic development and well-established infrastructure, they simultaneously face challenges of high population density and strong mobility, leading to elevated infectious disease transmission risks. Notably, Chengdu presents a characteristic “low vulnerability-high risk” profile. In contrast, Panxi and northeastern Sichuan primarily consist of smaller cities or remote areas (e.g., Panzhihua, Guangyuan, and Ya’an), featuring lower population density, reduced human activity intensity, and higher natural environmental carrying capacity, resulting in comparatively lower comprehensive risk indices.

We conducted correlation analysis between the comprehensive risk index and actual cumulative infection cases to evaluate whether this risk indicator system could effectively reflect infection patterns (65). Spearman’s rank correlation analysis (Table 6) revealed a statistically significant positive correlation between the two variables (rs = 0.680, p = 0.002) at the 0.01 significance level. These results demonstrate that: increased comprehensive risk index significantly correlates with higher cumulative infection numbers; and the geographical distribution characteristics of the risk index effectively mirror the spatial distribution patterns of actual case numbers.

This study calculated the Coupling Coordination Degree (CCD) index to examine the interaction and coordination between infectious disease hazards and urban vulnerability. The overall CCD level across the study area was relatively low (mean = 0.384). Guangyuan and Panzhihua showed the highest CCD values (0.655 and 0.649, respectively), while Chengdu had the lowest CCD (0.031). As a megacity, Chengdu exhibited extremely low urban vulnerability but exceptionally high infectious disease risk, resulting in severe coupling coordination imbalance.

Based on the CCD model established in Section 2.7, we classified the study area into four coordination types using the following thresholds: coordinated development (0.6 ≤ D ≤ 1), marginal coordination (0.5 ≤ D < 0.6), near imbalance (0.4 ≤ D < 0.5), and declined imbalance (0 ≤ D < 0.4). Figure 9 reveals a spatial pattern where “lower CCD corresponds to higher risk.” Coordinated development cities were spatially dispersed, whereas declined imbalance cities were predominantly concentrated in the Chengdu Plain and southern Sichuan regions (Figure 10).

In-depth analysis reveals that the “hazard-dominant lagging” pattern predominates in the coupling coordination types across Sichuan Province. Cities achieving coordinated or marginally coordinated development, such as Panzhihua, Zigong, and Guangyuan, have made preliminary progress in aligning risk governance with spatial planning. Within the disordered and declined coordination categories, the regional risk patterns exhibit significant spatial heterogeneity: Chengdu demonstrates a characteristic “imbalanced-vulnerability lagging” type, while other cities predominantly follow an “imbalanced-hazard lagging” pattern. Empirical evidence confirms a notable positive feedback effect between urban vulnerability levels and infectious disease hazards, where heightened vulnerability not only exacerbates epidemic transmission risks by weakening systemic resilience but also perpetuates coupling system imbalances.

The aforementioned research reveals that the core mechanism influencing coupled coordinated development stems from multidimensional interactions among economic-spatial-social-environmental systems. This manifests in two distinct patterns: In rapidly developing regions (e.g., Chengdu, Mianyang), economic growth leads to excessive factor concentration, forming a “high investment-high density-high risk” transmission chain; whereas in underdeveloped areas (e.g., Guang’an), insufficient development momentum creates a vicious cycle of “low output-low protection-high vulnerability.” These findings align with the “economic foundation-spatial structure-governance capacity” synergistic mechanism proposed in Shekhar et al. (37), unveiling the coupled pathways of regional system complexity:

First, the economic-spatial coupling in the Chengdu Plain region demonstrates a significant positive feedback effect. Taking Chengdu and Mianyang as examples, the agglomeration of high-tech industries has driven population concentration and mobility (66), while simultaneously triggering land-use pattern restructuring. However, excessive intensive development has led to a decline in per capita public service resources (34). This “high-density, high-pressure” coupling model resulted in infection risks in core urban areas far exceeding surrounding regions during the pandemic, confirming the risk accumulation effects brought about by economically-driven spatial restructuring.

Second, the economic-social coupling in northeastern Sichuan manifests as a bidirectional inhibitory effect. A representative case was the “May 09” outbreak in Guang’an, where monthly infections surged to 1,299 cases. This episode revealed that the region’s per capita medical expenditure lagged behind the provincial average, with medical facility shortages directly accelerating epidemic spread. The event not only demonstrated how infrastructure vulnerability amplifies risk transmission, but more importantly exposed the underlying mechanism: sluggish economic development severely constrains public service investment (67), while inadequate social protection simultaneously restricts human capital development and suppresses economic growth by weakening consumption capacity.

Third, the economic-environmental coupling demonstrates distinct stress effects in industrial cities, particularly evident in Nanchong, Yibin, and Meishan. The expansion of traditional industries (e.g., liquor manufacturing in Yibin, textile and chemical production in Nanchong) has driven increased PM2.5 concentrations and water quality deterioration. Insufficient environmental protection investment further exacerbates health risks, empirically validating how the tension between development intensity and ecological carrying capacity translates into public health threats.

Fourth, the coupling coordination between spatial and economic systems is particularly prominent in Panzhihua. Through measures including intensive mining area redevelopment, functional diversification in central urban zones, and strict ecological conservation management, the city has achieved optimal alignment between spatial resource allocation and economic development needs. This spatial optimization strategy has established Panzhihua as a core city in China’s national comprehensive resource utilization pilot zone. Such a coordinated development pathway provides replicable practical solutions for peer cities to construct more resilient industry-space-ecological systems.

Fifth, the synergistic effects of space-society-environment systems are prominently demonstrated in Guangyuan City. As Sichuan Province’s highest CCD-scoring city (CCD = 0.655), Guangyuan has achieved virtuous interaction between social systems and environmental governance through its “low-density, high-investment” development model. Leveraging its mountainous topography to create a polycentric spatial configuration, the city maintains relatively low population density while compensating for economic development limitations through extraordinary healthcare resource allocation and robust social security systems. This governance approach—integrating decentralized spatial structures, premium ecological endowments, and targeted social investments—has not only significantly reduced epidemic transmission risks but also established a distinctive resilient development paradigm for mountainous cities. It provides valuable reference for coordinating socioeconomic development with ecological conservation in underdeveloped western regions.

These coupling mechanisms not only deconstruct the generative logic of risk heterogeneity, but also empirically validate through typical cases a transmission chain of: industrial agglomeration (economic) → land-use compactness (spatial) → service coverage (social) → ecological sensitivity (environmental). This provides a differentiated theoretical framework for regional resilience planning.

The coordinated development between infectious disease hazards and urban vulnerability constitutes a critical component of urban public health risk governance (68). These two systems interact through factor flows and feedback mechanisms, collectively determining regional comprehensive risk levels. This study has measured infectious disease hazards, urban vulnerability, comprehensive risks, and coupling coordination, with regional distribution patterns visualized through spatial mapping techniques.

This study developed a “hazard-vulnerability” risk coupling model based on the entropy method and Coupling Coordination Degree (CCD) model, systematically analyzing coupling coordination mechanisms and key influencing factors. The results reveal that the regional risk coupling system exhibits a distinct “high in central areas, low in peripheral regions” spatial differentiation pattern, where economic vulnerability, spatial vulnerability, and social vulnerability demonstrate complex nonlinear interactions.

To verify the reliability of research conclusions, this study conducted systematic validation through a triple-verification approach: First, correlation analysis between cumulative confirmed cases and the comprehensive risk assessment system across prefecture-level cities in Sichuan Province confirmed the predictive validity of the evaluation framework. Second, cross-validation between the risk model and coupling model revealed a significant negative correlation, demonstrating that improved coupling coordination degree effectively reduces systemic risks. Finally, verification analysis between regional case growth increments and coupling coordination degree further ensured the robustness of research findings. These validation results provide a solid empirical foundation for subsequent policy recommendations.

Building upon these research findings, this study proposes differentiated governance strategies:

1) Optimization strategies for the Chengdu Plain region

Given Chengdu’s distinctive “low vulnerability-high risk” profile, systemic optimization strategies should be implemented. First, within the Chengdu-Chongqing Economic Circle framework, priority should be given to developing regional growth poles such as Mianyang Science City and the Yibin-Luzhou industrial corridor. This can be achieved through industrial policy guidance and infrastructure interconnectivity to promote polycentric network development, thereby effectively alleviating single-core agglomeration pressures.

Second, urban spatial expansion should incorporate enhanced ecological resilience planning. This includes reserving ecological buffer zones along the outer ring expressway and connecting them through greenway systems to form emergency evacuation networks. Simultaneously, new urban developments should mandatorily reserve convertible emergency land parcels with pre-installed medical equipment interfaces.

Furthermore, structural adjustments to medical resources should be implemented via a “branch hospital system + tiered diagnosis and treatment” model.

These measures will ultimately establish a multidimensional prevention system encompassing “spatial decentralization—ecological buffering—emergency preparedness—medical resource balancing,” systematically mitigating public health risks in high-density urban areas.

2) Balanced development pathway for northeastern sichuan region

To address the concentration of impoverished populations and inadequate infrastructure in northeastern Sichuan, a systematic development strategy should be adopted.

First, priority should be given to cultivating competitive local industries such as specialty agriculture and green processing manufacturing, which can be achieved by establishing return-to-hometown entrepreneurship subsidies and industrial development funds to reverse outmigration trends and boost local employment.

Second, the returning population should be leveraged to drive local fiscal revenue growth, with newly increased public finances being preferentially allocated to upgrading medical facilities and improving transportation infrastructure to enhance public service delivery.

Ultimately, this integrated approach will establish a virtuous development cycle of “industrial revitalization → population agglomeration → fiscal expansion → service improvement,” fundamentally strengthening regional resilience. This cyclical model not only addresses current infrastructure deficiencies but also achieves sustainable development through endogenous growth drivers.

3) Transformation strategies for environmentally stressed cities

For environmentally stressed cities like Yibin and Nanchong, a development path coordinating industrial transformation with environmental governance should be adopted.

First, ecological transformation of key industries should be implemented to reduce pollution intensity through industrial structure optimization.

Second, environmental protection investments should be significantly increased through creating ecological compensation mechanisms, designating health protection buffer zones around major industrial parks, and building a tiered environmental risk prevention system.

Additionally, supporting environmental health monitoring networks should be constructed to enable real-time warnings of pollution sources and health risks.

Through this triple-intervention approach of “industrial upgrading → environmental governance → health protection,” environmentally high-risk cities can transition toward green development models.

4) Optimization and upgrading directions for Panzhihua

As a successfully transitioned resource-based city, Panzhihua has demonstrated outstanding performance in industrial-spatial-ecological coordination, yet requires targeted enhancements in healthcare and social security systems.

First, healthcare system improvements should be prioritized by increasing medical expenditure ratios and optimizing resource allocation.

Second, the current emergency supplies reserves are insufficient, necessitating the establishment of a three-tier allocation system encompassing municipal, county, and mining district levels.

Third, social insurance coverage expansion should be accelerated, particularly for vulnerable groups like miners, through customized insurance schemes and awareness campaigns.

Through this integrated approach of “medical investment → emergency preparedness → insurance coverage,” the city will systematically address existing gaps while reinforcing its sustainable development advantages.

5) Sustainable development pathway for Guangyuan City

As a exemplary model of resilient mountain city development, Guangyuan has achieved remarkable success in spatial-social-environmental coordination, yet requires focused improvements in economic development levels.

First, economic scale expansion should be prioritized by addressing the below-average per capita GDP through targeted growth initiatives.

Second, industrial support strengthening should be implemented by cultivating regionally influential leading industries to address the current lack of driving industries.

Third, fiscal capacity building should be enhanced by improving financial self-sufficiency rates and accelerating infrastructure development to overcome current limitations.

Through this integrated approach of “specialty industry cultivation → fiscal mechanism innovation → infrastructure improvement,” the city will effectively enhance endogenous growth drivers and sustainable development capacity.