As noted earlier, VHR is a key factor in ensuring animal health and supporting the development of the livestock industry (18). Differences in the quantity, structure, skill, and incentives of VHR directly affect the accessibility and professionalism of veterinary services at the local level (17). Based on this, the study builds a province-level evaluation index of VHR to measure regional differences in human resource allocation and service capacity.

Considering both data availability and comparability across provinces, the VHR subsystem is designed around four dimensions: workforce size, technical structure, work efficiency, and incentive mechanisms (see Table 1). These indicators aim to capture how VHR supports livestock development and rural veterinary services across different regions.

Specifically, the workforce size dimension is measured by the number of staff in TAHVS, reflecting the overall supply capacity of local human resources in each region (43). The technical structure dimension is represented by the proportion of staff with senior professional titles, indicating the technical composition and qualification level of local veterinary teams (44). The work efficiency dimension is assessed by the annual average output per staff, which reflects the economic contribution per person and the service performance of the workforce (45). The incentive mechanisms dimension is measured by the annual average salary of staff, representing the strength of institutional support and the attractiveness of the position (46).

Unlike existing studies that focus mainly on static workforce size or educational background, this study incorporates efficiency and incentive dimensions to enable a more comprehensive assessment of VHR. The index covers four aspects: quantity, structure, efficiency, and incentives. It gives a clear picture of the overall strength of local veterinary teams in different regions and provides a solid base for the next-step coupling analysis with PAAW.

As discussed above, PAAW has become an important social force driving improvements in local veterinary service systems (26). Public attention to animal diseases, prevention behaviors, and institutional services not only reflects responsiveness at the level of information behavior but also reveals potential preferences for services and a willingness to participate in governance (27). Therefore, building an index that can quantify and describe public attention across multiple dimensions helps identify how well supply and demand are aligned and where structural mismatches may exist.

This study draws on the three-dimensional animal welfare framework proposed by the World Organisation for Animal Health (WOAH), which includes physical health, emotional and mental state, and expression of natural behaviors (47). Based on this, the PAAW subsystem is developed using four dimensions: notifiable animal diseases, common diseases, preventive measures, and the service system. The subsystem is designed to reflect both the rural context and the characteristics of online behavior (see Table 2).

In selecting the indicators, priority was given to data availability on the Baidu Index platform, relevance to rural information-seeking behavior, and the representativeness of each keyword within animal welfare dimensions. For the notifiable animal diseases dimension, “avian influenza” and “foot-and-mouth disease” were selected to reflect public risk perception of highly pathogenic animal diseases and concern for the basic welfare condition of “freedom from disease” (32). The common diseases dimension includes “pig diseases” and “chicken diseases,” which are closely tied to the everyday language of rural livestock keepers and signal continued public attention to frequent health issues in daily farming practices (48). For preventive measures, the keywords “animal quarantine” and “animal epidemic prevention” correspond to routine tasks performed by local veterinary workers and reflect public awareness of basic disease prevention and service efforts (49). Finally, in the service system dimension, “animal husbandry and veterinary” and “livestock station” were used to gauge how the public perceives the capacity, accessibility, and structure of grassroots veterinary services (50).

All eight keywords selected in this study are directly related to the “physical health” dimension of animal welfare, covering animal disease prevention and control, treatment of common diseases, vaccination behaviors, and veterinary service supply. These represent the fundamental baseline of animal welfare. The keywords “animal quarantine” and “animal epidemic prevention” also reflect standardized operational procedures, which may influence animals’ emotional states and are therefore indirectly linked to the “emotional and mental state” dimension. Meanwhile, “livestock station,” as a representative of local service platforms, supports the “expression of natural behaviors” through its role in feeding guidance and environmental management.

Overall, the PAAW subsystem establishes a measurable, structured, and rural-adaptive indicator framework, providing both data and conceptual foundations for the subsequent coupling coordination analysis with the VHR subsystem.

The coupling degree is widely used to reveal complex coupling mechanisms among systems (58). The formula is as follows:

Here, C represents the coupling degree between the systems. U_V and U_A denote the comprehensive contribution values of VHR and PAAW to the combined system, respectively.

The level of coupling alone does not reflect the development status of the two subsystems. A high coupling degree may still occur even when both VHR and PAAW are at low levels, which differs in meaning from high-level coordination. Therefore, a coupling coordination degree (CCD) model is introduced to evaluate how well the two subsystems develop in a balanced and coordinated manner. The CCD is calculated as follows:

Here, D represents the CCD, and T is the comprehensive coordination index, reflecting the joint contribution of VHR and PAAW to overall coordination. α and β are coefficients to be determined. Since VHR and PAAW are complementary and hold equal weight in their interaction, both α and β are set to 0.5.

Following previous studies (59, 60), this study adopts a ten-level classification to interpret CCD values, ranging from extremely uncoordinated to perfectly coordinated. The classification is shown in Table 3. This approach allows for nuanced identification of coordination intensity across regions and time.

To explore the internal constraints affecting the coordinated development of VHR and PAAW, this study introduces the relative development degree (RDD), denoted as R, which is defined as the ratio between VHR and PAAW.

Based on the RDD values of the two subsystems and drawing on previous studies (61, 62), the coupling development relationship between VHR and PAAW can be classified into three types: VHR lag (R < 0.9), VHR-PAAW synchronization (0.9 ≤ R ≤ 1.1), and PAAW lag (R > 1.1).

As shown in Figure 3, there are clear regional differences in the development of VHR across provinces in China from 2014 to 2023. Some provinces—particularly in the southwest—demonstrate strong and consistent growth. For example, several provinces in this region reached their highest VHR levels in 2023, indicating robust workforce expansion in local veterinary institutions. In contrast, parts of northeastern and northern China experienced earlier peaks around 2015, followed by gradual declines or stagnation, suggesting weakening or plateauing capacity in workforce renewal. VHR levels in central and eastern provinces remain relatively low throughout the study period. Some of these areas reached their lowest points around 2020, likely due to shifting policy priorities or limited investment in grassroots animal health services. Meanwhile, certain western and northwestern provinces have shown a steady upward trajectory since 2018, reflecting progress in targeted rural veterinary workforce development.

To provide a more detailed view, Figure 4 presents VHR scores at the provincial level. Provinces such as Yunnan and Sichuan consistently perform well, with Yunnan ranking first nationwide in several years and Sichuan showing steady growth since 2017. This reflects effective talent attraction mechanisms and sustained local investment. In contrast, economically developed eastern provinces such as Shanghai, Zhejiang, and Guangdong have relatively low VHR levels. This is not due to weak economic strength but rather to their industrial structure—animal husbandry accounts for a smaller share of their economy, which limits the actual demand for veterinary talent.

Meanwhile, regions like Ningxia, Guizhou, and Chongqing in central and western China have seen steady improvements, suggesting early progress in human resource development. However, provinces such as Hainan, Qinghai, Gansu, and Anhui continue to lag, with little sign of significant improvement. The northeastern provinces show considerable fluctuations and have not yet achieved stable development.

Overall, VHR development across the country follows a regional pattern: the Southwest leads, the East lags, the Northeast remains unstable, and the Central and Western regions are catching up. This reflects a structural imbalance in VHR development and highlights the urgent need for region-specific strategies to promote coordinated regional growth.

Based on Figure 5, the development of PAAW reveals significant disparities across provinces from 2014 to 2023. The East region consistently recorded the highest level of public attention, peaking in 2016, indicating a high level of public awareness and sensitivity to animal welfare issues. The Northeast region also showed relatively high levels, with a trend like the East region, though with slightly greater fluctuations. The North and South Central regions maintained moderate levels, with a slight decline in recent years, possibly due to changes in industrial structure and information dissemination frequency. The Southwest region showed lower levels of public attention with noticeable fluctuations, while the Northwest region remained at the lowest level over time, reflecting a serious lack of public awareness.

To provide a more detailed spatial comparison, Figure 6 presents PAAW levels for each province and year. The overall distribution further confirms that eastern and southeastern provinces tend to have higher PAAW levels, while western and inland regions remain comparatively lower. Guangdong, Shandong, and Jiangsu consistently ranked among the top provinces, with Guangdong leading nationwide for several years. This reflects stronger public sensitivity to animal welfare issues and better access to information in economically developed areas. Attention levels were also relatively high in provinces such as Sichuan, Zhejiang, and Henan, suggesting that animal welfare awareness is gradually expanding in some central regions. In contrast, Tibet, Qinghai, Ningxia, and Hainan stayed at the low end throughout the period, showing continued gaps in information access, education, and local systems.

In general, the spatial differences in PAAW across regions come from many factors, including the level of economic development, access to information, public awareness, the quality of animal health service systems, and the distribution of the livestock industry. Future efforts should match local conditions and improve policies and outreach to help raise PAAW and support balanced development.

This study first checks whether the VHR and PAAW scores are stable over time. ADF tests for each province are used to simulate the IPS method (see Table 4). The results show that 80.6% of the provinces pass the test for VHR at the 5% level, meaning the data are stable. For PAAW, 61.3% of the provinces pass the same test. Based on the rule that most provinces must pass for the data to count as stable, both VHR and PAAW are considered stable overall. So, they are suitable for the PVAR model in the next step.

This study uses AIC, BIC, and HQIC to decide the number of lags. All three suggest using two lags. Based on this, a PVAR model with two lags is built. The Granger causality test and impulse response analysis are both carried out under this setting to check how stable and useful the model is. Table 5 shows the Granger test results for whether VHR and PAAW have a two-way causal link.

The results show that at the 5% level, VHR has a clear Granger causal effect on PAAW. This means that investing in VHR can help raise PAAW over time. On the other hand, the effect of PAAW on VHR is not clear. This suggests that public awareness alone is not strong enough to change the structure of veterinary services. These findings support the next step of impulse response analysis and provide a basis for building the coupling coordination model. They also show the direction in which VHR affects PAAW.

After confirming the Granger causality between VHR and PAAW, this study uses the impulse response function (irf) to explore how the two subsystems affect each other over time. Figure 7 shows the results of their two-way responses.

Findings indicate that a positive shock to VHR leads to a clear positive response in PAAW starting in the first period. The effect peaks in the second period, then gradually weakens and levels off after the fifth period. This suggests that VHR has a strong and sustained influence on PAAW in the medium term.

In contrast, the response of VHR to a shock from PAAW appears weak and volatile, with no consistent direction or pattern. While this supports the short-term unidirectional relationship identified by the Granger test, it does not necessarily rule out longer-term or indirect feedback effects. Institutional inertia or policy delays may obscure potential responses to public attention, which could become more apparent over longer time horizons. Further research could explore whether public attention exerts delayed or mediated influence on resource allocation through other channels.

In short, the IRF analysis confirms that VHR plays a leading and important role in raising PAAW. These results also provide evidence for the next step—analyzing the coordination between VHR and PAAW.

Based on Equations 8–10, the CCD between VHR and PAAW is calculated, reflecting the extent to which public demand aligns with veterinary service capacity. As shown in Figure 8, for visualization purposes, CCD values are displayed by grouping provinces into six traditional geographic regions. While this provides a general sense of spatial distribution, all analytical interpretations are conducted at the provincial level.

From 2014 to 2023, CCD trends varied across provinces. The South Central region provinces remained in the barely coordinated state throughout the period, indicating limited but stable interaction between the two systems. The Northeast, North, and East regions reached relatively high levels around 2015 but then declined. The drop in the East was especially sharp, with some years falling to the lower end of the coordinated range. The Southwest region remained stable, with small fluctuations, and consistently stayed in the barely coordinated range. The Northwest region remained at a low level throughout the period. In some years, it fell into the nearly or slightly uncoordinated state, reflecting weak interaction between the two systems.

A more detailed analysis at the provincial scale is presented in Figure 9, which shows CCD scores for selected years (2014, 2017, 2020, and 2023). The results indicate that most provinces remained in the “barely coordinated” to “moderately coordinated” range (between 0.5 and 0.8) throughout the observation period.

Provinces like Jiangsu, Shandong, Sichuan, and Yunnan maintained high CCD levels, mostly in the “Primarily” to “Moderately coordinated” range (between 0.6 and 0.8), with some years approaching “well coordinated” status. Beijing, Zhejiang, and Guangdong—despite being economically developed—showed relatively stable CCD. Though there were small fluctuations, they generally remained within the “barely” to “primarily coordinated” range (between 0.5 and 0.7).

In contrast, less developed western provinces such as Tibet, Qinghai, Ningxia, and Gansu remained mostly in the “Uncoordinated state” (below 0.5), indicating much weaker CCD performance between VHR and PAAW.

In summary, the CCD distribution across provinces reveals a clear spatial pattern. Provinces with stronger economic bases and more established veterinary systems tend to achieve better coordination between supply and demand. Conversely, provinces with limited investment in veterinary infrastructure or public awareness show persistently low coordination levels.

These findings underscore the need for province-specific strategies to strengthen the alignment between service capacity and public expectations. Improving CCD requires simultaneous attention to workforce development, digital outreach, and governance consistency—especially in provinces with historically weak performance. Tailored interventions in these areas will help promote a more balanced and integrated animal health governance system across China.

The relative development degree (RDD) is calculated using Equation 11. Based on the RDD classification, the coupling relationship between VHR and PAAW is divided into three types: VHR lag, PAAW lag, and VHR-PAAW synchronization (see Figure 10).

In terms of spatial and temporal patterns, the “VHR lag” type is the most common. It is widely found in South Central and East China, such as Henan, Hunan, Jiangxi, Anhui, and Fujian, as well as Sichuan in Southwest China. These provinces are characterized by intensive livestock production and rising PAAW. The persistent lag in VHR suggests that local veterinary services have not kept pace with growing public demand, especially in areas such as disease prevention, diagnosis, and health management.

The “PAAW lag” type mainly appears in the Northeast and Northwest regions, such as Heilongjiang, Inner Mongolia, and Gansu. Although these places have relatively strong veterinary capacity, low PAAW levels suggest challenges in information access, farming practices, or public awareness. The “VHR-PAAW synchronization” type appears only in a few provinces and specific years, such as Fujian, Liaoning, and Shaanxi. These cases often result from short-term factors like policy shifts, disease outbreaks, or media coverage. They tend to be temporary, which suggests that the coordination between supply and demand systems is not yet stable or sustained.

In summary, the overall relationship between VHR and PAAW shows a supply–demand mismatch. This reflects ongoing gaps between institutional support, resource allocation, service capacity, and public awareness. To achieve coordinated development, it is necessary to strengthen local veterinary service systems while also improving public understanding of animal welfare.

In terms of VHR, the Southwest region—especially Yunnan and Sichuan—shows strong capacity in resource allocation and continued investment. These areas have made clear progress in attracting veterinary talent and building local service systems. In contrast, eastern developed provinces like Shanghai, Zhejiang, and Guangdong have relatively low VHR levels. This is mainly due to their industrial structure, which focuses on services and high-end manufacturing, with limited demand for veterinary services because of the smaller livestock sector. These findings reveal structural differences in China’s animal health system across regions (63).

Regarding PAAW, public attention remains high in coastal and northeastern regions. Provinces such as Guangdong, Shandong, and Jiangsu consistently rank at the top in Baidu search data. In contrast, PAAW levels are much lower in Central and West regions, especially in pastoral and less developed provinces. This shows clear regional disparities, shaped by differences in information access, education, institutional support, and economic development (29, 34, 64, 65).

The Granger causality test shows that VHR has a significant impact on changes in PAAW, while the reverse path from PAAW to VHR is not significant. This supports a supply-driven mechanism, in which stronger veterinary capacity contributes to increased public awareness of animal welfare (37). It also reflects the role of institutional capability in shaping public responsiveness (66).

The impulse response analysis reinforces this finding. When VHR receives a positive shock, PAAW responds quickly, reaches a peak, and then stabilizes. This shows that improvements in VHR can have a sustained effect on public engagement in the medium term. It supports earlier findings that information and resource supply can influence public digital behavior in its early stages (67).

However, the response from VHR to changes in PAAW is weak and unstable. This does not necessarily mean that public attention has no influence. Rather, such influence may take longer to manifest, possibly mediated through policy feedback mechanisms, budget cycles, or media-driven political agendas (68, 69). Therefore, while current evidence highlights the leading role of VHR, future research should consider longitudinal approaches or qualitative methods (e.g., policy tracing, stakeholder interviews) to examine whether public demand gradually feeds back into institutional adjustments over time.

In terms of regional patterns, provinces like Jiangsu, Shandong, and Yunnan show high levels of CCD. This suggests a good match between public resource allocation and public demand. In contrast, regions in the west, such as Tibet, Qinghai, and Ningxia, remain close to or within the range of imbalance. These areas show weak system interaction and poor resource integration. Such regional differences point to an uneven coupling mechanism between service capacity and public demand across provinces (33).

The analysis of RDD shows that most provinces fall into the “VHR lag” group, where PAAW grows faster than local veterinary service capacity. This reflects a common pattern in developing countries, where investment in human resources often falls behind rising public awareness of animal health (13). The “PAAW lag” type is mostly found in the Northeast and Northwest, where basic veterinary services are in place, but public attention remains low. This may result from limited access to information, livestock structure, and cultural acceptance, suggesting that digital access and social norms shape public awareness (70). The “VHR-PAAW synchronization” type appears only in a few provinces and usually in short periods. These cases are often driven by sudden policy actions or media events, leading to short-term changes in public attention (29), and a stable coordination mechanism has yet to form.

The regional disparities in CCD and RDD patterns can be explained by differences in institutional environments, governance models, and communication infrastructures. For instance, Yunnan exhibits relatively high CCD levels despite being a less economically developed province. This may be attributed to strong grassroots demand for animal health services in its livestock-intensive areas and provincial efforts to expand rural veterinary access (71). In contrast, Guangdong, despite its economic strength and developed infrastructure (72), shows only moderate CCD levels, suggesting that public attention and veterinary resource allocation may not be well aligned or equally prioritized across sectors.

In the Northeast and Northwest regions, where PAAW lag is common, the presence of veterinary systems has not translated into high public engagement. Factors such as limited information access, traditional farming practices, and cultural attitudes toward animal welfare may have contributed to weak public awareness (73). Meanwhile, provinces like Tibet and Qinghai remain in a persistently uncoordinated state, which may be associated with weaker institutional coordination and limited investment in rural service infrastructure. These findings suggest that beyond economic development, policy alignment and targeted public outreach play a critical role in shaping VHR-PAAW coordination.

Overall, VHR plays a leading role in driving changes in PAAW. However, structural imbalances and coordination gaps remain between regions. Unlike traditional approaches that treat supply and public behavior as separate issues, this study builds a joint framework linking VHR and PAAW. It reveals the dynamic interaction among institutions, resources, and public behavior. This offers a new way to understand structural mismatches in animal health systems and helps guide policy design.