Strategies to inTerrupt RAbies Transmission for the Elimination Goal by 2030 In China (STRATEGIC): a modelling study

As a nearly always fatal but entirely vaccine-preventable viral disease, dog-transmitted human rabies remains a critical but neglected threat to global health in some countries, especially for children [1]. For example, given the vast population size, besides India, China reported the second-largest number of global rabies deaths [2], and rabies is still the second most common infectious cause of death for Chinese children, despite an annual decrease of 20% over the past decade [3]. In many countries, rabies elimination was successfully achieved through vaccination for both human and domestic dogs by the "One Health" approach [4]. In this regard, the World Health Organization (WHO) and partners have set a goal that, by 2030, up to 92% of countries in the world could eliminate dog-mediated human rabies deaths ("Zero-by-30"), and all nations would reduce dog-mediated human rabies deaths by 50% [5]. However, whether the global goal of "Zero-by-30" could be achieved in some countries, such as China, is still undetermined.

The WHO rabies modelling consortium has developed epidemiological and economic models to predict rabies deaths and investigate the effectiveness of enhancing post-exposure prophylaxis (PEP) access in 67 rabies-endemic countries during 2020–2035 [6]. This modelling study supported the "One Health" approach and suggested that mass dog vaccination programmes could eliminate dog-mediated rabies over this period, and improved PEP access was highly cost-effective, especially in combination with patient risk assessments by the integrated bite case management (IBCM). Unfortunately, as a populous and rabies-endemic country, China is underrepresented in this global modelling study.

Unlike other Western or Gavi-eligible nations [7], China has unique but complex health systems, with multiple levels of care and various health providers offering different services [8]. For example, PEP treatment for patients with potential rabies exposures is available at county-level centers for disease control and prevention, hospitals and township-based rabies clinics (health sector). China's current availability of PEP use seems sufficient, while patient risk assessments by the IBCM are not yet implemented [9]. In some provinces with high rabies burdens, such as Guizhou, Hunan and Guangxi, public insurance of the New Rural Cooperative Medical Scheme (NRCMS) covered PEP-related medical expenses and has contributed to the observed decline of human rabies cases in rural areas. However, most provinces in China have discontinued the reimbursement of PEP following the integration of the NRCMS and the Urban Resident Basic Medical Insurance in 2016. As a result, patients currently pay the whole PEP series as an out-of-pocket cost, which might hinder their possibility of adherence to the PEP treatment regimen [10].

On the other hand, dog vaccination coverage in China remains inadequate from the "One Health" approach perspective [10, 11]. As a nationally notifiable disease for both dogs and humans, dog-mediated rabies management involves multiple sectors in China [9], including the health sector responsible for human PEP and immunization programmes, the livestock/veterinary sector responsible for supplying veterinary rabies vaccine and implementing mass dog vaccination, and public security departments responsible for pet dog registration, vaccination, and stray dog management in cities. It is estimated that there are approximately 80–100 million dogs in China [9]. Before the revision and implementation of the Animal Epidemic Prevention Law of the People's Republic of China in 2021, there were no national, unified regulations for dog management and dog vaccination against rabies in China. As a result, dog management practices were not optimal, and the registration and vaccination rate was low, particularly in rural areas [12].

Taken togehter, evidence on the effectiveness and cost-effectiveness of comprehensive strategies of improved PEP access for humans, mass dog vaccination, and collaboration between the health and livestock/veterinary sectors through the use of IBCM in the context of the "One Health" approach is lacking in China. We used the updated data from China's national human rabies surveillance system and similar decision-analytic modelling methods as the WHO rabies modelling consortium in the Strategies to inTerrupt RAbies Transmission for the Elimination Goal by 2030 In China (STRATEGIC) study. Specific aims include: (1) to estimate whether the global rabies plan of "Zero-by-30" could be achieved in China under the status quo; (2) if not, to predict the time when the Chinese elimination goal will be accomplished; and (3) to find out the most cost-effective strategy for the purpose in China.

As described in the statistical analysis plan (Additional file 1: Statistical Analysis Plan) [6, 13‐15], we adopted a decision tree model to simulate human rabies dynamics in different strategies for rabies control, similar to the WHO rabies modelling consortium study for direct comparisons [6]. We used data collected by the Chinese Center for Disease Control and Prevention (China CDC) in the National Human Rabies Surveillance (NHRS) system and also in some provincial surveillance points, including Shandong (East China), Hunan (Central China), Tianjin (North China), Guangxi (South China), Shaanxi (Northwest China) and Guizhou (Southwest China), to investigate potential regional disparities in diverse areas (Additional file 1: Fig. S1). Our study followed the updated Consolidated Health Economic Evaluation Reporting Standards 2022 (CHEERS 2022) checklist [13] (Additional file 1: CHEERS 2022 Checklist). All analyses were performed in R (Version 4.0.5).

Cumulatively 8308 dog-mediated human rabies deaths in China were reported during 2011–2020, with declining trends annually (Additional file 1: Fig. S3). However, vast regional rabies-epidemic disparities were shown: most deaths (1071) were in Guangxi (South China), while the rabies cases in Tianjin (North China) were much fewer (36) (Additional file 1: Fig. S1). Totally 146.21 million vials of human-used PEP vaccines were signed in the same period of 2011–2020, and the annual amount in 2020 was 17.55 million, suggesting adequate access to PEP in China by far. Maintaining the current level of PEP access in China is expected to avert 19,270 (95%UI: 13,459–25,921) deaths or 304,106 (95%UI: 212,402–409,064) DALYs from 2024 to 2035, compared to the condition without PEP (Table 1). However, under the status quo, the predicted number of dog-mediated human rabies deaths will remain stable at 308 (95%UI: 214–411) in 2030 (Table 2 and Fig. 2), with a total of 3,695 (95%UI: 2,572–4,931) deaths or 58,317 (95%UI: 40,581–77,817) DALYs in China in the next decade.

If expanding PEP access in China (scenario 2), the annual deaths will decrease from 305 (95%UI: 212–407) in 2023 to 212 (95%UI: 147–284) in 2028 and then remain unchanged till 2035 (Table 2). Total deaths and DALYs will reduce to 2,737 (95%UI: 1,904–3,663) and 43,616 (95%UI: 30,336–58,367), respectively. Compared to the status quo, this strategy of improving PEP access alone could prevent approximately 958 (95%UI: 668–1,268) extra deaths during 2024–2035 (Table 1). In contrast, scaling up dog vaccination to the coverage of 70% could eliminate rabies deaths as soon as 2033 in China. With mass dog vaccination alone (scenario 3a), the cumulative number of deaths will reduce to 429 (95%UI: 95–1,244), thus averting approximately 3,265 (95%UI: 2,477–3,687) extra deaths in contrast to the status quo during 2024–2035. Under scenario 3a, by 2025, Chinese human rabies deaths will decrease to 116 (95%UI: 19–249), which is half of the counterpart in 2023 (i.e., achieving the global goal of a 50% reduction in human rabies mortality). Of note, even by combining improved PEP access and mass dog vaccination (scenario 3b), the earliest time to eliminate the Chinese rabies deaths will remain in 2033.

Concerning costs, a total of 304.57 (95%UI: 259.73–350.57) million human-used PEP vaccines will be required during 2024–2035 under the status quo, approximately 25.38 million vials annually, which already exceeds the signed vials in 2020 in China (Table 1). Furthermore, an additional 23.37 (95%UI: 19.93–26.91) million vaccine vials with extra costs of 323.60 (95%UI: 276.00–372.50) million US dollars will be needed to expand PEP access (scenario 2) from 2024 to 2035. The trend of rabies deaths under the scenarios that included IBCM (scenarios 4a to 4d) was similar to those without IBCM (scenarios 1, 2, 3a and 3b) (Additional file 1: Table S6 and Fig. S4). However, the use of IBCM could substantially reduce the PEP vaccine under the status quo from 304.57 (95%UI: 259.73–350.57) to 152.51 (95%UI: 130.07–175.54) million vials during 2024–2035, with the total cost decreasing from 9.38 (95%UI: 8.62–10.17) to 6.78 (95%UI: 6.39–7.17) billion US dollars.

From the cost-effectiveness perspective, the ICERs of the strategies of improving PEP access only (scenario 2) or mass dog vaccination alone (scenario 3a) are 337,771 or 962,920 US dollars per death prevented, respectively, compared with the status quo (Additional file 1: Table S7). Although the ICERs of both strategy 2 and 3a exceed the cost-effectiveness threshold of three times China's GDP per capita, the strategy of IBCM and mass dog vaccination (scenario 4c) dominates the status quo, with more deaths averted and less cost. In addition, compared with scenario 4c, the ICER of the combination strategy of enhancing PEP access, mass dog vaccination, and IBCM (scenario 4d) is also beyond the cost-effectiveness threshold. Thus, the former strategy of combining mass dog vaccination and the use of IBCM (scenario 4c) is the most cost-effective among all the elimination strategies (i.e., scenarios 3a, 4c and 4d).

In terms of regional disparities for rabies control, under the status quo, the trends of dog-mediated human rabies deaths in different areas of China are broadly similar to those at the national level. However, with improved PEP access (scenario 2), the deaths in Hunan (Central China) continue to decline, while the trends in Shandong (East China) remain almost unchanged. Scaling up dog vaccination could also facilitate the elimination goal in our six region-stratified analyses, showing that human rabies deaths in all those areas will be zero by 2029 (Table 2 and Fig. 2), which would be earlier than the goal of "Zero-by-30". Moreover, in area-specific cost-effectiveness analyses (Fig. 3 and Additional file 1: Table S8-S13), the ICERs of strategies are generally consistent with those at the national level that the one combining mass dog vaccination and use of IBCM (scenario 4c) would be the most cost-effective.

Sensitivity analyses showed that estimates of the rabies burden of the strategy of mass dog vaccination (scenario 3a) and the one combining mass dog vaccination with the use of IBCM (scenario 4c) might be most affected by uncertainty in the incidence of rabid dog bites (Additional file 1: Fig. S5). However, strategies with mass dog vaccination (i.e., scenarios 3a, 3b, 4c, and 4d) coordinated the reduction of rabies burden, and it seems less likely to change the magnitude of estimates of dog-mediated human rabies deaths and DALYs, with shorter uncertainty intervals in PSA of 1000 times of simulations (Additional file 1: Fig. S6).

According to the WHO recommendation [28], the average cost of mass dog vaccination is estimated to be 4.03 US dollars per dog, ranging from 1.56 to 11.33 US dollars. Given the budget constraints, as the cost of mass dog vaccination per dog rises to 13.69, 13.81, and 14.06 US dollars, the additional cost (compared to the status quo) of the strategy combining both mass dog vaccination and IBCM will reach the level equivalent to zero, one, and three times China's per capita GDP, respectively. It highlights the possibility that if the cost of mass dog vaccination per dog can be kept below 14 US dollars, the strategy incorporating both mass dog vaccination and IBCM may become cost-effective in China. Even if increasing the total costs of IBCM from 100 thousand to 652.10 million US dollars during 2024–2035 (i.e., 8.33 thousand to 54.34 million US dollars annually), the combined strategy of mass dog vaccination and use of IBCM (scenario 4c) will still dominate. When the annual cost of IBCM exceeds 62.84 million US dollars (about 6036 times China's GDP per capita in the year 2020), the ICER of this strategy (scenario 4c) will exceed three times GDP per capita, suggesting the robustness of our conclusions.

Our study indicated that, under the status quo, the burden of human rabies in China would remain considerable in the next decade. The annual rabies deaths have decreased in recent years with much effort by the Chinese government, such as the launch of the National Animal Rabies Prevention and Control Plan (2017–2020) and the revision of the Animal Epidemic Prevention Law (2021) [9]. Nevertheless, the trends of the rabies epidemic and the purpose of elimination still call for specific, actionable strategies in China. Globally, the WHO and partners are also focusing on the key strategic framework to guide the cross-sectoral collaborative activities in the forthcoming Joint Plan of Action for One Health (2022–2026) [30]. On this matter, our STRATEGIC study may fill the gaps in evidence on comprehensive strategies for rabies control in China and promote the achievement of the global plan of "Zero-by-30".

The WHO Rabies Modelling Consortium study suggested expanding PEP access to be highly cost-effective in 67 rabies-endemic countries [6]; however, its effectiveness is quite limited in China for the elimination goal of "Zero-by-30". Improving PEP access alone could only reduce less than half of deaths but is not enough to eventually end the dog-mediated rabies epidemic in China. Unlike those Gavi-eligible countries, China's current supplies and coverage of PEP vaccines seemed adequate 9, for example, about 94.64% of the outpatients from rabies PEP clinics completed the whole schedule of doses in Tianjin [21]. In contrast, the national level of dog vaccination coverage is as low as 40% currently in China [10]. What is worse, dog vaccination coverage was just 19.1% in rural Guangxi (South China) in 2021 [31]. Thus, there is still room for improvement in dog vaccination coverage rather than merely expanding human PEP access in light of the "One Health" approach [32, 33]

The current study showed that it is expected to achieve the elimination goal in China as soon as 2033 by implementing mass dog vaccination coverage of 70%, highlighting dog vaccination in the "One Health" approach for rabies control, which is consistent with findings from other countries [34, 35]. In India, for instance, if scaling up dog vaccination coverage to 7% or 13% in the next five years, rabies deaths could be reduced by 70% or 88%, respectively, suggesting a slight improvement in dog vaccination coverage yields substantial health benefits [36]. In this sense, it is practical to improve dog vaccination coverage gradually in some areas of South China (e.g., Guangxi), from the current 19.1% to the national average level of 40%, and further to the aim of 70%. Moreover, studies in the Philippines have indicated that dog population management programmes, such as dog spaying/neutering and shelters, can effectively manage limited resources and reduce human rabies cases by increasing dog vaccination coverage rates [37]. Further research is necessary for China to assess the impact of programmes for dog population management, specifically in rural areas and under the livestock/veterinary sector supervision.

The WHO proposed the role of IBCM in evaluating the risk of rabies exposure, which formally engages medical and veterinary multi-sectors from the "One Health" viewpoint; however, it has not been implemented in China at all [10]. The purpose of IBCM in other rabies-epidemic countries, such as Tanzania and Haiti, is to improve PEP compliance to reduce the rabies burden [38, 39]. For example, in Haiti, the probabilities of health-seeking, receiving PEP treatment, and completing the regimen were 54%, less than 50%, and less than 40%, respectively [38]. In contrast, the corresponding probabilities in China were 85%, 99%, and 91%, respectively [21, 31]. The WHO recommended that countries also avoid overuse of PEP, especially when successful control activities cause rabies to decline [28]. Since adequate PEP access has been provided in China [9], the primary aim of IBCM is expected to reduce the costs of unnecessary use of PEP; thus, the strategies combined with IBCM in our study dominate others in terms of cost-effectiveness. Nevertheless, the success of IBCM relies on close collaboration between the health and livestock/veterinary sectors to ensure timely dog bite reporting and a skilled workforce to assess animals and make diagnoses [40]. Cost estimates for IBCM should include factors such as the number of staff needed and their working hours [38]. Considering regional disparities in the rabies epidemic and dog vaccination coverage, a pilot study on implementing IBCM is needed in China.

To the best of our knowledge, this is the first study that used the updated national data to predict the future burden of human rabies and assess the cost-effectiveness of comprehensive rabies-control strategies in China. Moreover, we considered the regional disparities in the rabies epidemic and the economic level in China to inform the evidence-based practical implementation of these strategies for rabies elimination by areas. Nevertheless, as with other modelling studies, our study also has limitations that warrant consideration when interpreting the results. First, some of our assumptions in the model may vary over time; for example, the natural growth rate of the human population will change with time, while we assumed that the values of these parameters were constant for simplicity. In addition, we used similar methods and assumptions to the WHO Rabies Modelling Consortium study for international comparisons. Second, the parameters used in analyses were mainly from official data confirmed in the national rabies surveillance system or published results in other countries; for example, the estimation of the bitten incidence of rabid dogs was from Tanzania [6]. However, our intention was to highlight the cost-effective strategies that could make the elimination goal achievable, rather than just making predictions on the precise elimination time. In this regard, various sensitivity analyses allowed a comprehensive approach to the assessment of the robustness of our conclusions. Finally, with the same methods used in the WHO Rabies Modelling Consortium study, some region-specific parameters were obtained from the national-level values due to a lack of available data. Thus, these homogenous average estimates may inevitably result in underestimating regional disparities. Further investigations on region-specific surveys in more areas of China are therefore required.

In summary, the burden of Chinese rabies death remains stable in the next decade under the status quo; however, scaling up dog vaccination instead of expanding PEP access is crucial to achieving the elimination goal in China as early as 2033. The combined strategy of mass dog vaccination and IBCM could substantially reduce the rabies burden and costs of unnecessary PEP use. Eliminating dog-mediated rabies in China cost-effectively requires the "One Health" approach to prioritize progress in the livestock/veterinary sectors for widespread rural dog vaccination. The successful implementation of IBCM also calls for collaboration between the health and livestock/veterinary sectors. It would be the most cost-effective strategy recommended for interrupting rabies transmission in China, thus eventually contributing to the global plan of ending dog-mediated human rabies death.