The dual burden of animal and human zoonoses: A systematic review

Background Zoonoses can cause a substantial burden on both human and animal health. Globally, estimates of the dual (human and animal) burden of zoonoses are scarce. Therefore, this study aims to quantify the dual burden of zoonoses using a comparable metric, “zoonosis Disability Adjusted Life Years” (zDALY). Methodology We systematically reviewed studies that quantify in the same article zoonoses in animals, through monetary losses, and in humans in terms of Disability Adjusted Life Years (DALYs). We searched EMBASE, Web of Science, Scopus, PubMed, and Google Scholar. We excluded articles that did not provide the data to estimate the zDALY or those for which full text was not available. This study was registered at PROSPERO, CRD42022313081. Principal findings/Significance We identified 512 potentially eligible records. After deduplication and screening of the title and abstract, 23 records were assessed for full-text review. Fourteen studies were included in this systematic review. The data contains estimates from 10 countries, a study at continental level (Asia and Africa), and 2 studies on a global scale. Rabies was the most frequently reported zoonosis where zDALYs were calculated, based on the following included studies: for Kazakhstan 457 (95% CI 342–597), Viet Nam 5316 (95% CI 4382–6244), Asia 1,145,287 (90% CI 388,592–1,902,310), Africa 837,158 (90% CI 283,087–1,388,963), and worldwide rabies 5,920,014 (95% CI 1,547,860–10,290,815). This was followed by echinococcosis, the zDALYs in Peru were 2238 (95% CI 1931–2546), in China 1490 (95% CI 1442–1537), and worldwide cystic echinococcosis 5,935,463 (95% CI 4,497,316–7,377,636). Then, the zDALYs on cysticercosis for Mozambique were 2075 (95% CI 1476–2809), Cameroon 59,540 (95% CR 16,896–101,803), and Tanzania 34,455 (95% CI 12,993–76,193). Brucellosis in Kazakhstan were 2443 zDALYs (95% CI 2391–2496), and brucellosis and anthrax in Turkey 3538 zDALYs (95% CI 2567–6706). Finally, zDALYs on leptospirosis in New Zealand were 196, and Q fever in Netherlands 2843 (95% CI 1071–4603). The animal burden was superior to the human burden in the following studies: worldwide cystic echinococcosis (83%), brucellosis in Kazakhstan (71%), leptospirosis in New Zealand (91%), and brucellosis, and anthrax in Turkey (52%). Countries priorities on zoonoses can change if animal populations are taken into consideration.


Unfunded studies
Enter: The author(s) received no specific funding for this work. Describe where the data may be found in full sentences. If you are copying our sample text, replace any instances of XXX with the appropriate details.
If the data are held or will be held in a public repository, include URLs, accession numbers or DOIs. If this information will only be available after acceptance, indicate this by ticking the box below. For example: All XXX files are available from the XXX database (accession number(s) XXX, XXX.).
• If the data are all contained within the manuscript and/or Supporting Information files, enter the following: All relevant data are within the manuscript and its Supporting Information files.
• If neither of these applies but you are able to provide details of access elsewhere, with or without limitations, please do so. For example: Data cannot be shared publicly because • All relevant data are within the manuscript and its Supporting Information files. Scripts are available at https://github.com/LizPNZ/Dual-burden-of-zoonosis of [XXX]. Data are available from the XXX Institutional Data Access / Ethics Committee (contact via XXX) for researchers who meet the criteria for access to confidential data.
The data underlying the results presented in the study are available from (include the name of the third party and contact information or URL). This text is appropriate if the data are owned by a third party and authors do not have permission to share the data. Zoonoses can cause a substantial burden on both human and animal health. Globally, estimates 10 of the dual (human and animal) burden of zoonoses are scarce. Therefore, this study aims to 11 quantify the dual burden of zoonoses using a comparable metric, "zoonosis Disability Adjusted Life 12 Years" (zDALY).

13
Methodology/Principal Findings 14 We systematically reviewed studies that quantify in the same article zoonoses in animals, through 15 monetary losses, and in humans in terms of Disability Adjusted Life Years (DALYs). We searched 16 EMBASE, Web of Science, Scopus, PubMed, and Google Scholar. We excluded articles that did not 17 provide the data to estimate the zDALY or those for which full text was not available. This study 18 was registered at PROSPERO, CRD42022313081.

Conclusions/Significance 20
We identified 512 potentially eligible records. After deduplication and screening of the title and 21 abstract, 23 records were assessed for full-text review. Fourteen studies were included in this 22 systematic review. The data contains estimates from 10 countries, a study at continental level 23 (Asia and Africa), and 2 studies on a global scale.

24
The total burden from these 14 studies in zDALYs is 11,015,438 (CI 95%: 6,235,806,100), 25 most of which is attributable to rabies and echinococcosis worldwide, including the Animal Loss 26 Equivalent (ALE) is 4,936,233 (CI 95%: 3,512,357,435). These results are based on ten 27 zoonotic diseases (rabies, echinococcosis, cysticercosis, brucellosis, leptospirosis, anthrax, Q-28 fever, CCHF, tularemia, and toxoplasmosis) which had the biggest impact on the public health 29 sector.

30
Author Summary 31 Zoonoses impact humans and animals in several ways. Unfortunately, the burden of zoonoses is 32 usually not characterized and quantified through integrated human and animal metrics. Our study 33 is the first systematic review to assess the dual burden of zoonotic diseases in humans and animals 34 globally. In the considered set of human and animal burden of zoonoses, 45% of the zDALY was 35 due to animal disease. Therefore, metrics encompassing both burdens are likely to change 36 decision-making regarding the prevention and control of zoonoses. Implementing a One Health 37 approach will require the application of such metrics. We believe that quantification of the dual 38 burden of the diseases is a key to improving zoonosis prioritization decision-making, and resource allocation. This study outlines the need for integrated studies on zoonoses and reporting of data 40 with a comparable metric.

41
Introduction 42 Zoonoses are diseases that can be transmitted directly or indirectly from animals to humans (and should be organized in an integrated way [4]. This allows for a comprehensive risk assessment and 54 the design of appropriate responses [5].

55
The business case for a One Health (OH) approach to mitigation of zoonoses has been presented 56 as a framework [6] which includes the creation of one health surveillance and response programs 57 for future emerging diseases. Animal health surveillance data can be used to inform public health 58 messaging, control measures along the food chain, and establish public health surveillance if a 59 pathogen is present in the human population and public health action is required.
In general, the impact of zoonotic diseases on the human population is measured by financial cost,

66
The direct impact of animal disease is studied using various economic models. For example, the 67 burden of diseases can be quantified through the money spent on the disease intervention 68 programs, or money accounted for the loss of animal productivity (less milk/meat yield, etc.). The 69 challenge of economic analysis in a OH context is that the boundaries of the system for which costs 70 and benefits incur can be extended or restricted arbitrarily and hence alternative economic 71 models are needed.

72
A pragmatic approach to consider the combined burden on human and animal health has been 73 proposed as "zoonosis Disability Adjusted Life Years" (zDALYs) [9]. The zDALYs extends the DALY 74 framework to domestic animals. The idea behind this indicator is that the animal burden estimated 75 as monetary losses can be converted to Animal Loss Equivalents (ALE). The ALE is basically a metric 76 that reflects the time trade-off for human life years to "replace" the animal loss, e.g., it is the 77 amount of time that a farmer would need to spend to recover the losses.  We computed the zDALY, adding the DALY of the findings to the ALE that we estimated.
To account for the uncertainty of all estimates, we generated random numbers between the lower 115 and upper bounds of the distributions from the previous studies. We set 100,000 iterations for 116 each estimation. According to the original studies, we reported the 50, 2.5, and 97.5 percentiles 117 of the estimates, and 50, 5, 95 percentiles. We have also kept the terms that previous studies used 118 to express uncertainty. 119 We performed the analyses in R 4.1.3. Scripts are available at https://github.com/LizPNZ/Dual-120 burden-of-zoonosis 121 The stochastic approaches used to calculate zDALYs imply that sensitivity is included in the 122 calculations. 123 We estimated ALEs and zDALYs for all countries with available data over the study period. We  This study is registered at PROSPERO, CRD42022313081.

131
We identified 552 articles through electronic database searches (Figure 1). After removing 140 132 duplicates, 412 articles were screened for titles and abstracts. The full texts of 23 articles were 133 reviewed and 9 were excluded at this stage. Thus, 14 articles are included in this review (Table 1,     For brucellosis, the Kazakh study only accounted for losses due to slaughtering of the animals and 178 subsequent compensation. Whereas the Turkish study also considered reduced productivity.

179
Besides, the Turkish study was the only one that included bacterial, parasitic, and viral zoonoses.

180
However, we only determined the ALE for brucellosis and anthrax since the animal loss was only 181 available for those diseases. We calculated the total zDALY for all the diseases included in this 182 study.

183
Adding all the adjusted estimates for each study, we obtained a zDALY of 11,015,438 (95% CI:  Since the studies that already estimated zDALYs did not meet the inclusion criteria, we added 187 their findings in the supporting information p 4.

188
The full ROBIS assessment is provided in the supporting information pp 8-12. Overall, the risk of 190 bias for this study is low. According to the signaling questions, there were no concerns regarding 191 all the domains (study eligibility criteria, identification, selection of studies, and data collection).

192
Therefore, the review is likely to include a high proportion of relevant studies.

193
However, the last domain (synthesis and findings) outlines that no meta-analysis was performed.

194
We report the reasons in the discussion.

195
The PRISMA checklist is provided in the supporting information pp 13-15.

197
We report the first systematic review that estimates the dual burden of zoonoses in humans and   one must ask is whether this is due to a lack of data from the animal population or if it is because 287 only losses to farmers due to animal zoonosis account for the ALE.

288
Our estimations are based on the results of previous studies which is a limitation of this study, 289 besides the small number of papers. In some cases, the data available for humans and animals 290 were not from the same period, reducing the accuracy of the estimations. Only three studies 291 shared their code for the analysis (one of them partially), making the rest of the studies not 292 reproducible. Also, the lack of availability of datasets following the FAIR principles did not allow us 293 to obtain the confidence intervals of our choice. This shows the need for FAIR data application in 294 the health area [37][38][39]. The lack of data continues to be a challenge, as the approach that is used 295 to analyze it. We did not perform a meta-analysis due to the high variability among studies, including the type of study, and analysis design. This is also evidence of a lack of standardized 297 methods to unify the burden caused by zoonoses in humans and animals in the past, and the 298 unfamiliarity of the existing metrics available for that aim.

299
The strength of this study consists of an extensive literature search in different databases without 300 an initial time restriction. Considering that the GBD study does not include most of the zoonoses 301 burden, as well as the animal burden of zoonosis, we integrated this data into the human burden 302 among the studies available worldwide. The DALY is a metric used to prioritize international 303 disease-control investments. However, its use has been debated for various, primarily ethical, 304 reasons. Among which is a limited applicability to neglected tropical diseases (NTDs). Most NTDs 305 in this study have a low chronic morbidity that accounts only for a small portion of DALY. In low-306 income settings, where poverty is dominant, this low morbidity raises little attention. Half of the 307 world's hungry are subsistence farmers and rely heavily on agriculture for their livelihoods. [40] 308 However, subsistence farming and hard physical work are common in those settings and the 309 disabling effects of the NTDs are a main source of poverty. This circular causality cannot be 310 captured through DALY calculations. The zDALY, at least, allows to include the burden from animal 311 health losses, which are highly relevant in most poverty settings. How much subsistence farmers 312 lose due to a zoonotic disease and how long it will take them to recover their losses should receive 313 more attention in public health policy as it addresses an important determinant of human health.

314
Regarding vector-borne zoonoses, the only reported were tularemia and Crimean-Congo 315 hemorrhagic fever (CCHF) in Turkey but without a direct association of their animal losses. We 316 suggest establishing databases that incorporate human and animal diseases for each country, thus on a global scale. For example, complement the GBD database with ALEs to move towards better 318 integration of human and animal health policies.

319
A remaining challenge for the zDALY are animals without traded economic value. Therefore, other 320 methods for estimating the ALE component of the zDALY (e.g., willingness to pay, pairwise 321 comparisons or direct time trade off) in analogy to ecosystem services should be explored. Not 322 only are more comprehensive metrics needed, but also a more integrative effort and support to 323 face zoonosis in LICs and LMIC. For this endeavor, we consider the zDALY represents a step 324 towards progress in zoonosis prioritization.