A scoping review of One Health interventions 2005–2025: What has been done, are they effective, and what are the best practices?

Abstract

Background

One Health research and practice are evolving rapidly, necessitating solid evidence-building and quality assessment of One Health interventions. We conducted a scoping review to (1) characterize how many and what kinds of One Health interventions are published in peer-reviewed literature; and (2) assess whether the identified interventions demonstrated effectiveness. We hypothesized that the evidence was limited in numbers and ability to demonstrate effectiveness.

Methods

The review was reported according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews. The initial search was conducted on Medline (Ovid), Embase, PubAg, and Web of Science in May 2024 and updated in December 2025. Inclusion criteria were: Primary research in which the investigators intervened with humans, animals, or their shared environment (at least two of three domains); an outcome assessment was included, and full text was available in English, published on or before December 16, 2025. Two researchers independently screened and analyzed all manuscripts.

Results

Over 18,000 references were screened, and 151 studies were included. Studies were conducted in 53 countries representing all six WHO Regions. The highest number of interventions were reported from the United States of America, Tanzania, and India. Among 79 interventions that targeted zoonoses, the most common diseases were general zoonoses, rabies, and cysticercosis. Over half (55%) of the study designs were quasi-experimental, followed by 23% randomized control trials (RCTs) and 19% non-randomized experiments. Less than half (41%) of the studies had a control or comparison group. The most common perceived sources of bias were immigrative (28%) and emigrative selection bias (11%). Eight field-based RCTs identified that a One Health intervention was effective in reducing disease occurrence or pathogen detection compared to the control. Exemplary practices included a comprehensive intervention spanning human, animal, and environmental domains, using a participatory approach, assessing longer-term outcomes, triangulating results using more than one method, not overstating findings, and providing a visual aid to describe the intervention.

Conclusions

A small number of field-based RCTs clearly demonstrated the comparative effectiveness of One Health interventions, while it was challenging to assess the effectiveness of other studies without a more granular definition of effectiveness. Rigorous studies that assess the effectiveness of One Health interventions will collectively improve buy-in from governments, researchers, and policymakers, and advance the evidence base for employing the One Health approach.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s42522-026-00229-0.

Background

One Health is a collaborative and holistic approach that integrates across domains of human, animal, and environmental health; therefore, interventions that employ a One Health approach are transdisciplinary experiments, projects, or programs that aim to improve the health among these domains. These approaches contrast with siloed approaches, where human, animal, and environmental health challenges are addressed separately, without recognizing their interconnected nature, missing the opportunity to address upstream drivers of health and to take advantage of shared opportunities for intervention delivery. The One Health approach has gained traction globally in academic settings [1, 2], public health practice [3, 4], and international pandemic agreements [5]. As the field of One Health grows, so does the need for solid evidence and quality assessment of past and current One Health initiatives. For example, donors need to evaluate evidence to justify sustained funding; scientists need to provide specific evidence for the benefits of employing a One Health approach; and policymakers need to demonstrate why One Health can better achieve improved health outcomes and cost-savings compared to more traditional, but siloed approaches.

In response to these gaps, researchers have begun to assess the effectiveness of One Health initiatives. For example, in 2016, Baum et al. assessed One Health case studies and frameworks and found that only 7 of 39 eligible studies reported quantitative outcomes, with the effectiveness of One Health often assumed or not supported by data [6]. In 2018, Falzon et al., reviewed primary research and reported that a majority of studies demonstrated positive or partially positive monetary and health outcomes of One Health initiatives, while few papers reported negative monetary outcomes or non-significant improvement in health outcomes [7]. In 2025, Omuse et al. reviewed One Health interventions conducted in rural smallholder farming settings in Africa and recommended the need for “harmonised outcome evaluation metrics to avoid under- or over-estimation of OH benefits.” [8] These efforts consistently highlighted the need to assess the effectiveness of One Health interventions based on evidence, not on assumptions.

Therefore, the current review was conducted to evaluate the evidence base for One Health interventions, contributing over a decade of new studies supported by growing investment in One Health. This review also provides one of the first systematic quality assessments of the literature—specifically the reproducibility, generalizability, perceived sources of bias, and study limitations. The research questions addressed in this paper include: (1) how many and what kinds of One Health interventions are published in peer-reviewed literature? (2) Do the identified manuscripts demonstrate the effectiveness of the One Health approach? With the hypothesis that evidence in the literature remains limited to demonstrate the comparative effectiveness of interventions that employ a One Health approach versus more traditional approaches, the intent of this scoping review is to summarize the evidence, inform future hypotheses for a systematic review with a focused research question, as well as serve as an evidence base for policymakers, funders, and designers and implementers of interventions. The primary objective of this study was to conduct a scoping review to identify and characterize the literature about interventions that employ a One Health approach; the secondary objective was to conduct a preliminary assessment regarding whether there is sufficient evidence that One Health interventions are as, or more effective than traditional approaches.

Methods

This scoping review was reported according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews (PRISMA-ScR) [9]. There were no existing protocols that suited our goals to the best of our knowledge, so we developed a new protocol. We preregistered this study before conducting the scoping review on the Open Science Framework [10] (Registration Digital Object Identifier https://doi.org/10.17605/OSF.IO/C9SMB).

Definitions

In this paper, we defined the following terms as below:

One Health is an “integrated, unifying approach that aims to sustainably balance and optimize the health of people, animals and ecosystems,” which recognizes that “the health of humans, domestic and wild animals, plants, and the wider environment (including ecosystems) are closely linked and interdependent.” [11] Under this definition, the natural or built environments, ecosystems, and plant health all fulfilled the environmental health domain, and either or both domestic or wild animals fulfilled the animal health domain.

A One Health intervention is “an activity jointly undertaken by at least two sectors (multi-sectoral), engaging multiple disciplines (multi-disciplinary), aimed at improving the efficacy and efficiency of different human, animal (terrestrial and aquatic), plant and environment health systems services.” [12] The choice to use at least two sectors (e.g. human and animal, animal and environment), rather than three (e.g. human, animal and environment) as has been espoused [13], was pragmatic and meant to be more inclusive, given how few interventions exist that address human, animal, and plant or environmental dimensions simultaneously under the same project or program. In the first step, where we defined One Health, we required that the intervention encompass at least two of the three sectors. In the second step, where we conducted the title and abstract screening, two analysts independently assessed the content to determine whether it employed a One Health approach. Although this approach was more time-intensive, it allowed us to balance the tension between casting a sufficiently wide net and including only relevant papers. We chose this approach based on our preliminary assessment before conducting the scoping review, in which we identified very few manuscripts on One Health interventions.

Study designs were defined as shown in Table 1.

Table 1.

Study design definitions applied to One Health interventions

Study design	Definition
Randomized Control Trial	An experimental study in which the investigator randomly assigns the intervention to the intervention and control groups. Randomization units include individuals and groups. (i.e., cluster randomized control trials)
Non-randomized Experimental Trial	An experimental study in which the investigator assigns the intervention non-randomly to the intervention and control groups.
Quasi-experimental study	A catch-all term for an experimental study that is not a randomized controlled trial or a non-randomized experimental trial.
Qualitative Study	A study that uses qualitative methods, including key informant interviews, focus group discussions, and observations, to investigate people’s perceptions or investigates research questions that focus on “how” and “why” things occur.
Economic evaluation	A study that measures the economic metrics of an intervention, including benefit-cost analysis, return on investment, etc.

Manuscript identification

We defined eligible manuscripts in a structured manner. First, we searched for review papers on One Health interventions in PubMed. Given that there were limited reviews conducted on this topic and that no protocols or data extraction forms had been validated for identifying and documenting One Health interventions, we developed our own search terms and data extraction templates. Then, we identified five key papers that fit the scope of our review and developed search terms based on them. The five key papers were identified through the initial search on PubMed, which included papers that were as diverse as possible in terms of study themes covered (e.g., food safety, antimicrobial resistance, and zoonotic diseases), outcomes measured (e.g., knowledge, attitudes, and practices, willingness to pay, and disease prevalence), and geographic distribution, including studies based in East Asia, Southeast Asia, and sub-Saharan Africa. The population, intervention, comparison, outcome (PICO) model [9] for this review is shown in Table 2. Although the Population, Concept, Context (PCC) framework is commonly used in scoping reviews, PICO was more appropriate for our study, because the topic of the scoping review was on interventions.

Table 2.

Population, intervention, comparison, outcome (PICO) model for One Health interventions

P: Population	Human, animal, and the environment (e.g., water, air, soil, bacteria, and other components of the ecosystem). No restrictions on place, age group, or animal species.
I: Intervention	An intervention conducted by investigators to improve health outcomes in human or animal populations or environmental components.
C: Comparison	A group of humans or animals that did not receive the intervention. A comparison or control group was not required.
O: Outcome	Any health outcome in humans, animals, plants, or the environment.

We searched four databases: Medline (Ovid), EMBASE (biomedical science), PubAg (agriculture), and Web of Science (general science) to represent the transdisciplinary scope of One Health. We conducted the search in late April to early May 2024, then an updated search in mid-December 2025.

Contrary to some previous literature reviews for One Health, which required the title or abstract to explicitly use the term “One Health,” we cast a broader net to identify manuscripts with a One Health scope in two ways. First, we used adjacency searches on Medline (Ovid) and EMBASE, which enabled us to search for titles and abstracts that used the words “human”, “animal”, “environment”, and “health” within 5 words of each other. Second, in addition to the words “One Health”, we included other conceptually similar terms such as EcoHealth, Planetary Health, and One Medicine. The final search terms are shown in the supplemental material.

Inclusion and exclusion criteria

Inclusion of papers was selective in that we did not automatically include study themes that are typically recognized as One Health, such as zoonotic disease and antimicrobial resistance. The inclusion and exclusion criteria are shown in Table 3.

Table 3.

The inclusion and exclusion criteria for the scoping review of One Health interventions

Inclusion Criteria

Exclusion Criteria

Primary research studies published as a scientific manuscript or book chapter, in/for which:  • Investigators intervened with humans, animals, or their shared environment (at least two of three domains fulfilled without the analysts having to make inferences)  • An outcome assessment for the intervention was included.  • Full text was available in English.  • Published on or before December 16, 2025. No earlier time limit.

Primary research studies published as a scientific manuscript or book chapter, but:  • Investigators intervened with one domain (human or animal or environment) only  • Did not include an outcome assessment for the intervention.  • Full text was not available in English.  • Were retracted. OR Commentaries, perspectives, strategic plans, program descriptions, training protocols, and conference abstracts not tied to a manuscript.

If the work was a conference abstract and not a full scientific manuscript, we identified and included the corresponding manuscript whenever possible by searching the title on an online search engine. If the manuscript became available online first, then as a print later across two years (e.g., published online in late 2020, and published as a journal issue in early 2021), we included the earlier year as the year of publication. To fulfill the “animal” component, we required the study to include live animals, including embryos. In other words, if the study only included an animal product, such as meat, it did not satisfy the animal component according to our pre-determined definition.

We used the following rules for this review: (i) the participating animal population must be alive (i.e., a meat product alone would not fulfill the animal component); (ii) a microbiome study would not be automatically included, even though the human or animal body is the environment for commensal bacteria, unless it demonstrated a One Health approach; (iii) studies of pathogens of humans would not be automatically included (e.g., pathogenic bacteria alone do not fulfill the environmental component, pathogenic parasites alone do not fulfill the animal component) unless it demonstrated a One Health approach; (iv) studies that mention the downstream One Health impacts of the study alone would not be sufficient to be considered (e.g., discussing that a One Health approach is needed to solve the challenge alone does not suffice); and (v) studies that tested the efficacy of refined plant extracts or isolated strains of bacteria as an experimental testing agent (e.g., strain of Lactobacillus or neem oil) were not considered a part of the environment. Retracted articles were removed.

Manuscript screening and data extraction

Search results from each database were imported into a citation management program, Zotero (Corporation for Digital Scholarship, Virginia, United States), and duplicates were removed. The unique records were then imported into Covidence systematic review software (Veritas Health Innovation, Melbourne, Australia) for title/abstract and full text review.

Two researchers independently screened and analyzed all manuscripts. One researcher had a background in veterinary medicine, field epidemiology, and infectious disease surveillance and outbreak response. The other researcher had a background in laboratory science, microbiology, zoology, and epidemiology. Where there was disagreement in their decisions, the two analysts discussed the discrepancy in reaching an agreement. If there was further disagreement, the researchers had a procedure to seek the opinion of a third researcher to reach a decision, but they did not need to do so. The following information was extracted in Covidence (Veritas Health Innovation, Melbourne, Australia), which was downloaded in tabular format using Microsoft Excel (Microsoft Corporation, Washington, USA). Final analyses were conducted on R Software version 2022.12.0 (R Core Team, Vienna, Austria). When appropriate, we contacted the corresponding author of the manuscript by email or by online public research portals such as Research Gate to verify the information. We summarized the main findings and assessed perceived biases, generalizability, reproducibility, and perceived conflicts of interest.

The data extraction template (see Supplemental material) included components such as journal, year published, years the study was conducted, country, the aims, design, and theme [14, 15] of the study, possible sources of bias, type of population included (human, animal, or environment) and their descriptors (sample size, data collection method of each population), type of animal population, ethics statement, pre-registration, inclusion/exclusions criteria, type of analysis, descriptor of intervention(s), types and detail of outcomes including effect estimates, assessments of reproducibility and generalizability, stated and perceived limitations, and a semi-quantitative One Health grading assessment. The data extraction template was piloted, revised, and re-piloted on predefined papers until two analysts agreed that the template was finalized. All papers were assessed using the finalized template. No simplifications were made when extracting data.

Process outcome

We calculated interrater reliability using Cohen’s Kappa statistic, which is a measure of agreement of decisions between two analysts accounting for their agreement by chance. At the 50% mark of the initial data extraction (July 2025), the two analysts held a discussion on their positionality and reflexivity to identify potential implicit biases.

Quality assessment and synthesis

Quality assessment of each manuscript was conducted using the following categories: The scope of One Health (i.e., the breadth of human, animal, and environment that it covers), description of methods for reproducibility (adapted from a previous scoping review for One Health by Falzon and colleagues [7]), level of generalizability, reviewers’ perceived bias, reviewers’ perceived conflict of interest, and barriers to implementation stated by the authors. We also assessed whether the One Health paper best fit one of the following categories suggested by the Checklist for One Health Epidemiological Reporting of Evidence (COHERE), which is a checklist for epidemiological reporting in One Health [13]: One Health, One Medicine, Environmental Health, and Veterinary Preventive Medicine. When definitions from COHERE did not fit our colloquial use of the definitions for some manuscripts, we listed both. Additionally, we conducted a semi-quantitative assessment of the breadth of One Health using a point scale of 1.25 to 3.00, indicating the level of active involvement of, intervention on, and measurement of human, animal, and environmental disciplines. We did not make assumptions or simplifications to the data. We synthesized the results by conducting descriptive analyses.

Assessment of the extent of “One Health-ness”

The One Health-ness of the manuscripts was assessed in two ways: (1) a semi-quantitative grading based on whether and to what extent human, animal, and environmental domains were involved in the interventions (Table 4), and (2) previously validated categories identified in COHERE (One Health, Veterinary Preventive Medicine, Environmental Health, and One Medicine) [13]. The semi-quantitative grading scale was developed to incorporate whether humans, animals, or the environment were involved in the study, as well as how “active” these domains were within the study being scored. The first digit delineated how many domains of One Health were fully involved, active, and had measurements taken within the study. The decimals measured the level of involvement of other domains that were technically participating in the study but were not fully active or did not have measurements taken. This led to increments of 0.25, and the highest achievable value on the score was 3, showing that three domains were fully involved in the intervention.

Table 4.

Definitions used for the semi-quantitative grading of One Health interventions

Semi-quantitative One Health grading	Definitions
3.0	Humans, Animals, and the Environment are all active participants in the study and are having measurements taken on all three populations.
2.75	Humans, Animals, and the Environment are all involved in the study, but only two are active participants with measurements being taken on all three populations.
2.5	Humans, Animals, and the Environment are all involved in the study, but only one is an active participant with measurements being taken on at least two populations.
2.25	Humans, Animals, and the Environment are all involved in the study, but only one is an active participant with measurements being taken on one population.
2.0	Two of the three populations are active participants in the study and are having measurements taken on both populations.
1.75	Two of the three populations are active participants in the study and are having measurements taken on one of the populations.
1.5	Two of the three populations are involved in the study, but only one is an active participant with measurements being taken on both populations.
1.25	Two of the three populations are involved in the study, but only one is an active participant with measurements being taken on one population.

*"active participants" also include actively that influences the outcome related to the environment. The final set of included manuscripts were downloaded from Covidence onto Zotero

The following sources of bias were assessed: immigrative selection bias, emigrative selection bias, measurement error, ecological bias, confounding, and other. These sources of bias were assessed by two analysts based on study design characteristics described in the manuscript, including sampling strategy, randomization, data collection methods, and participant attrition. Categorizing biases required in-depth discussions between the two investigators.

Generalizability (external validity) was assessed by the extent to which the intervention could be conducted in other settings and generate similar results by examining the participating populations, sampling strategy, intervention implementation, and outcome measurement.

Reproducibility was assessed by the extent to which the intervention could be repeated by other investigators, by considering the authors’ clarity in describing and justifying their decisions related to participant selection, intervention implementation, and other study characteristics.

Ethics

Ethics approval was not required for this study. The secondary data generated consisted of publicly available, peer-reviewed publications. There were no primary participants in this study.

Results

Scoping review process

A total of 36,224 references were imported from Medline (Ovid), Embase, Web of Science, and PubAg. Once duplicates were removed, 18,344 references went through title and abstract screening, and 296 references were considered for full text review. Ultimately, 151 studies were eligible, and the details were extracted (Fig. 1).

Fig. 1.

Scoping review process for identifying One Health intervention manuscripts

In the first step of title and abstract screening, Cohen’s Kappa, which measures the inter-rater reliability between the two analysts, was 0.28, interpreted as “fair agreement” [16]. In the second step of full text review, Cohen’s Kappa was 0.65, interpreted as “substantial agreement.” Reflections on investigator reflexivity and possible implicit biases were related to research interests, education, familiarity with authors, expectations of studies being conducted in high- vs. low- or middle-resource settings, and field of expertise (e.g., one investigator had more training in laboratory science while the other had worked in disease surveillance and response).

Study characteristics

The studies were conducted in 53 countries across all six WHO Regions: Africa, the Americas, Eastern Mediterranean, Europe, Southeast Asia, and the Western Pacific (Fig. 2). One study was a multi-country study conducted across six Latin American countries, and another was conducted in Spain and Portugal. The largest numbers of studies were conducted in the United States (n = 16), Tanzania (n = 11), and India (n = 9), followed by Brazil and Thailand (n = 8 each), then Australia and Germany (n = 7 each). Among the top three countries, the study topics differed. In the United States, the study topics were diverse, including antimicrobial resistance (AMR), climate change education, lead poisoning, hand hygiene, risk communication on rabies, and providing companion animal care among underserved communities. In Tanzania, the topics were predominantly focused on zoonoses, where six of 11 interventions were conducted to control Taenia solium, followed by schistosomiasis (n = 2), rabies (n = 2), and sanitation and hygiene (n = 1). In India, three studies targeted rabies, and others studied integrated pest management, anthrax, brucellosis, arboviral diseases, and AMR.

Fig. 2.

Countries in which 151 One Health interventions were conducted (53 countries) Boundaries shapefile: geoBoundaries (https://www.geoboundaries.org/). Projection in Equal Earth

The most commonly reported study themes were zoonoses and emerging and reemerging infectious diseases (52%, 79/151), followed by environmental contamination (21%, 32/151), and food and water safety/security (19%, 28/151) (Fig. 3a). Over half of interventions were associated with education and knowledge, attitudes, and practices (KAP) (64%, 97/151), programmatic interventions (47%, 71/151), mass drug administration/mass vaccination (14%, 21/151), and farming practices (11%, 16/151) (Fig. 3b). Four studies were conducted in a laboratory. There was a marked increase in the number of publications between 2005 and 2025, with the largest number of publications recorded in 2025, with 36 publications (Fig. 3c).

Fig. 3.

Study themes (a), intervention types (b), and publications by year (c) of One Health intervention studies identified in the scoping review (n = 151)

Among 79 manuscripts that had an intervention targeted at addressing zoonotic diseases, the five most common types of diseases were general zoonosis, followed by rabies, cysticercosis/taeniasis (Taenia solium), Chagas disease, and dengue (Table 5). Although not a zoonotic disease, two additional manuscripts had an intervention to address transboundary animal diseases, including foot and mouth disease.

Table 5.

The types of zoonotic disease targeted by the One Health interventions (n = 79)

Type of zoonosis target in the One Health intervention	Number of manuscripts
General zoonosis	12
Rabies	12
Cysticercosis/taeniasis (Taenia solium)	11
Chagas disease (Trypanosoma cruzi)	6
Dengue	6
Opisthorchiasis (Opisthorchis viverrini)	5
Anthrax	3
Avian influenza	3
Schistosomiasis	3
Vector-borne disease (General)	3
Cystic Echinococcosis	2
Malaria	2
Nipah virus disease	2
Tungiasis	2
Bacteria contaminating milk	1
Brucellosis	1
Cryptosporidiosis	1
Guinea worm	1
Intestinal parasitic infections	1
Methicillin-resistant Staphylococci and ESBL-Escherichia coli	1
Porcine gastrointestinal nematodes and ectoparasites	1
Q fever (Coxiella burnetii)	1
Rocky Mountain spotted fever	1
Soil-transmitted helminths	1
Tick-borne encephalitis	1
Visceral leishmaniasis	1

*There were four manuscripts that addressed more than one zoonotic disease. Therefore, the right column does not sum up to n = 79

Studies were published in 92 unique scientific journals focused on various topics, including public health and primary care, medicine, veterinary science, One Health, EcoHealth, infectious diseases, agriculture, environmental health, and entomology. The top 15 journals that most frequently published One Health intervention studies are listed in Table 6.

Table 6.

Journals that published One Health intervention manuscripts

Rank by frequency	Journal	Number of papers identified
1	EcoHealth	9
2	One Health	7
3	PLOS Neglected Tropical Diseases	6
4	Acta Tropica	5
4	Preventive Veterinary Medicine	5
6	BMC Medical Education	4
6	BMC Public Health	4
8	American Journal of Tropical Medicine and Hygiene	3
8	Animals	3
8	CABI One Health	3
8	Frontiers in Veterinary Science	3
8	International Journal of Infectious Diseases	3
8	Journal of Veterinary Medical Education	3
8	One Health Outlook	3
8	Transactions of the Royal Society of Tropical Medicine and Hygiene	3

When searching for the use of the terms “One Health”, “EcoHealth”, or “Planetary Health” in the title or abstract, 58% (87/151) of manuscripts used at least one of the three terms, while 42% (64/151) did not use any. Among papers that used one of the three terms, 68% (59/87) used “One Health,” 20% (17/87) used “Planetary Health,” and 16% (14/87) used “EcoHealth.” All but three manuscripts that included at least one term (97%, 84/87) exclusively used one of the three terms, whereas three manuscripts [17–19] used “One Health/EcoHealth” in the abstract.

Study design

The study designs of the included papers were 55% (83/151) quasi-experimental studies, 23% RCTs (35/151), including one study that used a case-crossover design, and 19% (29/151) non-randomized experimental studies. Less than half (41%, 62/151) of the studies had a control or comparison group. There were 7% (11/151) of papers that conducted economic analyses, including benefit-cost analyses, and the calculation of disability-adjusted life years (DALYs) averted and cost-per dose of vaccines. Two other papers reported on the price of intervention materials or reported a change in the income of farmers, although they were not formal economic analyses. Eighteen studies (12%, 18/151) used qualitative methods with key informant interviews (n = 10), focus group discussions (n = 10), investigator observations (n = 4), and participants taking photographs of the intervention experience (n = 1). Three studies exclusively used qualitative methods, whereas the other 15 were mixed-methods designs that combined qualitative data with a survey.

Fourteen papers (9%, 14/151) mentioned pre-registering their study to a clinical trial registry, including ClnicalTrials.gov (USA), German registry for clinical trials, the Australian New Zealand Clinical Trials Registry, and the Uganda National Council for Science and Technology. The length of the intervention ranged from less than one year to 25 years for those who provided years during which the study took place.

“One Health-ness”

The median score for the semi-quantitative One Health grading was 2.0. The data was skewed, where 25% of papers (38/151) scored 1.25, indicating that only one of three domains (human, animal, or environment) was an active participant with measurements being taken on one population. Most of these papers were educational interventions that measured knowledge, attitudes, and practices in people without involving animals or the environment in the study. Another 19% (28/151) were graded as 2.25, and 15% (23/151) were graded as 2.0 (Fig. 4). Only 4% (6/151) of papers were graded at 3.0, which is the highest level achievable where humans, animals, and the environment are all active participants in the study, with measurements being taken on all three populations. The categorization of the number of manuscripts that represented human, animal, and the environment is represented as a Venn diagram in Fig. 5.

Fig. 4.

The number of One Health intervention papers categorized by the semi-quantitative One Health grading where 1.25 is the lowest and 3.0 is the highest achievable score

Fig. 5.

Human, animal, and environmental components that were targeted in 151 One Health interventions, and examples of the topics covered in each domain. Papers that reported only one domain did not meet our inclusion criteria (i.e., no number shown in the Venn diagram for papers that were categorized into the human-only, animal-only, and environment-only domains)

Among 120 studies in which animals were involved, the categories of animals that were targeted were wild or feral animals (47%, 56/120), domesticated species (40%, 48/120), pets or companion animals (29%, 35/120), and the general animal population (3%, 3/120). Twenty studies (17%, 20/120) involved more than one animal type.

Bias, reproducibility, generalizability, and exemplary practices

The most common perceived source of bias was immigrative selection bias (42/151, 28%), followed by emigrative selection bias (16/151, 11%), measurement error (10/151, 7%), confounding (5/151, 3%), and ecological bias (4/151, 3%). One study introduced a bias by selecting a control community based on low disease prevalence, while the intervention community was not selected based on that criterion [20].

Regarding reproducibility, 64% (97/151) of manuscripts had the study design and intervention characteristics clearly explained and reproducible, 28% (42/151) lacked certain information and were partly reproducible, and 5% (8/151) were insufficiently described. Four papers were marked as “other,” including two studies that were clearly explained but still only partly reproducible; one study was a qualitative study for which dependability would have been more appropriate to assess rigor; and one was a short communication piece for which the methods were not intended to be reproducible. For generalizability (external validity), 91% (137/151) of papers were categorized as generalizable to similar settings, 5% (7/151) were categorized as not generalizable, 3% (5/151) were categorized as widely generalizable, and two others were marked as “other” where generalizability was not appropriate to assess rigor.

Several exemplary practices were identified in the papers regarding study design, implementation, and reporting, presented in Table 7 with their definitions and examples. The full list of manuscripts is presented in Supplemental Table 2.

Table 7.

Summary of exemplary practices that were identified among One Health interventions, categorized by their types of strengths

Exemplary practice	Definition	Examples
Comprehensive intervention	The intervention is comprehensive and actively involves humans, animals, and the environment.	In a trial to reduce schistosomiasis among school children in Senegal, investigators removed invasive submerged vegetation in water, and used removed vegetation as feed for livestock. They assessed disease prevalence, water quality and biodiversity, and the benefit-to-cost ratios of using vegetation compared with traditional feeds [21].
Uses a participatory approach	Community actor groups are engaged in the participatory co-design and implementation of the intervention.	Authors conducted participatory action research involving community members and local decision-makers to co-design and implement an environmental health program to address challenges, including lead, dengue, waste management, and polychlorinated biphenyls (PCBs), to meet community needs in Mexico [22].
Comprehensive outcome measurement	Investigators take measurements, samples, or observations from humans, animals, and the environment to assess the effectiveness of the intervention.	An integrated intervention to control cystic echinococcosis in Italy detected the pathogen or measured disease occurrence (incidence and prevalence) in dogs, sheep, goats, cattle, buffaloes, humans, and detected the pathogen in water and on raw vegetables [23]. An intervention to improve biosecurity in pig farms assessed the intervention by collecting farm biosecurity scores by interviews, questionnaires, and veterinarians’ observations, as well as environmental sampling to measure Adenosine Triphosphate (ATP) levels using an ATP rapid test, and prevalence and genotypic analysis of Methicillin Resistant Staphylococci and Extended-spectrum beta-lactamase (ESBL)-Escherichia coli collected from farms [24].
Longer-term follow-up after the intervention	The effectiveness of the results is followed up after the intervention ended to measure the sustained effectiveness of the intervention.	Investigators measured changes in knowledge of cysticercosis one year after an education intervention in Tanzania [25].
Triangulation of results using more than one method	An outcome is compared with or verified using two or more methods of measurement.	In a zoonotic disease education intervention in South Africa, investigators used focus group discussions to triangulate findings from a survey to assess change in knowledge [26].
Creative use of setting a comparison	Sets a control or comparison group even when there are limitations in data availability.	In a cattle vaccination program against Coxiella burnetti, unvaccinated cattle served as the control group until they were eligible to be vaccinated [27]. In an economic study assessing a rabies control program in India, the authors used a hypothetical counterfactual model to compare Disability Adjusted Life Years with and without the program [28].
Succinctly answers the research question	Succinctly answers the main research question using evidence generated and does not include information that is not directly related to the research question.	The author of an ecological intervention succinctly described the impact of invasive water hyacinths on snail hosts of schistosomiasis in Lake Victoria, Tanzania [29]. In an intervention assessing the effectiveness of poultry vaccination against H5/H7 in China, authors succinctly addressed the research question in a short report with two figures using human surveillance and environmental sampling data [30].
Clear writing	Clearly guides the readers through details of complex interventions.	Authors clearly described an integrated intervention involving mass drug administration against soil-transmitted helminths in humans and mass vaccination against rabies in dogs, which was easy to follow despite the many components involved [31].
Not overstating findings and is transparent about study limitations, including barriers to study implementation	Does not overstate the effectiveness of the intervention and considers other sources that may have led to the outcome. Shows critical thinking and intellectual humility by thoroughly communicating the limitations of the study.	In a risk-based vaccination program of foot and mouth disease in cattle in the Lao People’s Democratic Republic, the authors were transparent about changes in data collection and sampling once field work started, uncertainties about whether the program was carried out according to plan, and disruptions due to the COVID-19 pandemic [32]. In a trial to reduce Nipah virus infection exposure in Bangladesh, authors considered the potential effect of social desirability bias in self-reported behavior, comparability of intervention and control groups, and altered human behavior while being observed (Hawthorne Effect) [33]. In a dog-assisted education to improve emotional comprehension among 6 to 7-year-olds in Italy, authors critically interpreted their data, including the selection of their own framework, a small convenience sample, and the shortcomings of the outcome measurement to assess the complex process of animal-assisted education [34].
Provides a visual aid to describe the intervention or outcome assessments	Includes photographs or other formats of visual aid to support the reader in easily understanding the intervention.	Authors included photographs of how dogs were confined, how cages were disinfected, how people were educated using virtual reality headsets to control cystic echinococcosis [23]; and photographs of the laboratory conditions for rabbits, including an enriched floor pen (intervention) and a steel laboratory cage (control) [35].

Reporting outcomes

Studies that included a statistical analysis often conducted multiple types of tests. Of these, 36% (54/151) conducted categorical tests (Chi-Square, Fisher’s Exact, McNemar, and Mann-Whitney U Tests), 59% (89/151) conducted numeric tests (T-test, ANOVA, Kruskal-Wallis, and Wilcoxon signed rank tests), and 33% (50/151) conducted regression analyses (e.g., logistic (n = 22), linear (n = 17), Poisson (n = 13), negative binomial (n = 4), and Cox (n = 1) regressions). There were 13 studies that used a mixed-effects model for their regression, accounting for hierarchically clustered data or repeated measures over time. Two studies used imputation to address missing data.

Effectiveness of the interventions

While some studies demonstrated that One Health interventions were more effective than their comparison groups or control arms, others did not. For example, eight RCTs conducted in the field (not under controlled laboratory conditions) out of 35 total RCTs showed the effectiveness of their interventions, demonstrating a reduction in disease occurrence or pathogen detection, as described in depth in Table 8. Among 62 studies that had a control or comparison group, 35 (56%) showed effectiveness of the treatment groups compared to the control or comparison group, 18 (29%) reported mixed results depending on the different outcomes measured, 5 (8%) were comparing the effectiveness before and after the intervention among different types of treatment groups but not between each other, 3 (5%) reported no difference in effectiveness, and one study (2%) reported inconclusive results. For example, in an intervention that aimed to assess the mutual effects of a combined weight loss program for humans and dogs vs. humans only in the Netherlands, no significant difference in weight loss outcomes was reported between the intervention and control arms [36].

Table 8.

Summarized results of eight field-based randomized controlled trials of One Health interventions identified in the scoping review, ordered by year of publication

Title	Country, year of publication	Summary of findings
Community-based control of Aedes aegypti by adoption of eco-health methods in Chennai City, India	India, 2013	A cluster randomized controlled trial (cRCT) involved actor groups to assess the needs of the community and groups currently involved to limit Ae. aegypti and used this information to further support these groups as well as further involve the community. The authors found a significant decrease in Ae. aegypti indicating that community involvement and integration can create a successful and sustainable impact on disease control [44].
Entomological impact and social participation in dengue control: a cluster randomized trial in Fortaleza, Brazil.	Brazil, 2015	A cRCT in Brazil assessed the effectiveness of Aedes aegypti mosquito control strategies as a surrogate control for dengue by utilizing community partnership, holding workshops, and providing supplies to cover water containers. These investigators saw a decrease in mosquito infestation between the intervention and control clusters; they also measured participation of community members in each cluster and assessed why clusters differed from each other within the intervention [43].
Geographically targeted interventions versus mass drug administration to control Taenia solium cysticercosis, Peru	Peru, 2021	In a cRCT aimed at lowering the seroincidence of Taenia solium cysticercosis in pigs, investigators compared the effectiveness of three control strategies, each with and without pig treatment: (1) treating T. solium positive humans around T. solium positive pigs; (2) treating all humans around T. solium positive pigs; and (3) treating all humans regardless of pig positivity [41]. All trial arms successfully reduced T. solium seroincidence in pigs during the two-year intervention [41].
Entomological impact of mass administration of ivermectin and dihydroartemisinin-piperaquine in The Gambia: a cluster-randomized controlled trial	The Gambia, 2022	A cRCT assessed the impact of a three-month mass drug administration campaign of ivermectin and dihydroartemisinin-piperaquine to reduce the human reservoir of infection for malaria, found that the density of mosquitoes, including Anopheles gambiae, and the entomological inoculation rate decreased in the intervention arm compared to the control arm [45].
First bovine vaccine to prevent human schistosomiasis - a cluster randomised Phase 3 clinical trial	The Philippines, 2023	In a cRCT with six trial arms aimed at lowering the human incidence of schistosomiasis in the Philippines, an integrated intervention involving cattle vaccination, human mass praziquantel administration, and snail mass niclosamide administration (molluscicide to control the intermediate host) was conducted. The trial arm with bovine vaccination combined with snail molluscicide resulted in a lower relative risk of 0.69 (95% CI 0.61–0.78) in human schistosomiasis incidence compared to the control arm [40].
A planetary health innovation for disease, food and water challenges in Africa	Senegal, 2023	A cRCT in Senegal assessed the effectiveness of the removal of invasive submerged vegetation on schistosoma prevalence and water quality. They found that schoolchildren at intervention sites had lower infection rates of Schistosoma mansoni than at control sites [21].
Amplification of avian influenza virus circulation along poultry marketing chains in Bangladesh: A controlled field experiment	Bangladesh, 2024	An RCT aimed to assess how the transport conditions of backyard chickens and broilers affect the shedding of avian influenza virus (AIV) from farms to live bird markets in Bangladesh between biosecure (i.e., separating chicken batches and cleaning and disinfecting surfaces) vs. conventional transport environments. Investigators found that chickens transported in the biosecure intervention arm had lower odds of AIV detected in oropharyngeal swabs at 12 h after the transport compared to the less biosecure control arm [42].
Integrated Eco-Health approach significantly reduces helminth infections in endemic Khong islands with emphasis on Schistosoma mekongi	Lao People’s Democratic Republic, 2024	In a stepped-wedge design RCT, investigators assessed the effectiveness of mass drug administration in humans and dogs, two health education campaigns, and latrine construction to control Schistosomiasis. They found that the prevalence of Schistosoma mekongi in humans and dogs decreased after the multi-pronged interventions were implemented [18].

Among 89 studies that did not have a control or comparison group, 80 studies (90%) reported that the intervention was effective, five studies (6%) reported mixed results depending on the different outcomes measured, two studies (2%) reported the intervention to be ineffective, and two studies (2%) were exploratory qualitative studies for which effectiveness was not applicable. Some studies reported that the One Health intervention was effective, but the two analysts could not fully assess it because there were no control arms or comparisons groups, and limited considerations for threats to internal validity in experimental studies, including history (events extraneous to the intervention), maturation (natural changes within a participant), testing (testing to measure outcomes itself changes the participant response), instrumentation (change may occur in the measurement tool), and regression (extreme outcomes regress toward the mean the next time they are measured) [37]. For example, a quasi-experimental study assessed the effectiveness of a 14-year EcoHealth intervention to control Chagas disease by decreasing Trypanosoma cruzi transmission among vectors in Argentina. The comprehensive intervention included education, surveillance, insecticide spraying, improving animal pens, building wells and handwashing sinks, and improving roofs, which reduced the infestation of Triatomine bugs inside the house from 18% to 0.2%, and near the house from 20% to 3% [38]. While the reduction in Triatomine bugs may be plausibly associated with the intervention, it is challenging to discern whether the community was affected by extraneous changes during the long intervention.

Many papers that conducted educational interventions only reported an improvement in knowledge, practices, or behavior. However, the common shortcomings of these educational interventions were that most measured knowledge immediately after the intervention without a longer-term follow-up, knowledge was self-reported without further triangulation, and the courses were commonly elective courses, making the study population a selective group of students who were already interested in One Health and planetary health. For example, there were 35 (35/151, 23%) studies that measured self-reported knowledge or perception immediately after an educational intervention without a control or comparison arm, and that did not have a longitudinal follow-up. An additional four studies (4/151, 3%) measured outcomes of an educational intervention that were not self-reported instead of, or in addition to, self-reported knowledge. The outcome measures included findings from physical exams of livestock, microbial contamination of milk, the quantity of rejected milk, bacterial isolates, and the quantity of personal protective equipment purchased.

Although most studies did not assess the effect of the intervention longitudinally, three (3/151, 2%) reported that the intervention was effective at first, but the effect of the intervention trended back to baseline at follow-up. For example, an intervention involving human education, cleaning of public spaces, and rodent trapping aimed to reduce rodent infestation in Argentina showed that rodent activity decreased from 35% to 16% but increased to 22% at 90 days after the intervention [39].

There were eight high-quality, field-based RCTs. These studies were able to adjust for known and unknown confounders by randomly assigning the intervention, offering insights into how an intervention will work in a field setting. We present the findings from these papers in Table 8.

Discussion

Our scoping review of One Health interventions with data extracted from 151 studies, conducted in 53 countries, and published from 2005 to 2025, addressed a diverse range of health challenges. This scoping review contributes to the literature in its size and comprehensive approach to reviewing One Health interventions to date. The number of studies that were published increased substantially during the two decades evaluated. We build on prior efforts, including a literature review that assessed the effectiveness of One Health case studies and frameworks, which was published in 2017 [6], a scoping review that quantified outcomes of One Health approaches applied to global health, which was published in 2018 [7], and a scoping review that characterized the interface between One Health and global health, which was published in 2023 [14]. As expected, “zoonoses and emerging/re-emerging infectious diseases” was the most frequently represented study theme. However, the specific diseases represented did not include some common zoonotic diseases such as toxoplasmosis, foodborne trematodiasis, viral hemorrhagic fevers (e.g., Ebola and Marburg virus diseases), and Lassa fever, all of which are mentioned in the World Health Organization’s Research Priorities for Zoonoses and Marginalized Infections [46]. Among the human, animal, and environmental domains, the environmental domain was least well represented, as previously reported [47, 48]. However, to our surprise, environmental topics such as contamination with lead and mercury, pesticide exposure, polychlorinated biphenyls, waste management, and the built environment were represented in some papers.

The effectiveness of the One Health interventions was mixed and challenging to assess. Quantifying the number of studies that demonstrated the effectiveness of One Health interventions vs. those that did not was challenging, because the interventions included in the 151 papers were diverse, often measured multiple outcomes, including mixed results, or did not include a comparison or control group. Attempting to do so would have reduced the complexity of the studies and omitted the nuances to interpret the findings. However, eight field-based RCTs clearly demonstrated that a One Health intervention lowered the occurrence of disease or pathogen detection, offering insight into the effectiveness of One Health interventions when they are conducted well, even in a non-laboratory setting.

RCTs are the gold standard, but other study designs also hold merit. For example, some societal contexts may not lend themselves to setting up a rigorous RCT. For example, Sanders et al. reported a situation in Mali where an intervention to eliminate guinea worm was carried out amid government mistrust and violent conflict and provided insight on how security challenges can be incorporated into pragmatic One Health interventions [49]. Although a quasi-experimental design that reported the outcomes of a program, this paper was useful to fill a gap in the literature on implementing programs at the intersection of security and disease control.

The hypothesis at study conception was that limited evidence existed to examine the comparative effectiveness of One Health interventions. To our surprise, we found 151 studies that fit our inclusion criteria. This may be due to our comprehensive search strategy that included papers that did not explicitly use the terms “One Health”, “Planetary Health”, or “EcoHealth” in the title or abstract. In fact, almost half of the papers identified did not use any of these terms. For future literature searches, it would be important not to exclude papers that are not explicitly identified as One Health, Planetary Health, or EcoHealth, because it would lead to an incomplete representation of relevant literature. The tradeoff for conducting a comprehensive One Health literature search is resource availability. This comprehensive scoping review was arduous, requiring many hours of scanning, filtering, and in-depth discussions between two analysts.

A large proportion of the papers that were considered to match our One Health descriptor of at least 2.25 on our semi-quantitative scale were studies that conducted an education intervention that solely measured self-reported knowledge, attitudes, and practices (referred to here as education/KAP papers). These mostly involved the ideas or concepts of animals and the environment, but did not directly target them in the intervention nor measure the outcomes of animals or the environment these domains of One Health. This is shown in Fig. 4 in which the majority of papers in the 1.25 and 2.25 scores were education/KAP papers. Our results are likely affected by the number of education/KAP papers that met our criteria. Without education/KAP papers, we had 4 studies conducted in the USA, 5 in India, and 6 in Tanzania. We also see that the number of papers that involve only the human and animal domains is reduced from 34 to 26, and those that involve only the human and environment domains were reduced from 11 to 3.

Education and knowledge transfer are necessary but insufficient components of behavior change [50, 51]. We attempted to assess whether education/KAP papers were effective at improving knowledge or changing behaviors. While many of the papers stated a significant improvement in KAP, we often found that there was more nuance in the results. There may be an increase in knowledge scores, but not behavior change [52]. The control group also changed with the experimental group, which may have been due to the transfer of education [53]. One study achieved a significant result depending on the type of model run and whether interactions between variables were included [54]. It is not uncommon to run multiple models during an analysis, and the authors were transparent in their results; however, it does add nuance as to whether interventions are effective or not. Due to this nuance, we cannot confidently conclude that education/KAP interventions are or are not effective.

Overstating the significance of results has become a point of scrutiny in research. While we did not directly assess how many papers overstated their findings, we documented this practice in some of the manuscripts. This observation also aligns with other literature that found a majority of papers have a propensity to overstate findings or have positive results, including systematic reviews [55, 56]. We recommend that future studies always include limitations in a manuscript for transparency and self-critique, which will benefit the authors, reviewers, readers, and future researchers [57]. We hope that these recommendations will help advance One Health to an approach that is backed by rigorous studies and evidence, and to further the credibility of the One Health approach within peer-reviewed literature as a legitimate interdisciplinary research approach to create evidence-based policies and interventions [58]. We also recommend that authors include barriers to intervention implementation, as this may assist in implementing interventions in other areas. Having a comparison or control groups to assess the effectiveness of an intervention is also critical to improving the quality and evidence of the research being conducted. We also recognize that many studies that use secondary data, such as surveillance data, may not have the funds or ability to have a control or comparison. For this, we recommend using a hypothetical counterfactual akin to Larkins, et al. 2020 [28]. Although hypothetical counterfactuals would rely on certain assumptions made by authors, we believe it could provide more evidence and information to assess the efficacy of an intervention. If setting a separate control group is logistically challenging, investigators may consider analyzing participants categorized by the “dose” of exposure to the intervention. Reporting of interventions could also be strengthened in the analysis stage using correlation or regression analyses.

In addition to setting a comparison or a control, we identified exemplary practices that are presented in Table 4, which we believe will strengthen the study rigor of One Health intervention studies. These include designing a comprehensive intervention, using a participatory approach, conducting comprehensive outcome measurements, conducting longer-term follow-ups after the intervention, triangulating results using more than one method, answering the research question succinctly, writing clearly, and not overstating findings, being transparent about study limitations, and providing visual aids to describe the intervention.

Within this review, we also wanted to assess how anthropocentric One Health literature currently is. As seen in Fig. 5 and 90% of eligible papers included a human component, and 79% included an animal component. In comparison, only 70% of eligible papers included an environmental component. This may be due to One Health largely researching zoonotic diseases, as found in our review, in which humans are the predominant interest with animals as the source. These findings match with Meisner, et al.’s critique that the field has taken the role of a reductionist approach to addressing biomedical topics which predominantly focus on human health [59]. It may also be attributed to the difficulty of defining what an environment is and how it contributes to the overall health of all three One Health domains, because they may be upstream drivers of disease. It is also more difficult to obtain environmental and animal data. This is likely why so many education/KAP papers – including ones scored as a 2.25 – involved the concept of animals and the environment rather than using these domains as active participants. This is likely due to education/KAP research being relatively cost-effective and requiring less effort to collect data reliably compared to field studies. Another possibility for some studies taking a relatively anthropocentric focus is the donor’s priority, but this needs to be investigated further. From high-scoring One Health papers, we appreciated how difficult it may be to include all One Health components of all three domains in a succinct and clear way for a journal article and its readers to gather pertinent information. This can often be at odds when describing complex and interconnected research, such as One Health research.

There are notable strengths to this research. First, as stated above, we used broad search terms not only restricted to “One Health”, including adjacency searches, and four databases that do not solely focus on medical literature. Second, all papers were screened, assessed, and extracted by two independent researchers (ET and LH). The double screening method was important because analysts had different training and backgrounds and therefore brought unique perspectives and decisions about papers. Third, although not always required for a scoping review [60], we conducted a quality assessment on all papers, including perceived biases, reproducibility, generalizability, and a critical assessment of whether the outcome could be attributed to the intervention. We also identified and provided examples of exemplary practices that future researchers could consider incorporating into their study design and reporting.

Given the topics and types of papers ranged from basic, highly controlled experiments to large-scale surveillance analysis, reproducibility and generalizability were difficult to assess. We found that most of these papers were well described, with the ability to repeat the study and intervention. Assessing reproducibility sheds light on the importance of including detailed descriptions not only on the study design but also on the intervention itself. Regarding generalizability, most were focused on providing interventions that were tailored to a specific region or community. This, in turn, reduced the number of papers we scored as “widely generalizable” and would likely not yield the same outcomes if the intervention were to be conducted in different settings.

This study has several limitations. First, defining the boundaries of One Health and defining the scope of what constitutes an intervention were challenging. Iterative discussions and boundary-setting were required between two researchers to agree on whether a paper fits within the scope of our search, which involved judgment calls. If two different investigators had reproduced this scoping review, they may have ended with a slightly different collection of manuscripts. This challenge in the subjectivity of a literature review has been described previously [61]. To address this point, we used peer-reviewed definitions of One Health and One Health interventions, iteratively clarified definitions, documented decision points, and intentionally practiced reflexivity and possible biases in decision-making. Additionally, the quality assessment that we conducted was somewhat subjective, based on two analysts’ professional judgment. However, this was the result of following recommendations for scoping reviews, which advise against systematically conducting risk of bias assessments, because the primary objective is to map the available evidence [60]. In the future, established Risk of Bias frameworks [62] for randomized trials and non-randomized studies could be used to assess biases. Additionally, we did not separate plant health from environmental health, while recognizing that the new definition of One Health published by the One Health High-Level Expert Panel in 2022 distinguished plant health from environmental health [11]. This was a pragmatic decision to focus our already-broad literature search criteria. Future studies could focus on addressing how plant health is represented in One Health literature. Finally, we required that the One Health dimensions be included in a single manuscript. There may be studies that took a One Health approach but reported findings in different manuscripts, and some of these may not have been included based on the search strategy.

Conclusion

We identified a diverse set of One Health interventions in this scoping review. A small number of field-based RCTs clearly demonstrated the comparative effectiveness of One Health interventions, while it was challenging to assess the effectiveness of other studies without a more granular definition of effectiveness. Continuing to build evidence based on rigorous studies that assess the effectiveness of One Health interventions for improving health outcomes and the cost-benefits compared to siloed approaches will further improve buy-in of the One Health approach from governments, researchers, and policymakers, and advance the evidence base for employing the One Health approach.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

ET and MFD conceived of the study concept. LR advised and oversaw the formulation and execution of the search strategy and data management. ET and LH conducted all steps of the manuscript screening, extraction, and discussions around study consensus. ET conducted the analysis. ET and LH wrote the first draft of the manuscript. MFD oversaw the direction of the study. All authors read and approved the final manuscript.

Funding

ET was supported by a grant from the U.S. Centers for Disease Control and Prevention, National Institute for Occupational Safety and Health to the Johns Hopkins Education and Research Center for Occupational Safety and Health (award number T42 OH0008428).

Data availability

The dataset generated during the current study is included in the supplemental material. A detailed version of the dataset is available from the corresponding author upon request.

Declarations

Ethics approval and consent to participate

Ethics approval was not required for this study. The secondary data generated consisted of publicly available, peer-reviewed publications. There were no primary participants in this study.

Consent for publication

Consent for publication was obtained from all authors.

Competing interests

Eri Togami is an Associate Editor for One Health Outlook. The authors declare they have no other competing interests.