One Health Surveillance: Integrating Human, Animal, and Environmental Data

By Dr. Zubair Khalid, DVM, MS, PhD · 2026-08-11

Key Takeaways

• One Health surveillance necessitates the systematic integration of data from human, animal, and environmental sectors to detect and respond to emerging health threats, particularly zoonotic pathogens which constitute approximately 75% of such diseases.
• Whole genome sequencing (WGS) is a critical technical tool, enabling nucleotide-level comparison of isolates from human, animal, and environmental sources to accurately attribute outbreaks and reconstruct transmission pathways for foodborne and zoonotic pathogens.
• Structural and legal barriers, including siloed governance, data-sharing restrictions, and unclear inter-agency responsibilities, significantly impede effective cross-sectoral collaboration and data integration.
• Climate change is a significant environmental driver, altering pathogen and vector distributions and necessitating expanded geographic scope and adaptive strategies within surveillance systems, particularly for tick-borne and mosquito-borne diseases.
• Common failure modes include silent fragmentation of data streams, data incompatibility due to disparate identifiers and formats, and signal dilution from over-aggregation, all of which undermine the timely and accurate interpretation of integrated health information.
• Effective integrated surveillance requires a clearly defined surveillance question, a robust data integration architecture with shared case definitions and common data dictionaries, and strong political will supported by formal inter-sectoral agreements.

One Health surveillance is the systematic collection, validation, analysis, and interpretation of data from human, animal, and environmental domains, followed by the dissemination of that information to inform decisions for more effective health interventions. The concept responds to a practical problem: pathogens do not respect administrative boundaries between health sectors, and hazards that emerge in wildlife, livestock, or ecosystems frequently reach human populations through routes that no single discipline monitors completely. For the veterinary researcher, this framework reframes familiar surveillance questions. Instead of asking only whether a pathogen is circulating in a target animal population, the One Health approach asks how information from companion animal clinics, production medicine, wildlife ecology, public health laboratories, and environmental monitoring can be assembled into a coherent picture of transmission risk.

This article examines the conceptual foundations, operational design, and practical implementation of integrated surveillance systems. It addresses the veterinary researcher who needs to design, evaluate, or participate in cross-sectoral surveillance programs, and it answers a specific question: what distinguishes a genuinely integrated system from a collection of parallel surveillance activities that merely share a name? The discussion covers the scientific logic of integration, the structural and legal barriers that impede it, the role of molecular epidemiology and genomic methods, and the emerging influence of climate-driven changes on surveillance priorities. Throughout, the emphasis is on decision criteria, named failure modes, and published evidence instead of general advocacy.

At a Glance

Defining Integrated Surveillance

The term One Health surveillance describes more than the sum of its parts. A working definition, articulated at the Second International Conference on Animal Health Surveillance, holds that it is the systematic collection, validation, analysis, and interpretation of data on humans, animals, and the environment, with dissemination of findings to support evidence-based health decisions. This definition carries a specific operational implication: integration is not achieved by running three surveillance streams in parallel and sharing reports at annual meetings. It requires deliberate design of data flows, case definitions, and decision pathways that connect sectors.

The published discussion from the International Conference on Animal Health Surveillance identified that early success stories concentrated on the obvious interfaces between human and veterinary medicine, particularly zoonoses and food safety. These are the domains where the value proposition is easiest to demonstrate, because the transmission pathway is direct and the surveillance question is clear. The same source noted persistent challenges: the public health sector and the environmental sector have been engaged less fully than veterinary medicine in integrated activities, and legal issues, data-sharing hurdles, unclear responsibilities, and structural barriers between ministries continue to prevent coordinated action.

The Scientific Logic of Integration

The case for integrated surveillance rests on transmission ecology. Pathogens commonly acquired from food are not always transmitted by that route. They may spread through contact with animals, other humans, or environmental reservoirs, and outbreaks are frequently traced to food contaminated from these non-food sources. A surveillance system that monitors only human clinical cases will detect outbreaks late, after substantial exposure has occurred. A system that monitors only animal health will miss the human consequences that justify public investment. The review of whole genome sequencing applications in foodborne disease surveillance makes this point directly: presumed foodborne outbreaks are best investigated through a One Health approach that works across human, animal, and environmental sectors.

The epidemiological logic extends to detection timing. Because zoonotic pathogens amplify in animal or environmental reservoirs before spilling over into human populations, monitoring those reservoirs offers an earlier warning than waiting for human cases. This is the temporal advantage that justifies the added complexity of cross-sectoral data collection. The advantage is conditional, however. It depends on the surveillance system having sufficient sensitivity in the animal or environmental compartment to detect changes in pathogen prevalence before human exposure occurs, and on the data being shared rapidly enough to trigger preventive action.

The Role of Whole Genome Sequencing

Molecular epidemiology has changed what integrated surveillance can achieve. Whole genome sequencing has become the standard for molecular surveillance of foodborne pathogens, and it provides a common language across sectors. A veterinary isolate, a human clinical case, and an environmental sample can be compared at nucleotide resolution, and the relatedness of those isolates can be interpreted within an epidemiological framework. The review of WGS applications in One Health surveillance notes a specific technical challenge: outbreak strains that have propagated and evolved in non-human sources often show more sequence variation than typical monoclonal point-source outbreaks. This variation complicates cluster detection, because the genetic distance between isolates from a common source may exceed the thresholds used for point-source outbreak investigation. Flexible case definitions and integration of epidemiological data are required to interpret genomic findings correctly.

Structural Barriers to Integration

The obstacles to integrated surveillance are predominantly organizational instead of technical. A qualitative study of Australian academic experts identified four recurring themes. First, the absence of a clear definition and shared vision for One Health acts as a barrier to interdisciplinary collaboration. Second, siloed approaches by different sectors restrict the ability of professionals to work collaboratively. Third, an understanding of disease transmission across species and environmental compartments is a necessary requirement for successful integration. Fourth, political will is essential for the structural changes that integration requires.

These findings align with the technical and organizational challenges identified at the international conference level. The conference discussion on One Health surveillance noted that policy makers in the health sector often perceive One Health as a veterinary-driven initiative, a perception that undermines the cross-sectoral ownership needed for sustained integration. Legal constraints on data sharing and unclear responsibilities between ministries compound the problem. The practical implication for veterinary researchers is that technical competence in surveillance methods is necessary but insufficient. Successful integration requires engagement with governance structures, data-sharing agreements, and the political processes that determine whether cross-sectoral systems receive sustained funding.

Environmental Dimensions and Climate Change

The environmental component of One Health surveillance has historically been the weakest of the three domains. The systematic literature review of One Health surveillance system characteriztics found that confusion and uncertainty regarding practical application, outcomes, and impacts prevail, partly because a conceptual and methodological framework for defining the characteriztics of integrated systems has been lacking. This gap is particularly visible in the environmental sector, where the link between ecosystem monitoring data and health outcomes is often indirect and the relevant data sources are fragmented across agencies responsible for water, soil, wildlife, and climate.

Climate change intensifies the need for environmental surveillance. Changes in temperature and precipitation alter the conditions for pathogens and vectors of zoonotic diseases, and the review of climate change and zoonoses documents the increasing spread of West Nile and Usutu viruses and the establishment of new vector species, such as specific mosquito and tick species, in Europe and other regions. These shifts create new challenges for maintaining human and animal health, and they expand the geographic areas where surveillance must operate. The review of tick-borne disease geography found that spatial approaches can synthesize data generated by integrated surveillance systems, but that data on the enzootic cycle of tick-borne pathogens is severely underutilized and mapping efforts are mostly limited to Europe and North America. The same review suggests that available methods can be applied to track tick-borne disease distributions in Africa and Asia, combining medical and veterinary surveillance for maximum impact.

Operational Design of Integrated Surveillance Systems

Defining the Surveillance Question and Scope

The first decision in designing an integrated system is specifying what the surveillance is for and which sectors must contribute data to answer it. A system aimed at early detection of an exotic zoonosis in livestock differs structurally from one tracking endemic foodborne pathogens across a national food chain. The surveillance objective determines the data types, the sampling frame, the frequency of collection, and the legal authority for data sharing.

For a zoonotic disease with an environmental reservoir, the system must include environmental sampling points. For a pathogen transmitted primarily through food, the human clinical and veterinary diagnostic streams are the core, with environmental data used for traceback. The characteriztics of One Health surveillance systems described in the systematic literature review by Bordier and colleagues emphasize that the organizational structure must follow the hazard, the objective, and the context. A system designed for a wildlife-livestock interface pathogen such as brucellosis requires different partners than one designed for a foodborne pathogen such as Salmonella.

The scope decision also includes the geographic and administrative boundaries. Administrative borders rarely align with ecological or epidemiological ones. A system that stops at a national border misses the movement of wildlife reservoirs, vectors, and traded animals. The challenges of implementing an integrated One Health surveillance system in Australia identified that separate sectors with limited communication and the absence of a shared vision were primary barriers. Defining the question jointly, with all sectors at the table from the start, is the mechanism that prevents this fragmentation.

Data Integration Architecture

Integrated surveillance requires a data architecture that allows records from different sectors to be linked without compromising the integrity of any single source. The minimum components are a shared case definition, a common data dictionary, and a unique identifier that can be traced across human, animal, and environmental records.

The case definition is the most consequential element. A human clinical case of a zoonosis may present with non-specific symptoms, while the animal case may be detected through routine slaughter surveillance or abortion investigation. The CDC principles of epidemiology in public health practice provide the standard framework for case definitions, including clinical, laboratory, and epidemiologic criteria. In an integrated system, the case definition must specify which combinations of criteria from which sectors constitute a reportable event. For example, a single human case of a pathogen with a known animal reservoir may trigger an animal investigation even if no animal cases have yet been detected.

The data dictionary must reconcile different vocabularies. Human medicine uses ICD codes, veterinary medicine uses species-specific diagnostic codes, and environmental monitoring uses physical and chemical parameters. The integration layer maps these to a common ontology. This mapping is labor-intensive and requires ongoing maintenance as codes change.

The unique identifier is the most sensitive component. Human medical records are protected by privacy law in most jurisdictions, and animal records may be commercially sensitive in production systems. The system must use a pseudonymised identifier that allows linkage for epidemiological purposes without revealing personal or commercial identity. The WOAH animal health surveillance standards provide the international framework for animal health data reporting, and these standards should be consulted when designing the animal-side data fields.

Selecting Data Sources and Sampling Strategies

The choice of data sources determines the sensitivity and timeliness of the system. Passive surveillance, based on routine diagnostic submissions and clinical reporting, is inexpensive but subject to under-reporting and bias. Active surveillance, based on scheduled sampling, is more sensitive but more costly. Integrated systems typically combine both, using passive surveillance for ongoing monitoring and active surveillance for specific high-risk populations or during outbreak response.

The table below summarizes the main data source categories and their selection criteria.

The whole genome sequencing approach to One Health surveillance of foodborne diseases described by Gerner-Smidt and colleagues illustrates how genomic data changes the sampling strategy. Outbreak strains that have evolved in non-human sources show more sequence variation than typical point-source outbreak strains. This means that the sampling frame must include potential animal hosts and environmental sources, also clinical cases, and that the genomic case definition must be flexible enough to accommodate this variation.

Monitoring Parameters and Signal Interpretation

Each data stream generates signals that must be interpreted in context. The monitoring parameters differ by data type.

For human and animal clinical data, the parameters are incidence, case fatality, and the proportion of cases with a confirmed laboratory diagnosis. A rise in incidence in one sector without a corresponding rise in the other may indicate a change in reporting, a change in diagnostic testing, or a genuine epidemiological shift. The interpretation requires knowledge of the baseline and the factors that influence each stream independently.

For environmental data, the parameters depend on the hazard. For vector-borne diseases, the relevant parameters are vector abundance, vector infection rate, and the proportion of competent vectors in the population. The trends and opportunities in tick-borne disease geography review by Lippi and colleagues found that data on the enzootic cycle of tick-borne pathogens is severely underutilised in mapping efforts. Vector infection rates can provide an early warning that precedes human and animal cases by weeks, but only if the sampling design captures the relevant tick populations.

For genomic data, the parameters are cluster detection, single nucleotide polymorphism distance, and the identification of genetic markers associated with virulence or antimicrobial resistance. The interpretation of genomic clusters requires integration with epidemiological data. A genomic cluster without epidemiological linkage may represent an undetected common source, while an epidemiological cluster without genomic support may represent a misdiagnosis or a common exposure to multiple strains.

Documentation and Governance

The documentation requirements for an integrated system exceed those of a single-sector system. Each data contribution must be traceable to its source, with metadata describing the collection method, the laboratory method, and the quality control procedures. The data quality assurance standards for veterinary surveillance systems provide a framework for assessing completeness, validity, and timeliness, and these principles apply equally to human and environmental data.

Governance structures must define who can access which data, who interprets the integrated signals, and who authorises a public health response. The WOAH terrestrial animal health code sets out the international obligations for notification of animal diseases, and these obligations may conflict with the desire to delay reporting until an integrated picture is complete. The governance framework must resolve this tension by specifying the conditions under which a single-sector signal triggers immediate action and the conditions under which action awaits integrated confirmation.

The governance structure also determines how the system adapts. Surveillance systems require periodic review of their performance against their stated objectives. The review should assess sensitivity, timeliness, and cost, and should include an explicit assessment of whether the integration adds value over the sum of the single-sector systems. Where integration does not add value, the system should be simplified.

Recognized Complications and Failure Modes

Integrated surveillance systems fail in characteriztic patterns. The most common is silent fragmentation, where partner agencies nominally participate but continue to operate their own databases, case definitions, and reporting timelines. Detection depends on auditing data flow at each interface. A veterinary laboratory that cannot confirm whether its submissions reached the human health authority within the agreed window indicates a broken handoff, not a technical glitch.

Data incompatibility represents a second failure mode. Human clinical records, livestock movement databases, and environmental sampling logs rarely share identifiers, spatial resolution, or temporal granularity. When a system cannot link a farm-level outbreak to a regional human case cluster, the integration has failed functionally even if all partners report successfully. Early detection requires routine linkage testing using historical outbreaks as benchmarks.

A third failure is signal dilution. Integrated systems that aggregate data across sectors often bury weak but genuine signals beneath unrelated background variation. The discriminating check is whether the system retains the ability to detect a single-species event, such as an unusual mortality cluster in wildlife, with the same sensitivity as before integration. If sensitivity drops, the aggregation rules require revision.

Governance failure manifests as unresolved disputes over data ownership, publication rights, and decision authority. These disputes typically surface during the first cross-sectoral outbreak response. The operational test is whether the system can issue a joint risk assessment within the required timeframe without seeking ad hoc legal clearance. Systems that cannot do so have not achieved integration regardless of their formal agreements.

Common Errors in Implementation

Less experienced teams frequently confuse data sharing with integration. Sharing spreadsheets between agencies does not create a surveillance system unless the data are jointly analyzed against shared objectives. The corrective action is to define a single analytical question that requires all three sectors, then design the data flows around that question.

A second recurring error is designing the system around available data instead of the surveillance objective. Teams adopt existing laboratory submissions or passive reporting streams because they are accessible, then discover the data cannot answer the defined question. The corrective step is to specify the minimum data elements required for the decision, then negotiate access or collection accordingly.

A third error involves case definition mismatch. Human health authorities may define a probable case clinically, while veterinary authorities require laboratory confirmation. The resulting counts are not comparable, and trend analysis becomes meaningless. Teams should harmonise case definitions across sectors before launch, using the most specific definition that all partners can support operationally.

Finally, teams underestimate the effort required for ongoing data quality assurance. Integrated systems multiply the points where errors can enter, and a single misclassified record can propagate across sectors. Routine validation against source records, automated range checks, and periodic inter-laboratory comparisons are necessary controls, not optional refinements.

Limitations of the Current Evidence

The evidence base for One Health surveillance remains uneven. Systematic review work has identified the organizational and functional characteriztics of integrated systems, but the literature contains few controlled comparisons of integrated versus conventional surveillance performance One Health surveillance systems systematic review. Claims of improved cost-effectiveness rest largely on plausibility instead of measured outcomes.

Expert opinion diverges on several points. Some authorities argue that integration should begin with zoonotic pathogens of proven epidemic potential, while others advocate building broad platforms that can accommodate unknown future threats. The former position prioritizes demonstrable value, the latter emphasizes preparedness. Both positions have merit, and the choice depends on local governance capacity and political will, which qualitative research identifies as decisive factors in implementation success integrated One Health surveillance challenges in Australia.

The role of environmental data remains contested. While climate change clearly alters pathogen and vector distributions climate change and zoonoses review, environmental surveillance data are often the least standardized and most difficult to interpret. Whether routine environmental sampling adds sufficient predictive value to justify its cost is unresolved, and current guidance is to include environmental components only where a specific transmission pathway justifies them.

Escalation and Referral Criteria

Veterinary clinicians should escalate to public health authorities when a diagnosed animal disease has known zoonotic potential and human exposure has occurred. The threshold for reporting is defined by national legislation and international standards, and clinicians must know the requirements in their jurisdiction WOAH animal health surveillance standards. When in doubt, consultation with the relevant authority is always preferable to delayed reporting.

Specialist laboratory involvement is warranted when a cluster involves an unusual species, an atypical clinical presentation, or a pathogen with ambiguous identification. Reference laboratories provide confirmatory testing, typing, and antimicrobial susceptibility data that local facilities cannot generate. Whole genome sequencing has become the standard for molecular surveillance of foodborne pathogens, and its use requires laboratory capacity that most primary diagnostic facilities lack whole genome sequencing for foodborne disease surveillance.

Referral to an epidemiologist is appropriate when a suspected cluster cannot be confirmed or refuted using routine surveillance data. Epidemiologists can design the analytical studies needed to establish whether cases are linked, estimate the exposure window, and identify the source. This consultation should occur early, because delays in investigation reduce the probability of identifying the vehicle of infection.

Regulatory reporting obligations differ by species, production system, and region. Livestock producers and companion animal practitioners face different requirements, and clinicians should maintain current knowledge of the notification list applicable to their practice. International trade implications arise when reportable diseases are confirmed, and the relevant veterinary authority must be notified without delay to prevent wider dissemination WOAH terrestrial animal health code.

Frequently Asked Questions

How do we estimate the cost of an integrated surveillance system before committing resources?

Cost estimation should begin with a scoping exercise that defines the hazard, the decision the surveillance must inform, and the sectors that must contribute data. The systematic review of One Health surveillance system characteriztics identifies organizational and functional features that drive expense, including data governance mechanisms, cross-sectoral coordination bodies, and laboratory capacity. Budget for personnel time for inter-sectoral meetings, data standardization, and joint interpretation, also laboratory testing. Pilot a limited geographic or hazard scope first, then scale. Compare the integrated approach against the cost of parallel single-sector systems, including the cost of missed detections. Where formal economic analysis is unavailable, document assumptions explicitly and revisit them annually.

What is the minimum viable approach when full integration is not feasible?

Prioritize data sharing over structural integration. A functional minimum requires a shared case definition, a secure channel for exchanging laboratory results, and a named contact in each participating agency. The Australian implementation study found that siloed approaches and unclear definitions were primary barriers, so formalise even minimal agreements in writing. Use existing reporting channels where possible, such as notifiable disease lists, and add cross-sectoral flags instead of building new databases. Whole genome sequencing can be reserved for outbreak confirmation instead of routine surveillance, as described in the review of WGS for foodborne disease surveillance. A modest system that produces shared, interpretable data outperforms an elaborate system that only one sector maintains.

How should surveillance priorities differ between companion animal and production animal settings?

Companion animal surveillance should emphasize syndromic signals from clinical records, because individual animals present sporadically and laboratory confirmation is often delayed. Production animal systems can rely on structured population-level data, including mortality, feed intake, and slaughter checks, and can justify more intensive sampling. The WOAH animal health surveillance standards apply across species but are most directly operationalised in production settings with defined populations. Companion animal practitioners should contribute to integrated systems by reporting unusual clusters, also confirmed zoonoses, and by recording travel and exposure histories. Environmental sampling is more tractable in production systems with known cohorts. In both settings, the surveillance question determines the data source, not the species.

What records must a veterinary practice maintain to support integrated surveillance?

Maintain complete clinical records that include signalment, presenting signs, vaccination status, and a standardized exposure history covering travel, contact with wildlife, and other animals in the household. Record diagnostic test results with laboratory accession numbers and retain samples where storage permits. Document any communication with public health authorities, including the date, the person contacted, and the information exchanged. The CDC principles of epidemiology in public health practice describe the surveillance cycle of collection, analysis, and dissemination, and practice records should support each step. Use consistent terminology for clinical signs to enable syndromic aggregation. Privacy considerations apply to owner information, so separate clinical data from personal identifiers before sharing with surveillance systems.

How do I explain integrated surveillance to a client who is concerned about data sharing?

Explain that the purpose is to detect health threats earlier than any single sector could alone, and that the information shared is typically de-identified or aggregated. Use a concrete example, such as a cluster of diarrheal illness in animals and people linked to a shared water source, where integration allows a faster, more targeted response. The One Health surveillance concept paper describes how integrated data informs decisions for more effective interventions. Clarify that veterinary records remain subject to professional confidentiality and that only specified data elements are shared. Offer the client the option to ask questions about what is shared and with whom. Most clients accept the rationale when the benefit is framed as protecting both animal and human health in their community.

When should a veterinary practice escalate a finding to public health authorities?

Escalate when a case meets statutory notifiable disease criteria, when an unusual cluster of similar cases appears in animals or people, or when a zoonotic pathogen is confirmed in a species or region where it is not expected. The WOAH terrestrial animal health code defines international notification obligations, but local requirements vary, so know the relevant jurisdictional list. Escalate early instead of waiting for laboratory confirmation if the clinical presentation is strongly suggestive and the potential for human exposure exists. Document the clinical findings, the exposure history, and the samples submitted. If the first contact does not respond within a reasonable timeframe, escalate to the next level. Uncertainty about whether to report should favour reporting, with the receiving agency making the formal classification.

Related Clinical & Scientific Guides

• Evaluating Veterinary Surveillance System Attributes
• Network Analysis for Infectious Disease Spread in Animal Populations
• Randomized Controlled Trials in Veterinary Field Settings

References and Further Reading

• One Health surveillance - More than a buzz word?. 2015.
• Whole Genome Sequencing: Bridging One-Health Surveillance of Foodborne Diseases.. 2019.
• Characteriztics of One Health surveillance systems: A systematic literature review.. 2020.
• The challenges of implementing an integrated One Health surveillance system in Australia.. 2018.
• Climate Change and Zoonoses: A Review of Concepts, Definitions, and Bibliometrics.. 2022.
• Trends and Opportunities in Tick-Borne Disease Geography.. 2021.
• WOAH Animal Health Surveillance Standards. WOAH.
• CDC Principles of Epidemiology in Public Health Practice. CDC.
• MSD Veterinary Manual, Professional Edition. MSD Veterinary Manual.

Related Articles

• Risk-Based Surveillance in Animal Health
• Data Quality Assurance in Veterinary Surveillance Systems
• Syndromic Surveillance in Veterinary Practice
• Time Series Analysis for Veterinary Disease Surveillance
• Designing Case Definitions for Veterinary Surveillance

This article is educational professional reference material for veterinary audiences. It is not a substitute for veterinary diagnosis, individual clinical judgment, current product labeling, or applicable regulatory requirements.