Beyond the buzzword: The real-world challenges of data integration for One Health

As the interdependence among human, animal, and environmental health is increasingly recognized, One Health emerges as a critical framework for addressing complex health challenges at the human-animal-environment nexus. Effective implementation requires systems-based approaches capable of modelling the dynamics, relationships, and feedback loops within complex social-ecological systems to support prediction and decision-making, positioning data integration as a cornerstone of actionable One Health intelligence. However, relevant data streams remain fragmented across sectors, institutions, and platforms, with surveillance systems operating in silos that limit cross-sectoral analysis and forecasting. Despite growing investments, major barriers continue to limit One Health analytics. Data sharing, alongside the adoption of standardized metadata, remains a major bottleneck. The diversity of repositories and lack of data/metadata standards make dataset discovery and reuse inefficient, while semantic and technical incompatibilities further limit interoperability. For example, human health data uses standardized classifications such as ICD codes, which have no direct equivalent in animal health; more broadly, data across sectors vary widely in format, spatial resolution, and temporal granularity. Institutional fragmentation exacerbates these challenges, with veterinary services often lacking mandates for wildlife, and environmental health underrepresented within One Health initiatives. Beyond technical and institutional barriers, global inequities in data availability persist. High-income countries dominate data production, while animal and environmental surveillance systems remain under-resourced, particularly in low- and middle-income settings. Legal, ethical, and trust-related barriers, including data sovereignty concerns and fears of reputational damage, further impede data sharing. These converging challenges raise critical questions: how can we detect anomalies or “shocks” without comprehensive visibility across system components? And how can effective One Health interventions be designed if their cross-sectoral impacts cannot be adequately measured? Emerging initiatives demonstrate both the challenges of structuring heterogeneous data and the value of integrated analytics. However, translating these lessons into sustainable, scalable frameworks remains an open challenge, limited by insufficient funding and capacity constraints. Advancing FAIR data principles, ensuring equitable access to infrastructure, and promoting ethical data practices are essential, particularly in the context of responsible AI use. Moving beyond buzzword to operational reality, data integration requires rethinking incentives, strengthening cross-sectoral workforce capacity, and embedding equity within data governance. Ultimately, the goal is to enable integrated One Health intelligence across scales to support early detection, prevent spillover at source, and strengthen health system resilience.