Knowledge integration in One Health policy formulation, implementation and evaluation

Abstract The One Health concept covers the interrelationship between human, animal and environmental health and requires multistakeholder collaboration across many cultural, disciplinary, institutional and sectoral boundaries. Yet, the implementation of the One Health approach appears hampered by shortcomings in the global framework for health governance. Knowledge integration approaches, at all stages of policy development, could help to address these shortcomings. The identification of key objectives, the resolving of trade-offs and the creation of a common vision and a common direction can be supported by multicriteria analyses. Evidence-based decision-making and transformation of observations into narratives detailing how situations emerge and might unfold in the future can be achieved by systems thinking. Finally, transdisciplinary approaches can be used both to improve the effectiveness of existing systems and to develop novel networks for collective action. To strengthen One Health governance, we propose that knowledge integration becomes a key feature of all stages in the development of related policies. We suggest several ways in which such integration could be promoted.


Introduction
The concept of One Health initially arose from integrated research on zoonoses, 1,2 but now covers all of the interconnections between human, animal and environmental health. The concept is a collaborative, interdisciplinary and intersectoral multi-institutional approach, linking many different forms of knowledge and expertise. [3][4][5] One Health is represented by a complex biological and social system that involves multiple actors and processes and their interactions over time, at local, national and global levels. 5 To date, relatively little attention has been given to the epistemological, institutional, political and social factors associated with the implementation of a One Health approach. 2,6 This is illustrated by the almost complete lack of literature on One Health governance.
There is an existing framework for global health governance: a combination of the formal and informal institutions, rules and processes that influence global decisions on health policy. 7,8 Ideally, such a framework should transcend national boundaries, embrace multisectoral and interdisciplinary approaches and engage with the whole wide range of relevant actors. 9 In reality, however, the current framework is affected by fragmentation of health interests, programmes and sectors, a general lack of societal participation and by professional focus on very limited areas of expertise, so-called professional silos. 10,11 The dysfunctionality of the current framework, in terms of the core elements of the One Health concept, emphasizes the need for a dedicated framework for One Health governance. 12 It has been suggested that some of the current framework's shortcomings could be overcome by the development of coordinated supranational bodies, the promotion of specialized training and career opportunities and the creation of dedicated funding mechanisms. 9,[12][13][14] We suggest that the framework may also be strengthened by improving the integration of its management 4,7,15 and by integrating knowledge at all stages of any related policy development. [16][17][18] In 2012, knowledge integration was listed as one of the United States National Cancer Institute's key recommendations for improving 21st century epidemiology. 19 Since 2014, about 230 experts and representatives of governments and nongovernmental organizations, from the fields of environmental, public and veterinary health and associated sciences, have come together in the Network for Evaluation of One Health. 20 This network's main aim is to develop standards for assessing integration in One Health. Since 2016, this work has been enhanced by a core group of experts on complex systems, governance and knowledge integration. This paper summarizes the results of this group's investigation of knowledge integration in governance, as a mechanism for multi-institutional learning to improve the governance and coordination of One Health implementation in the absence of hierarchical chains of command.

Coordination and governance
In policy cycles, multiple rounds of agenda setting, policy formulation, decision-making, implementation and evaluation lead to the creation, implementation and revision of policies. 21 We believe that, in terms of the interdisciplinary, intersectoral and multi-institutional One Health approach, knowledge integration at every stage of policy development, in every policy cycle, could strengthen the coordination and governance of One Health implementation. Although some integration of knowledge from different disciplines, institutions and sectors can, and does, take place intuitively, in many circumstances, we believe that it needs to become a regular, routine and institutionalized process at project, programme and policy levels. 8,18,22 In the development of health policies, knowledge assessment is often confined to the last, that is evaluation, stage of each policy cycle. 23 We believe that, to op-timize the coordination and governance of the One Health approach, knowledge integration should be central at every stage of policy development.
In its broadest sense, knowledge integration has been defined as the building of shared and meaningful syntheses between distinct mental models, based on a recognition and explanation of the relevant differences between the models. 24,25 Rather than seeking consensus, knowledge integration can be used to build a common framework that allows an understanding of the links between the knowledge of multiple individuals. Such integration has been likened to the weaving of multiple perspectives into a central vision or a search for coherence and correspondence. 26,27 The fostering of effective knowledge integration in a policy cycle is a multidimensional challenge because it requires the integration of cognitive concepts, organizational and social interests and perspectives as well as communicative and cultural factors. The relevant literature distinguishes target knowledge from systems and transformation knowledge. Target, or normative, knowledge relates to objectives and interests, while systems, or descriptive, knowledge relates to perspectives on factual processes. Transformation, or prescriptive, knowledge relates to the transformation of the current version of a system towards a more desired version. 28 The integration of these three forms of knowledge throughout a policy cycle can be facilitated by three different approaches: multicriteria analyses for target knowledge, systems thinking for systems knowledge and transdisciplinary approaches for transformation knowledge.

Multicriteria analyses
The key to integrating target knowledge is to understand the often-conflicting interests, preferences and values of the multiple actors, as a first step to mediation, negotiation and, ultimately, collective action. 29 Multicriteria analyses can assist such integration because they elicit and structure value systems in a way that accommodates a multiplicity of information sources and types. 30 Such analyses can incorporate any objective that has relevance to the point of view under consideration, rely on nonmonetary units and apply valuation methods that are independent of pricing mechanisms. This makes these analyses particularly suited for priority setting in implementation of the One Health ap-proach, which typically involves equity, intergenerational justice and non-marketed goods. 31,32 When combined with systems analysis for strategic, long-term assessments, multicriteria analyses offer a flexible yet systematic method of valuation that can bridge the gap between governance and action. 33,34

Systems thinking
Systems knowledge refers to an understanding of the complex interactions, between the many actors and processes in the fields of human, animal and environmental health that emerge and feed back over long time scales. To integrate such knowledge, the management discipline known as systems thinking can be used. Systems thinking can assist human thought by permitting the analytical inference of dynamic consequences, from complex nets of long causal chains that often have feedback loops and unintended effects. System thinking also allows information from multiple sources, e.g. quantitative data, expert knowledge and stakeholders' experiential insight, to be combined systematically. 35,36 These different sources of information are complementary because of missing data, methodological differences and interestbased selective perception, even among members of the same scientific team. 37,38 By using all of the available relevant information to understand the possible outcomes of policy interventions and by linking diverse bodies of relatively abstract information with the narratives that guide everyday experience, systems thinking can reduce uncertainty in complex governance problems. 15,39,40 Transdisciplinarity Transdisciplinary approaches, which are sometimes called boundary management, are designed to build a bridge, at the science-policy interface and between potentially diverse knowledge systems, by facilitating communication, mediation and translation across cultural, disciplinary, institutional and/ or sectoral divides. Although multiple analytic methods may be employed, 29,37 the distinctive characteristics of such approaches are mainly sociocultural and aim to foster collective action towards societal transformations. 38,41,42 They include the selection of actors that legitimately represent the interest groups of relevance to the research problem. Coleadership helps to ensure the equitable representation of interests and perspec-tives and to mitigate power differentials. The joint negotiation and definition of research objectives and hypotheses is a crucial step in linking target knowledge, building mutual understanding and enabling successful collaborations. Linking narratives and experiential perceptions with conceptual or explanatory, systems knowledge is a central challenge. This challenge can be overcome by careful consideration and the development of a deep understanding in experiential encounters, by repeatedly exposing the different bodies of knowledge to each other and by working towards joint outputs. The sustained commitment of the varied stakeholders needs to be supported by strong leadership, trust building and conflict management. 28,29,37 Transdisciplinary approaches may make three crucial contributions to societal transformations. First, they create social contexts for successful knowledge integration, even where such contexts do not occur naturally. Second, as a result of their collaborative and interactive nature, they tend to produce knowledge that is generally perceived as credible, legitimate and salient. Finally, by fostering collaboration among societal and scientific partners, they can build trust and networks that are independent of any hierarchical chains of command.

Case studies
We believe that the effective implementation of the One Health strategy, as an interdisciplinary and intersectoral approach that links different forms of knowledge and expertise across multiple institutions, depends on knowledge integration. Six case studies support this view: three general One Health initiatives and three integrated health initiatives that included multicriteria analyses, systems thinking or a transdisciplinary approach ( Table 1).

Integration of target knowledge
The integration of target knowledge has been fostered by including stakeholder perspectives in agenda setting and decision-making, through either explicit co-leadership and negotiation 43,45,48 or changes of perspective in collaborative work assignments. 44,46,48 In Quebec, Canada, a rigorous multicriteria analysis, of Lyme disease surveillance and control strategies was used to support the public health authorities' decision-making and programme direction. 47 In the latter (. . .continued) investigation, a participative approach that involved health professionals and other stakeholders from governmental and nongovernmental organizations was used to compare several surveillance strategies in terms of their likely animal, environmental and public health and socioeconomic impacts. The stakeholder group provided input during the definition of management strategies, the assessment of objectives and their relative importance and the scoring of the strategies in terms of their likely attainment of the objectives. Since stakeholders represented their institutional perspectives, the process presumably assured the balanced representation of each of the relevant institutional viewpoints. The analyses allowed preference rankings of several possible intervention strategies for the management of Lyme disease, facilitated a better understanding of the conflicts between the key objectives and the relevance of such conflicts to each stakeholder group, and apparently improved each stakeholder group's appreciation of the preferences and priorities of the other stakeholder groups. In short, the analyses contributed to resolving trade-offs and setting a common vision and direction. While multicriteria analyses have mostly been focused on the early stages of policy development, e.g. agenda setting and policy formulation, they have important evaluative elements and can build consensus, to strengthen collective action, during policy implementation ( Table 1).

Integration of systems knowledge
The integration of systems knowledge has been used in the joint definition of broad conceptual bases for the collection and assessment of evidence 43,45,47,48 and in facilitating group understanding of the evidence collected via collaborative data analysis and validation (Table 1). 43,45,47 In the United Kingdom of Great Britain and Northern Ireland, a comprehensive intersectoral review of the activities, culture, effectiveness, policies and social relations within the child-protection sector demonstrated how One Health governance could be supported by structured and rigorous systems thinking. 46 This review engaged a reference group of relevant stakeholders, e.g. representatives from charities, the civil service and other government departments, an adoptive mother and young people who had been through the child-protection system themselves, and drew on evidence from databases, written sources and individual stakeholders' perspectives. The collaborative development of causal loop diagrams, with 60 variables, facilitated both a better understanding of the systemic outcomes of interdependent decision-making processes and a comparative assessment of potential policy interventions. The recommendations drawn from this review's results were largely accepted by the commissioning government authority and triggered substantial policy changes. 46 Systems thinking can therefore transform complex mixtures of individual observations into coherent narratives that state how situations emerge and how they may unfold in the future. While systems thinking has mostly focused on the evaluation stage of policy development, it usually includes target knowledge, as a determinant of behaviour, and its participative nature can also build trust and foster mutual learning between stakeholders and scientists.

Integration of transformation knowledge
Most One Health and related initiatives rely on a multi-institutional network of actors. This network often contributes to the integration of transformation knowledge in two ways: via the institutional support provided by relevant decisionmakers 43,46,47 and via the collaboration of individuals who have a broad range of implementation-related skills and expertise in many specialist fields. 43,44,46,47 The potential usefulness of transdisciplinary approaches for coordinating and managing such interdisciplinary, intersectoral and intercultural collaboration, even in challenging societal contexts, was illustrated by a collaboration in Guatemala. 48 The main aim of this collaboration was to bridge the gaps between the knowledge systems of biomedical doctors and those of traditional Maya healers and, in so doing, promote collaboration and mutual learning between the two groups. After facilitating joint patient diagnosis and subsequent treatment reconstruction, the collaboration was deemed useful and relevant by both groups of subjects and appears to have reduced the longstanding prejudices held by each group towards the other. Scientific institutions that, in terms of these prejudices, were perceived as neutral acted as intermediaries and helped ensure the credibility of the results. The process provided multiple opportunities for the building of mutual trust, via dialogue and experiential exchange and also triggered reflection, by pointing out the shortcomings of the current health systems, and appears to have educated all of the participants. In short, it developed and/or strengthened the networks for collective action. While the Guatemalan study focused on the implementation stage of policy development, the transdisciplinary approaches also had effects on agenda setting, by influencing the actors' target knowledge and on evaluation, by enabling process assessments that were more inclusive of the divergent knowledge systems (Table 1). Furthermore, the new networks and increased levels of trust helped achieve consensus and collective action at every stage of policy development.

Discussion
We believe that knowledge integration is both an integral element of successful One Health governance systems and a prerequisite for the effective implementation of the One Health approach. The combined use of multicriteria analyses, systems thinking and transdisciplinary approaches (Fig. 1) could contribute to more systematic and successful collaborations within and across existing institutions and form a procedural backbone for converting the aspirations of the One Health concept into institutional processes. In general, the aim of multicriteria analyses, systems thinking and transdisciplinary approaches is to create, maintain and inform collective action by broad coalitions of societal partners. If successfully implemented over extended time spans, they could contribute to the building of trust, networks and institutions that are not primarily dependent on any existing hierarchical structures of government. Although multicriteria analyses, systems thinking and transdisciplinary approaches mainly focus on different, crucial aspects of One Health governance, they are complementary and overlapping rather than mutually exclusive. They provide methods to resolve trade-offs and set a common vision and a common direction across disciplines, institutions and sectors. They serve as a toolbox for systemic monitoring and feedback to transform observations into narratives detailing how situations emerge and might unfold in the future. Finally, they contribute to the development and/or strengthening of Knowledge integration in One Health governance Martin Hitziger et al.

摘要
"唯一健康"政策制订、实施和评估中的知识整合 "唯一健康"概念涵盖了人类健康、动物健康和环境 健康之间的相互关系 , 需要多个利益相关方在诸多文 化、学科、机构和部门之间的合作。然而 ,"唯一健康" 的实施方法似乎受到全球卫生治理框架中的缺陷的阻 碍。知识整合方法在政策制订的各个阶段都可以帮助 解决这些缺点。关键目标的确定、权衡的解决以及建 立一个共同愿景和展开叙述可以通过多重判据分析来 实现。通过系统性思考 , 可实现基于证据的决策并将 networks for collective action towards a common vision. Potentially, therefore, as a decisive element in policy development, knowledge integration could help resolve the main shortcomings of the current global framework for health governance, by managing complexity and shaping interactions between actors and institutions towards joint learning. 17,18 Knowledge integration could also be used to complement educational and institutional measures for improving the imple-mentation of the One Health approach. 14 We therefore propose that policy cycles relevant to One Health should aim at knowledge integration and make the best possible use of multicriteria analyses, systems thinking and transdisciplinary approaches. Whenever they are used as elements of the implementation of the One Health approach, the processes involved in knowledge integration should be reported explicitly in the associated scientific articles. Ideally, such reporting should be based on standardized criteria and systematic evaluation frameworks, like the one proposed by the Network for Evaluation of One Health. 5,20 To develop and improve best practices in One Health, the practitioners and scientists in active One Health networks should be educated on knowledge integration and encouraged to discuss their ideas with those of more established governance actors, ideally in programmes supported by permanent professional associations or organizations. Finally, attention should be directed towards developing and implementing efficient technical mechanisms to facilitate stakeholder involvement and brokering at all levels of health governance, from local to global level. ■