Building Health Surveillance for Decision Support at the Animal, Human, Environment Nexus

John Berezowski, Craig Stephen, and Luis Pedro Carmo

The complex intersectoral and multi-level nature of the health of humans, animals, and environments means that it is not possible to prescribe an all-embracing policy that ensures decisions made to protect the health of one species, or generation, do not harm another. The world is a dynamic place where recent factors such as globalization, big data, climate change, and others have accelerated change from local to global levels. Changing conditions result in new problems and new opportunities with which societal organizations must deal. When new problems arise, existing information can be insufficient. New information will be needed to support decisions that enable optimal solutions. Information and knowledge are essential components of decision-making when uncertainty is present (Yim et al., 2004). Continuous production of new information and knowledge has become essential for organizations to be effective in this environment. Decision-makers need an expansive perspective so that decisions are less likely to have harmful unintended consequences, are more adaptable, and help reduce vulnerabilities to emerging and persistent harms. The nature of health under global change, and how it may be influenced by alternative decisions, is uncertain. Managing under these conditions requires new collaborations to gather information and perspectives to support integrated health management decisions.

Despite an urgency for interdisciplinary w'ork in the Anthropocene, most surveillance programs remain bound to disciplinary conventions. During the last century, most population health programs were developed to deal with single diseases that occur in a single population of a single species. These single disease- species systems were effective for many contagious diseases that were important in the previous century. However, these systems were not completely successful and there is growing recognition that new approaches are needed for dealing with more complex problems, such as cross-species disease emergence, the effect of climate change on human, animal and ecosystem health, and antimicrobial resistance (Wilson et al., 2020). Since humans, animals, and other species live in an interconnected w'orld that is constantly changing, we should not be surprised that health and disease constantly change and that new complex health problems constantly emerge. As organizations begin to operationalize new collaborative, cross-sectoral programs, they require a constant stream of new information. Surveillance at the interface of animals, health, and society needs to be reconceived as an information system made up of a chain of subsystems, each with its own knowledge of an aspect of the social and ecological drivers of health or harm. It must be conceptualized as a connected system of human knowledge and wisdom distributed throughout the socio-ecological chain of interactions that affect vulnerability and resilience.

HEALTH SURVEILLANCE AS AN INFORMATION SYSTEM

Decisions are built on a pyramid of data, information, knowledge, and wisdom (DIKW) (Figure 7.1). The DIKW hierarchy has been used to understand the role of information and knowledge in risk assessment in engineering (Aven, 2013), and we have adapted it to health surveillance.

Data are symbolic representations of observations of the real world. These can be observations of health outcomes, hazards, or other indices tracked by surveillance systems. Data alone have no intrinsic value. Information is data that are made useful, through processing and analyzing. Knowledge adds understanding (or knowing) of the importance of the information created. This entails examining information within the prevailing state of knowledge. Wisdom arises when knowledge is considered w'ithin its social and ecological reality. Wisdom helps us make the best decision based on the knowledge created, given the real-world opportunities, constraints, perspectives, and priorities.

DIKW systems can roughly be classed as static or dynamic. Static systems produce the largest volume and variety of DIKW used for health decision-making. It includes basic sciences - many applied disciplines such as pathology, epidemiology, sociology, human and veterinary medicine, engineering, animal science, experiential knowledge and wisdom held by community stakeholders, and more.

The data, information, knowledge, and wisdom (DIKW) hierarchy adapted to health surveillance

FIGURE 7.1 The data, information, knowledge, and wisdom (DIKW) hierarchy adapted to health surveillance.

These DIKWs are essential for understanding the complexity underlying health problems and solutions proposed to deal with them. The information and knowledge produced by static systems tend to change relatively slowly and, therefore, do not provide timely information about rapidly changing events.

Dynamic systems continuously generate DIKW through ongoing processes such as surveillance and monitoring. This type of information is often relatively simple but is essential for identifying changing health conditions or the emergence of new problems such as epidemics, as they occur. Population health surveillance, whether for humans, plants, or animals, can be defined as a continuous, dynamic process producing information that is used to inform decisions for dealing with health or disease-related problems (El Allaki et al., 2012). Surveillance systems create information that helps describe current health situations or helps forecast future health states. Since health surveillance is a continuous activity, it is particularly important for decisions that must be made in a constantly changing environment.

Health surveillance data are often collected from cases. What constitutes a case varies considerably, depending on the information output required. Cases can include laboratory-confirmed cases of a disease, health care provider (medical or veterinary) identification of suspect cases, patient self-reports of clinical symptoms, owner reports of clinical signs in animals, various surrogates for disease (for example counts of sales of diarrhoea medications as a surrogate for occurrences of diarrhoea in a population), antimicrobial-resistant bacterial or antimicrobial resistance genes, and many others (Antoine-Moussiaux et al., 2019). Other descriptive data are commonly combined with case data to provide contextual information. These include, at a minimum, the date and geographic location of case occurrences but can also include other data such as age, sex, breed, production type, and socio-ecological data such as socio-economic status of the case or the case owner, weather, nutrition, and other data. The additional data help to provide an epidemiological profile of cases and context for the cases that is useful for forecasting the risk of future events in specific times, places, or groups.

Surveillance data become information when they are processed and analyzed to answer specific questions. Examples include: How many cases have occurred in each region in the most recent period? Has the number of cases increased/ decreased compared to previous time periods? Are there more cases among individuals with specific characteristics (e.g. age or production type)? This information is used to make inferences about whether the disease risk has changed and to make predictions about whether the risk is likely to change and among which regions or subpopulations.

Surveillance knowledge is understanding the importance of the information produced. It is required to answer the question: Does the most recent disease risk information warrant an action? Decision-makers must have access to additional knowledge to make decisions, most importantly an understanding of the potential harm that is likely to occur in a population for different disease risk estimates. Information and knowledge about the epidemiology of a disease, the expected spread of the disease, the vulnerability of the population, and the expected economic and other losses can enhance the value of surveillance information. One Health decision-makers who operate programs aimed at benefiting the health of humans, animals, and the environment require rich contextual knowledge. For example, they will need to understand the complex linkages between the health of many species and the environment if they are going to effectively estimate the potential harm that an increase in disease risk in one (or more) species may have on all species involved. Currently, population health is mostly domain specific, and decision-makers are educated mainly within the domain they work. One Health decision-makers will require skills and capacities to acquire, understand, and use the additional knowledge needed to make effective decisions.

Wisdom is being able to use surveillance information and knowledge in the most appropriate way, resulting in the least harm and greatest benefit to as much of the population and as many of the stakeholders as possible. Wise decisionmakers must understand the significance of risk estimates and be able to design the most effective and appropriate actions. As we consider more species and populations in decision-making, the limitations of single species or single disease- focused surveillance quickly becomes apparent.

Health surveillance is typically a technical activity focused mainly on data creation and analysis leading to information creation. These activities are conducted by highly trained specialists who often perform their tasks with limited input from people outside their specialty. This specialist approach has elevated the importance of data and analysis, shifting emphasis aw'ay from knowledge and wisdom, and excluding many stakeholders who could provide valuable knowledge to enhance surveillance information.

The data information centric approach works well when health problems are well known, clearly defined, and the decision being made is unambiguous. For example, the U.S. Centers for Disease Control (CDC) defines an epidemic of listeriosis (caused by Listeria monocytogenes) as the occurrence of two or more related cases. The detection of two or more cases evokes an action (an outbreak investigation to find the source of food or drink contamination) (CDC, 2020). Behind this data information system is knowledge and wisdom gained from experiencing past epidemics such as knowing (i) that the occurrence of two related cases of listeriosis in the United States is abnormal and could indicate that an epidemic is ongoing in the population; and (ii) that listeriosis has a high case fatality ratio, making it important to respond immediately in order to minimize the number of people that become infected. The additional wisdom is understanding that the source of food or water contamination must be identified before the epidemic can be stopped, and therefore the best action is to conduct an outbreak investigation.

The importance of knowledge and wisdom, or the lack of it, is more evident in less w'ell-defined situations. In the COVID-19 pandemic of 2020, the earliest piece of surveillance information was the identification of cases of a previously unseen, severe respiratory syndrome in Wuhan, China (Wang and Wang, 2020). At the time, knowledge of the importance of this information w'as uncertain, and as a result, some decision-makers decided not to immediately initiate a response (Liu et al., 2020). As the epidemic spread, it became more certain that the new disease was rapidly spreading and had a relatively high mortality rate. Decisionmakers decided to respond based on wisdom gained from the first severe acute respiratory syndrome (SARS) epidemic in 2003 (Liu et al., 2020). Even after it became evident from the experience in China that the epidemic was a significant threat, decision-makers in different countries responded differently (Kluge, 2020). In some countries, the threat was immediately recognized and responses were rapidly implemented. In others, the threat was not recognized as early and disease control actions w'ere delayed. The variation in rapidity and type of national responses could be attributed to many factors, but uncertainty in information and differences in the know ledge and wisdom of decision-makers in different countries are likely to have played a role.

Recognizing the importance of emerging disease signals identified by surveillance and implementing the most effective and efficient disease mitigation strategies requires decision-makers with considerable knowledge and wisdom. It is unlikely that any single individual will have the necessary knowledge and wisdom. Transdisciplinary teams that include experienced experts from a range of scientific and practical specialties as well as representatives of different parts of society are more likely to correctly identify signals that are important and develop interventions that are effective and efficient in terms of costs and negative effects on people, society, other species, and the environment. There are established methods to guide these types of evaluations such as rapid risk assessment methods to estimate the risk of a pathogen introduction and spread and the potential harm that it could cause (WHO, 2012).

 
Source
< Prev   CONTENTS   Source   Next >