(Authors: Nathan Mugenyi1, Prize Ninsiima2,Pauline Byakika-Kibwika3,Rose Nanyonga Clarke4)
1Faculty of Medicine, Mbarara University of Science and Technology, Mbarara, Uganda
2 School of Medicine, Uganda Christian University, Kampala, Uganda.
3College of Health Sciences, Makerere University, Kampala, Uganda.
4Clarke International University,Kampala,Uganda.
Concerns regarding antimicrobial use in farm animals grew considerably due to the growing prevalence of antimicrobial resistance (AMR) and the way this affects human health for almost the past two decades. AMR is a natural molecular process that results from the ability of microorganisms to quickly adapt to changing conditions through mutations, change in configuration of the active site, drug inactivation and efflux mechanisms which render antibiotics less effective. Indeed, the appearance of rare and advantageous mutations that neutralize the effects of antimicrobials is inevitable in large and dense microbial communities and the rapid generation times allow these mutations to quickly become prevalent in growing communities. Additionally, bacteria have the capacity to exchange mobile genetic elements, including resistance genes, via horizontal gene transfer within and between bacterial species, further enhancing their ability to adapt and resist.
But while AMR is a natural phenomenon, its increasing prevalence is most certainly not. In fact, it is fueled by anthropogenic factors such as the intensive clinical and agricultural use of antimicrobials worldwide, the growth of the world’s human population, changes in human lifestyle (such as, increased urbanization, migration and travel), and misconceptions and malpractices regarding antimicrobial use (AMU) . Over time, this increasing prevalence is predicted to have a significant impact on global health and wealth by potentially causing up to 10 million deaths each year, at a cumulative cost of $100 trillion to global economic output by 2050. To further contextualize this, the World Bank Group estimated that reductions in annual global GDP due to AMR (ranging between 1.1 and 3.8%) may be comparable to the losses caused by the 2008–2009 financial crisis, with the difference that the economic damage would continue for decades and would mostly affect low and middle income countries (LMICs).
To contain this serious threat to global health and wealth, the 194 member states of the World Health Organization (WHO) endorsed a global action plan (GAP) in 2015 and committed to establishing national action plans (NAPs) based on the “One Health” approach, which recognizes the interaction between human health, animal health and the environment.
By 2018, 60% of Member States declared having a NAP in place and 33% reported that they were in the process of developing one. This global attempt to contain AMR with a One Health approach seems timely, since it was estimated that by 2030, antimicrobial consumption, which has now been repeatedly associated with AMR would increase by 67% in livestock, by 33% in aquaculture and by 15, 32 or 202% in humans, depending on the scenario.
To address the challenges posed by antimicrobial resistance, countries have been advised to invest in AMR containment through AMR surveillance, infection Prevention and control (IPC) and through curbing the prevalence of antimicrobial resistance via optimal antimicrobial prescription and use in both human and veterinary medicine. Regarding the latter, the institutionalization of Antimicrobial use (AMU) as well as the reduction of antimicrobial dependence is necessary in order to achieve a sustainable use of antimicrobials.
In livestock production, antibiotics play a crucial role since they are not only a therapeutic but also an economic asset. The preventive use of antimicrobials to treat at-risk herds or animals (prophylaxis) as well as clinically healthy animals sharing premises with symptomatic animals (metaphylaxis) allows the limitation of economic risks and labor costs. Outside of Europe and the USA, antimicrobials are also used as feed additives, which are thought to improve animal growth, feed conversion and yield and allow farmers to keep pace with the demand for meat while lowering the prices.
To reduce this reliance on antimicrobials, the focus is often on information and technological innovations as vaccination and alternatives to antimicrobials. But while investments in therapeutic innovations are foreseen in the GAP, and presumably the NAPs based thereon, the promise of new technologies might not be enough. In fact, therapeutic alternatives to antimicrobials are currently not sufficiently developed in order to effectively replace antimicrobials. Considerable investment in research and development will be needed, which means that these options will not be widely available in the coming years. Moreover, it is very likely that such options will offer short-term solutions since we are engaged in an infectious arm’s race with microbes that always find a way to accommodate to new therapeutics.
In this regard, Smith suggests that the vision for AMR control is currently focused on technological and biomedical innovations, the benefits of which could be short-lived if our society remains heavily dependent on antibiotics. In addition, there is no guarantee that alternatives will be immediately adopted by farmers, as it was the case for the live oral Lawsonia vaccine in pigs, which was not widely used despite positive results. Studies cut across many disciplines have shown that the adoption of new technologies by farmers can be influenced by numerous factors, for example, environmental factors such as land use and land characteristics ; personal features such as age, human capital or risk preferences; economic attributes such as market intervention by regulators and costs of acquiring the technology; extension services as well as cultural and social factors including social identity, social networks and peer group influence. It is therefore clear that farmers’ behavior is embedded in both biophysical and social landscapes and that decision-making processes are complex and context dependent. In addition to this, other actors in the social landscape may also indirectly influence farmers’ behavior by voluntarily or involuntarily creating physical (that is, land appropriation) or social structures (such as, norms) that restrict, or enlarge, farmers’ opportunity space. To better understand farmers’ behavior while considering the systemic complexity in which it is embedded, several frameworks and systemic approaches have been developed with the hope that this would help design research that represents farmer’s behavior more realistically and that it would lead to the development of more effective sustainable agricultural policies.
In this regard, our objective is to add to an interdisciplinary research agenda by providing a perspective on strategies for reducing the dependence on AMU and the threat of AMR from a social science and economic point of view. This perspective was inspired by how social scientists and economists contributed to environmental policies. We discuss how knowledge about farmers’ behavior and the system in which they operate can contribute to answering central concerns for the development of policies and strategies and can provide a clear taxonomy of AMR interventions in livestock production along the value chains.
In livestock production, antimicrobial resistance (AMR) is considered an externality as it is the undesired result of preventive and curative antimicrobial use. To address this biosocial issue, our objective is to present an approach based on interdisciplinary research to develop strategies and policies that aim to contain AMR.
To effect this, we addressed fundamental concerns on which control policies and strategies for agricultural pollution problems are centered in the light of AMR. To ensure the technical, economic, behavioral and political feasibility of the developed measures, we demonstrated the usefulness of systemic approaches to define who, what and how to target by considering the complexity in which the ultimate decision-maker is embedded. We then define how voluntary or compulsory behavioral change can be achieved via certain routes, introducing a clear taxonomy for AMR Interventions.
Fundamental Concerns for Defining Policies and Strategies to Mitigate AMR in Livestock farming
Since AMR in livestock and agricultural pollution are both externalities, the design of policies and control strategies for the latter might also be useful for the former. We therefore used three fundamental questions on which the design of policies and control strategies for agricultural environmental pollution is centered to develop an approach to design new policies and strategies to mitigate AMR in livestock. The first fundamental question is: who among those who play a role in the production of an externality should be targeted. The second question aims to determine the basis for measuring effectiveness or, in other words, what variable(s) control policies and strategies wish to change. Finally, the third question is how to target, i.e., by what mechanism(s) the intended actors and bases should be targeted.
Identifying Key Actors in Antimicrobial Decision Systems
To reduce antimicrobial use and dependence in livestock production, it is necessary that farmers, as ultimate users, and veterinarians, as antibiotics prescriber, change their behavior. It is thus only logical that policies and strategies target them.
However, the solution to the question “who to target” does not need to be limited to the actual users of the antimicrobials. Indeed, when dealing with externalities, it has been suggested that, in addition to the actual source, others could be targeted. This idea has been reinforced by systemic approaches that suggest that there are many more actors who have an influence on how food is being produced and that often, farmers are end-of-pipe decision makers largely influenced by the practices and demands of other actors in the system.
In this regard, value chain approaches for the analysis of animal health systems grew in popularity as they allow for the analysis of the different actors involved, their roles, and the interactions between the actors as well as how this influences practices. Since value chains are in turn embedded in a bigger biological, social, economic, and regulatory context, analytical frameworks have been developed to study such big and complex systems. For example, Lamprinopoulou et al developed a framework to analyze Agricultural Innovation Systems (AIS), which consists of innovation processes that encompass all type of knowledge that all actors in an agricultural system demand and provide, as well as the interaction between these actors. The framework allows to define the functions and structures, that is, identification and classification of actors, of an AIS and to assess how, at a micro level, systemic failures may affect the contribution of actors to the fulfillment of the functions of the AIS. Moreover, the functioning of the entire system is also explored by investigating if basic structural components and functions are sufficiently coordinated, aligned and harmonized. This approach was further used by Rojo Gimeno et al to comprehensively depict swine health systems by identifying key actors and their functions as well as merits and failures at micro and macro level that impact functions.
When considering the value chain, it becomes more clear that besides farmers and veterinarians, other actors of the value chain could be targeted. In the upstream part of the value chain, input suppliers such as feed mills could be subject to policies, that is, by banning or further regulating the production of certain inputs like non-medicated feed to avoid cross-contamination with antimicrobial residues. Targetable actors can also be found in the downstream part of the chain, as, for example, the knowledge of truck drivers regarding the health of animals for transportation could be regulated and the compliance of transportation companies to strict rules regarding the cleaning and disinfection of Lorries could be controlled. Lately, labeling systems can also be set up to provide information about the antimicrobial use during the production of animal products.
The policy domain comprises several levels such as (supra) national and regional governments. The intermediary domains refers to actors that, on the one hand, advise governments and may perform governmental activities and, on the other hand, may influence the value chain as well as the research and education domain via collaborations and the development of awareness campaigns. Finally, the research and education domain comprises schools, research institutions and universities developing and providing knowledge for the other actors as well as private and public extension organizations. Such external actors could also be targeted by policies by for example subsidizing control agencies or farmers’ organizations to develop communication campaigns to raise awareness. Universities could be expected to improve courses on AMR in the curriculum of veterinarians or farmers. Educational campaigns about AMR, antimicrobial stewardship or biosafety could be promoted for farmers, veterinarians and also advisors (from feed mills or companies that work on farm equipment). Such campaigns can also be supported by industries, such as pharmaceutical companies. Investment in R&D could allow the development of new tools.
Defining a Basis for Policies and Strategies
Here, we will explore different options that may provide an optimal basis for measuring impact, or in other words, a variable that policies and strategies are intended to change. To serve as an optimal base to formulate a regulation or strategy and to measure compliance with that regulation/strategy, any elements in the input/technology-production and AMR relationship can be used as long as they are correlated with AMR, enforceable and targetable in space and time.
Taxonomy of AMR Interventions: Routes to Behavioral Change
Once it is defined who and what should be targeted, the next step is to determine the mechanisms through which the intended actors and variables can be targeted. This entails the choice for a policy instrument, an advisory approach, a structural intervention and other types of options. To change behavior, several intervention frameworks have been developed for different contexts such as policy making. While the used vocabulary or level of detail regarding the categorization of the intervention type may differ, most of these frameworks consider regulation and coercion, norms, social influence and networks, knowledge, incentivization and enablement as factors that may influence one’s behavior.
According to Van Woerkum’s framework which we chose, a first way to achieve behavioral change involves regulation. This route differs from the others in that it attempts to make change compulsory, in contrast to the others that strive to induce voluntary change. Therefore, “bad” behavior is made illegal while “good” behavior is made mandatory. The best known regulations on antibiotics are the ban on the use of antimicrobials as growth promoters. To induce a voluntary change, the second route includes provisions and tools, which are instruments that are implemented to change the external material circumstances so that people become motivated to change their behavior. In some cases, the provisions can be restrictive (by making the ‘bad’ behavior less straightforward) and behavioral change is then coerced. However, in the case of AMR and AMU, most provisions and tools are rather enabling and give the intended person external motivation to voluntarily change by making it easier and more achievable to reduce and improve the use of antimicrobials. Examples of tools include coaching sessions to develop and implement farm health plans and alternatives treatments such as bacteriophage therapy.
Finally, the last route to voluntary behavioral change goes through communication and education. Through these mechanisms, change agents attempt to change the internal motivation of decision makers so that they become convinced that behavioral change is the best decision. Newly developed tools or economic studies that demonstrate the cost-effectiveness of measure to reduce AMU, such as improved management strategies (that is, biosecurity strategies) can be used as incentives. Within this category, typical extension instruments such as articles in agricultural magazines, demonstration farms, leaflets, study days, digital apps, and others are found. This is arguably one of the most used and investigated routes, mainly in the field of social veterinary epidemiology, which is the study of human behavior that affects the causes, spread, prevention and control of animal diseases and health problems, and related disciplines.
Strategies and policies that focus on reducing use and dependence to antimicrobials often do not take the behavioral character of AMU and AMR into consideration. To address this, we have introduced an approach that relies on interdisciplinary systemic approaches to comprehensively characterize antimicrobial decision system, hence identifying all actors influencing AMU in livestock production, adequate regulatory and intervention bases, which behavioral change strategies to use and whom should implement this.
Interdisciplinary systemic approaches enable the development of AMR policies and strategies that are technically, politically, economically and, last but not least, behaviorally feasible by allowing the identification of ;(a) all actors influencing AMU in livestock production, (b) power relations between these actors, (c) adequate regulatory and intervention bases, (d) what behavioral change strategy to use, (e) whom should implement this, as well as the cost-effective assessment of combinations of interventions. Unfortunately, AMR policies and strategies are often investigated within different disciplines and not in a holistic and systemic way, which is why we advocate for more interdisciplinary work and discuss opportunities for further research.
 J. O’Neill, “Tackling drug-resistant infections globally: final report and recommendations,” 2016.
 D. Abera, F. Desissa, and J. Endris, “Mechanisms of development of antimicrobial resistance in bacteria: A review,” 2016.
 A. Frenoy and S. Bonhoeffer, “Death and population dynamics affect mutation rate estimates and evolvability under stress in bacteria,” PLoS Biol., vol. 16, no. 5, p. e2005056, 2018.
 F. Baudoin, H. Hogeveen, and E. Wauters, “Reducing antimicrobial use and dependence in livestock production systems: a social and economic sciences perspective on an interdisciplinary approach,” Front. Vet. Sci., vol. 8, p. 584593, 2021.
 G. Maki and M. Zervos, “Health care–acquired infections in low-and middle-income countries and the role of infection prevention and control,” Infect. Dis. Clin., vol. 35, no. 3, pp. 827–839, 2021.
 B. Klepac Pogrmilovic et al., “A global systematic scoping review of studies analysing indicators, development, and content of national-level physical activity and sedentary behaviour policies,” Int. J. Behav. Nutr. Phys. Act., vol. 15, no. 1, pp. 1–17, 2018.
 M. Nathan, “Antibiotic resistance,” Community Engagem. curtailing Preval. Antimicrob. Resist., vol. 40, no. 1, pp. 63–75, 2021, doi: 10.1016/j.cnur.2004.08.006.
 C. Kirchhelle, “Pharming animals: a global history of antibiotics in food production (1935–2017),” Palgrave Commun., vol. 4, no. 1, pp. 1–13, 2018.
 W. H. Organization, “Food and agriculture organization of the United Nations,” Vitam. Miner. Requir. Hum. Nutr., vol. 2, pp. 17–299, 2004.
 V. M. Andres and R. H. Davies, “Biosecurity measures to control Salmonella and other infectious agents in pig farms: a review,” Compr. Rev. Food Sci. Food Saf., vol. 14, no. 4, pp. 317–335, 2015.
 M. Schlüter et al., “A framework for mapping and comparing behavioural theories in models of social-ecological systems,” Ecol. Econ., vol. 131, pp. 21–35, 2017.
 D. W. Cash et al., “Knowledge systems for sustainable development,” Proc. Natl. Acad. Sci., vol. 100, no. 14, pp. 8086–8091, 2003.
 F. Petetin, G. Bertoluci, and J.-C. Bocquet, “Decision-making in disruptive innovation projects: A value approach,” 2011.
 M. Carley, Rational techniques in policy analysis: Policy studies institute. Elsevier, 2013.
 R. Laxminarayan et al., “Antibiotic resistance—the need for global solutions,” Lancet Infect. Dis., vol. 13, no. 12, pp. 1057–1098, 2013.
 P. J. Jones, E. A. Marier, R. B. Tranter, G. Wu, E. Watson, and C. J. Teale, “Factors affecting dairy farmers’ attitudes towards antimicrobial medicine usage in cattle in England and Wales,” Prev. Vet. Med., vol. 121, no. 1–2, pp. 30–40, 2015.
 A. Surana, S. Kumara*, M. Greaves, and U. N. Raghavan, “Supply-chain networks: a complex adaptive systems perspective,” Int. J. Prod. Res., vol. 43, no. 20, pp. 4235–4265, 2005.
 M. Bowman, K. K. Marshall, F. Kuchler, and L. Lynch, “Raised without antibiotics: lessons from voluntary labeling of antibiotic use practices in the broiler industry,” Am. J. Agric. Econ., vol. 98, no. 2, pp. 622–642, 2016.
 A. Lekagul, V. Tangcharoensathien, M. Liverani, A. Mills, J. Rushton, and S. Yeung, “Understanding antibiotic use for pig farming in Thailand: a qualitative study,” Antimicrob. Resist. Infect. Control, vol. 10, no. 1, pp. 1–11, 2021.
 M. Anderson, K. Schulze, A. Cassini, D. Plachouras, and E. Mossialos, “A governance framework for development and assessment of national action plans on antimicrobial resistance,” Lancet Infect. Dis., vol. 19, no. 11, pp. e371–e384, 2019.
 G. Kok et al., “A taxonomy of behaviour change methods: an intervention mapping approach,” Health Psychol. Rev., vol. 10, no. 3, pp. 297–312, 2016.
 J. I. R. Castanon, “History of the use of antibiotic as growth promoters in European poultry feeds,” Poult. Sci., vol. 86, no. 11, pp. 2466–2471, 2007.
 H. D. Hedman, K. A. Vasco, and L. Zhang, “A review of antimicrobial resistance in poultry farming within low-resource settings,” Animals, vol. 10, no. 8, p. 1264, 2020.
Keywords: antimicrobial resistance, antimicrobial use, livestock production, systems thinking, behavioral change.
We highly acknowledge Seed Global Health and ReAct Africa Organization for the financial support they rendered to us during the course of this research work.
We equally acknowledge Prof Pauline Byakika from Makerere University and Associate Prof Rose Nanyonga Clarke for the research mentorship and guidance.