A One Health future to meet the AMR challenge?

This issue of the New Zealand Veterinary Journal focuses on antimicrobial resistance (AMR), because of its rising prevalence, its impact on morbidity and mortality, and uncertainty about our ability to continue to treat infections effectively into the future. While AMR genes are not of our making, the rise in prevalence of AMR, predicted since antibiotics were first used (Finland et al. 1946; Jawetz 1963), is undoubtedly anthropogenic. However the complexity of the relationships between antimicrobial use and resistance, and their impact on ecosystems including the humans and animals that inhabit them, makes AMR one of health’s “wicked problems” (Rittel and Webber 1973; Signal et al. 2013). In order to understand how best to respond to this problem, we need to pay attention to the complex systems that underpin it, in order to understand how best to respond.

Resistant organisms, or genes that confer resistance, have been identified in symptomatic and asymptomatic humans and animals, and in the physical environment (Huijbers et al. 2015). Transmission of organisms with AMR from animals to humans, directly or via the food chain, occurs but its magnitude and impact is unclear. Transmission from humans to animals probably also takes place. Transmission of resistant organisms from humans to other humans certainly occurs, and may of course occur via food handler contamination of animal products, making it difficult to estimate the importance of transmission from animals via food. The physical environment may be a reservoir of organisms with AMR and genetic material through contamination by humans or animals, or through the bacteria and fungi that have been manufacturing antibiotics for their own protection long before humans happened upon them (Holmes et al. 2016).

The ability of micro-organisms to survive in the face of compounds that are intended to kill them has a number of genetic bases. The increase in the proportion of such organisms amongst those that cause infections can be explained by antimicrobial use, which selects for resistant organisms and supports the establishment of them in new treated hosts. Some resistance mechanisms confer at least some fitness cost on those organisms carrying them. However, a large number of resistance genes have not been lost from the overall bacterial gene pool. Resistance mechanisms may be insufficiently costly, the level of antimicrobial use may be sufficiently high to maintain selection pressure, or co-location of resistance genes with others (e.g. plasmids), may enable their retention in the face of other selection pressures (Gullberg et al. 2014). Thus AMR continues to broaden and rise in prevalence.

How do we know how big a problem AMR is in New Zealand? Surveillance data exist but are patchy, and often data from different sources are hard to compare. Some AMR data are derived from active screening of healthy individuals (e.g. food animals at slaughter, hospital staff who are screened for methicillin resistant Staphylococcus aureus), some from clinical specimens, and some data are reported with both mixed together. The threshold for sending clinical specimens is usually unknown and probably varies between clinicians, making variability in prevalence of AMR between clinicians, institutions, infections, organisms, and species difficult to interpret. Nonetheless, reporting of extended spectrum β-lactamase-producing Enterobacteriaceae has increased in human clinical samples in New Zealand (Toombs-Ruane et al 2017), as has AMR in Escherichia coli isolated from canine urine clinical samples (McMeekin et al. 2017), and we need to respond (Laxminarayan et al. 2013).

A One Health approach to understanding and tackling AMR seems obvious and sensible, given its wide reach. Our incomplete understanding of how all the complex relationships between humans, animals, and the environment work to affect the prevalence of AMR must not prevent us from keeping the whole system in view, as we act to retain the effectiveness of antimicrobials for as long as possible. At the most basic level, we can learn from each other. The New Zealand Veterinary Association has set an aspirational goal that “By 2030 New Zealand Inc. will not need antibiotics for the maintenance of animal health and wellness”; a goal that could have some resonance for situations where doctors prescribe antimicrobials when there is no clear evidence of a bacterial infection that will respond better to antimicrobials than to supportive care. Hospitals are successfully employing antimicrobial stewardship pharmacists to directly influence antimicrobial prescribing, including improving the appropriateness of antimicrobial choice and enforcing rules such as limiting prescription of particular antimicrobials without special authorisation (Roberts et al. 2015; Nguyen-Ha et al. 2016). This type of approach might be a useful model for thinking about how to reduce the use of some products in veterinary medicine that are convenient but not best practice in many situations, e.g. prescribing injectable, long-acting and broad-spectrum antimicrobials for pets whose owners struggle to give pills. There is also some evidence that laboratories selectively reporting susceptibilities for appropriate first-line agents may help to change prescribing behaviour (Williamson et al. 2016).

In the late 1990s, a lot of attention was paid to AMR by editorialists (e.g. Kunin 1997) and policymakers (e.g. Anonymous 1998); it was considered a high priority by the New Zealand Ministry of Health in 2001 (Anonymous 2001), although the Ministry’s Antimicrobial Resistance Advisory Group was subsequently disestablished in 2010. Consideration of agricultural as well as medical use of antibiotics was included in recommendations for action (Anonymous 1998, 2001). However, since then there has been relatively little practical implementation of what would now be termed a One Health approach to tackling AMR. Denmark has been reporting human and animal antimicrobial use and resistance data in a single report since 1995 and, as noted by Toombs-Ruane et al. (2017), surveillance has demonstrated some successes in reducing antimicrobial use and resistance, but the links between animal and human health efforts are less clear.

New Zealand can no longer be complacent about antimicrobial use and resistance, both are now unacceptably high, at least in the human medical field (Thomas et al. 2014), with more data required for animals. Political, medical, veterinary, and epidemiological concerns around AMR are not the same, and in some instances are at odds with each other. Individual animal and human welfare should not be compromised by the imposition of regulation around antimicrobial use. However, to continue the trend of increased human consumption of antimicrobials would surely have a potential health and welfare impact on future generations; as would any imprudent antimicrobial use in domestic animals in New Zealand. While the use of antimicrobials for the “maintenance of health” in animals may not continue into the future, the stewardship of antimicrobials used for treating animal disease, and the surveillance of this use along with surveillance of resistant bacteria in New Zealand, requires collaboration between professions, investment of resources, and structured support from government.

In the last year, the World Health Organization, in coordination with the Food and Agriculture Organization of the United Nations and the World Organisation for Animal Health have recommended that all countries develop an AMR Strategic Action Plan, and promote a One Health approach to doing so (Anonymous 2016a). In New Zealand, the Ministries of Health and Primary Industries are collaborating to develop such a plan (Anonymous 2016b). We look forward to a clear plan for systematic, active surveillance of organisms with AMR and of antimicrobial use, with integrated reporting across animals, humans, and the environment, and for strategies for prudent antimicrobial use and infection control that are built on best practice from both human and veterinary medicine.