Antimicrobial resistance and antimicrobial use animal monitoring policies in Europe: Where are we ?

The World Health Organization has recognized antimicrobial resistance as one of the top three threats to human health. Any use of antibiotics in animals will ultimately affect humans and vice versa. Appropriate monitoring of antimicrobial use and resistance has been repeatedly emphasized along with the need for global policies. Under the auspices of the European Union research project, EFFORT, we mapped antimicrobial use and resistance monitoring programs in ten European countries. We then compared international and European guidelines and policies. In resistance monitoring, we did not find important differences between countries. Current resistance monitoring systems are focused on food animal species (using fecal samples). They ignore companion animals. The scenario is different for monitoring antibiotics use. Recently, countries have tried to harmonize methodologies, but reporting of antimicrobial use remains voluntary. We therefore identified a need for stronger policies.


Introduction
Antimicrobial resistance (AMR) is recognized as one of the major global public health threats. Reports emphasize its economic impact 1,2 and suggest a return to a ''pre-antibiotics'' era. 3 It is a perfect example of a ''one health'' issue, as any use of antibiotics in animals will ultimately affect humans (and vice versa). [4][5][6] It will have an associated environmental component 7-9 that recognizes no national boundaries. 10 The development of resistance to antimicrobial drugs is a natural phenomenon, but the overuse and inappropriate use of these drugs are associated with increased resistance. To control the problem, appropriate monitoring of use and of development of resistance is essential.
The World Health Organization (WHO) has recently recognized important gaps in (monitoring/surveillance) methods. No global consensus exists on standards for data collection and reporting of AMR across medical, veterinary, and agricultural sectors. 11 Comparison of results between countries is only possible when the results were obtained using the same (or similar/equivalent) procedures.
Under the Ecology from Farm to Fork Of microbial drug Resistance and Transmission (EFFORT) EU-FP7 project, we mapped current monitoring activities related to antimicrobial use and resistance in the ten European countries participating in the project: Belgium, Bulgaria, Denmark, France, Germany, Italy, The Netherlands, Poland, Spain, and Switzerland. From a 'gap analysis perspective,' we present the results as well as a critical comparison between the mapping results and current international guidelines and policies.

Materials and Methods
Initially, we took a very practical approach, having in mind the question: If a (European) country wants to set up monitoring systems to control the use and development of resistance to antimicrobials in animals, to which guidelines and policies should the competent authorities be looking? As a follow-up, we critically compared individual countries' policies, identifying gaps.
In a second step, we mapped what countries are currently doing to monitor antibiotic use and resistance, analyzing gaps. We considered activities related to both zoonotic and commensal bacteria. Zoonotic bacteria are those present in animals that can infect people, for example, by the food-borne route. Examples include Campylobacter, Salmonella, Listeria, and some strains of E. coli. Indicator (also called commensal) bacteria are those that are part of the normal flora of both humans and animals, such as enterococci. They can form a reservoir of resistance genes that may be transferred between bacterial species, including organisms capable of causing disease in both humans and animals.

International Policy Framework
The WHO Advisory Group on Integrated Surveillance of Antimicrobial Resistance (AGISAR) created a guidance document with key information for the design of programs of integrated surveillance of antimicrobial (use) and resistance. 12 Although not legally binding on countries, it does provide a generic overview of what countries need and shows how to achieve the mentioned goals.

European Policy Framework on Monitoring of Resistance in Animals
At the EU level, Directive 2003/99/EC 13 set out the goals for monitoring of zoonoses and zoonotic agents and related antimicrobial resistance. By definition, as a ''Directive'', it left it up to the countries to decide how to achieve these goals. This was followed in 2007, by the publication of Decision 2007/407/EC 14 (by definition, a ''Decision'' is a legislative act, binding on those to whom it is addressed) specifically focusing on the harmonized monitoring of AMR in Salmonella in poultry and pigs.
The most relevant current policy is probably Decision 2013/652/ EU. 15 It has a broader scope, addressing monitoring and reporting of AMR in zoonotic and commensal bacteria. The list of bacteria to be monitored and reported on includes Salmonella spp., Campylobacter jejuni (C. jejuni), and Campylobacter coli (C. coli) plus commensal Escherichia coli (E. coli). Commensal Enterococcus faecalis and Enterococcus faecium (E. faecalis and E. faecium) plus Salmonella spp. and E. coli producing Extended-Spectrum b-Lactamases (ESBL), AmpC b-Lactamases (AmpC) and Carbapenemases are also covered. EFSA provides the technical specifications for randomized sampling of harmonized monitoring of AMR in these bacteria. 12

Global and European Policy Framework on Monitoring of Use in Animals
At the global level, Chapter 6.3 of the Aquatic Animal Health Code, 16 OIE (Office International des Epizooties), provides guidelines on how to monitor the quantities and use patterns of antimicrobial agents in aquatic animals. The equivalent for food-producing animals is provided in Chapter 6.8 of the Terrestrial Animal Health Code. 17 Coordinated by the European Medicines Agency (EMA), the European Surveillance of Veterinary Antimicrobial Consumption (ESVAC) project was launched in September 2009, following a request by the European Commission to develop an approach for the harmonized collection and reporting of data on the use of antimicrobial agents in animals in the Member States. 18 ESVAC provides data collection, reporting, and analysis protocols that countries can follow. 13

Mapping Exercise
Initially, we mapped monitoring activities in place in 2014 related to antimicrobial use and resistance, including animals and food in countries participating in EFFORT. For this, we used a surveillance mapping methodology developed as part of another FP7 (The European Union's Seventh Framework Programme for Research and Technological Development) project RISKSUR, aimed at developing decision support tools for the design of cost-effective risk-based surveillance systems. 19 After an initial online training, data collectors received an MS Word template to be completed with information on the use of antimicrobials in their country, including • methods used to collect use data, • animal populations known (size), • indicator of use used, availability (or not) of Defined Daily Dose Animals (DDDAs -also known as DDDvet in ESAVC project), • the assumed average maintenance dose per day per kg body weight (for the main indication in a specified species), • classes of antimicrobials for which data were collected, specifications about the inclusion (or not) of data on premixes, • existence of specific policies to discourage or issue alerts about the overconsumption of antimicrobials, and • if veterinarians were allowed to sell antimicrobials.
We used a similar procedure to collect information about the monitoring of resistance, but via an MS Access database template. This database was completed with information about each monitoring activity/component, such as • the geographical focus, • legal framework, • target species and sectors, • sampling points and samples collected, • microorganisms tested, • means of data acquisition, • resistance criteria, and • whether the monitoring activity was funded and performed by the public or private sector.

Resistance
We did not find major differences in the way the monitoring of resistance is being performed in the European countries we studied ( Table 1). Testing resistance in microorganisms like Campylobacter jejuni, Campylobacter coli, E. coli, and Salmonella spp. is being done in all the countries. A few countries, France for example, have national programs for veterinary pathogens. Monitoring programs focus on the major food-producing species (poultry, cattle, and pigs). Slaughterhouses are the most common sample collection points, and the vast majority of the collected samples consist of fecal material. The monitoring activities are mostly active and under the control of the public sector.
Use Table 2 summarizes the monitoring activities for antimicrobial use in the ten countries. Sales data (from pharmacies, feed companies, wholesalers, or pharmaceutical companies) were the main source of data. They were used to derive/extrapolate use/consumption when such data were not available at the national level. Automated data collection exists in Denmark and The Netherlands and can be used in        combination with veterinary prescription data. In France, information is also collected via a retrospective longitudinal study. In Belgium, Bulgaria, and Italy, antimicrobial use data are not categorized by species. In other countries, it is not always possible to disaggregate the consumption to individual species. Dogs and cats are reported together in Denmark and France. There is no common way to group animal species. In Denmark, France, and Poland, ''cattle'' is reported as a single category (dairy + beef). In The Netherlands, usage data for rosé and white veal calves are reported separately from other cattle. In none of the countries is the size of the population of all the animal species known (live and slaughtered, when applicable).
''Total weight of Active Substance'' is the indicator of usage reported in seven countries: Belgium, Bulgaria, Denmark, France, Germany, and The Netherlands. In Belgium and France, ''Weight of Active Substance per biomass at risk to be treated'' (units: mg AS/PCU) is also used, and this is the only indicator used in Italy, Poland, Spain, and Switzerland. Additionally, France uses the Animal Level of Exposure to Antimicrobials (ALEA, the DCDA divided by the biomass). The Netherlands uses the ''Number of days treated per individual'' that is calculated as the consumption of antimicrobials per farm, as animal defined daily dosages per year (ADDD/Y); an ADDD/Y of 1 means that the average animal in the population was exposed to an antimicrobial for one day per year. This measure is similar to that proposed by ESVAC.
A list of DDDAs is not available in Bulgaria, Germany, Italy, Poland, Spain, or Switzerland. Belgium has established it by product (for pigs and poultry), active substance, administration route, and age group; Denmark by product, administration route, and age group; France by product, and The Netherlands by active substance and ATCvet category.
All the countries have data collected and available for all the antimicrobial classes in accordance with the ATCvet index list. Premix data -products containing more than one antimicrobial -are included in the usage information in all the countries, except in Germany and The Netherlands.
Unlike the other seven countries, Denmark, Germany, and The Netherlands have in place specific policies to discourage or to alert them to 'overconsumption' of antimicrobials.
Veterinarians are allowed to sell antimicrobials in Belgium, France, Germany, Poland, The Netherlands, and Switzerland.

Discussion
The EC Decision 652/2013 on the monitoring and reporting of antimicrobial resistance in zoonotic and commensal bacteria is binding on all the EU countries. 14 Thus, it was not surprising that we found few differences in the way countries are monitoring the level of resistance. There are, however, several aspects that this policy does not cover, and these should be addressed in future policies.
Current monitoring activities are mostly focused on food-producing animals. Aquaculture is not considered, as this is not a developed sector in many EU countries. Recent reports 15 suggest that policymakers should, in the near future, pay particular attention to aquaculture.
Mandatory monitoring is performed mostly using fecal material. Most of the 'prudent use guidelines' in veterinary medicine recommend a good diagnosis, antibiogramme (a test to evaluate in a microorganism the resistance or susceptibility to a specific antibiotic) use, and epidemiological knowledge of animal disease.
No strong regulations have been established to support best practice in laboratories dedicated to the testing of microbial agents of significance in veterinary medicine. With only fecal matter being tested, resistance-developing microorganisms that live in different organs (lungs, mammary gland, uterus, etc.) might be missed. Direct sampling and analysis of food samples is not mandatory. Including such analyses in the routine European mandatory activities could place strains on finances and on the workforce. Possible cross-contamination would have to be carefully taken into account.
Despite progress in the past few years, it would help to clarify, and above all quantify, the different transmission pathways for AMR between humans and animals. Significant knowledge gaps remain. Exposure to metal compounds such as silver oxide and zinc oxide may contribute to AMR selection, 20,21 but current policies do not mandate monitoring resistance to these agents. Antimicrobial-resistant bacteria have been repeatedly found in ''environmental'' samples, [7][8][9] but despite this, the monitoring of resistance seen in ''environmental'' samples remains voluntary. Such information would help explain the spread of resistance between humans and animals and in the environment.
Most current monitoring activities are 'active' as opposed to 'passive' collection of information. They are managed and funded by the public authorities of each country. As antimicrobials are a public good, these approaches make sense and facilitate both harmonization and transparency of methods. But it is private industry/owners that use antimicrobials (and benefit from them). They are the ones most directly affected by all the adverse consequences of resistance, both in humans and in animals. Increased involvement of private industry in monitoring would surely be desirable.
Monitoring the use of antimicrobials is very different and probably explained by the absence of any binding European policy that requires countries to report their use of antimicrobials in the animal sector. The legal framework for veterinary medicinal products, currently under revision, may offer an opportunity. Reporting sales will likely become mandatory, with mandatory monitoring to follow in 2 to 3 years.
The ESVAC project has contributed to the collection of harmonized data. But most of these data reflect sales, not use. This may change in the relatively near future. 16 Collecting real use data at the farm level is a demanding task that requires several agencies to be involved. Yet, at the same time, it is the most accurate way. An antibiotic sold is not an antibiotic used, and only the recording of actual use will avoid need for approximations, corrections, and relying on other indicators of use.
Knowing the animal population at risk, i.e., the denominator for antimicrobial use, is critical. In the countries we analyzed, the major food-producing animal populations are known (poultry, cattle, and pigs), but this is not always true for other 'minor' species. The use or sales of antimicrobials is reported in such a way that it does not always allow for differentiation between species, types, and stages of production. This is critical for identifying the species and stages in the food chain where prevention and control measures should be focused. Currently, the same antibiotic product can be commercialized for multiple animal species. This creates an additional hurdle when it comes to quantifying its use in a specific species.
Denmark has recognized the usefulness of having a ''yellow card'' policy. 17 Under this policy, a farmer receives a yellow card if he or she uses antimicrobials in a quantity two times higher than the national average. Such policies exist only in Denmark and The Netherlands. Similarly, policies that restrict sales of antimicrobials by veterinarians have had positive impact, 19 but such policies exist only in half of the analyzed countries.

Conclusion
European policies on the monitoring of antibiotic resistance in bacteria in animals currently provide specific guidelines for food-producing animals. However, most of the analyses call for fecal samples. Companion animals, food, and the environment are not even considered. These important gaps should be addressed in future policies. In another failing, no European policy requires countries to report their use of antimicrobials in animals, and most available data are based on sales, and not use.
Considering the universally recognized dimensions of the AMR problem, change is urgently needed. The 2006 EU ban on the use of growth promoters provided a strong global message, and the EU is recognized as an AMR best practices region (Martial Plantady, European Commission, Directorate General for Health and Consumer Protection. B-1049 Brussels. Belgium. 2016, personal communication). Developing and implementing the policies suggested herein should inform policy development in other parts of the world where similar activities may still be lacking.