Policy Webinar 5.
One Health and AMR Surveillance:
Approaches and Options
Professor Sabiha Essack, Senior Implementation Research Advisor to the International Centre for Antimicrobial Resistance Solution (ICARS), introduced the first presenter Professor Frank Møller Aarestrup. Professor Aarestrup is a professor at the Technical University of Denmark and Head of Division at the National Food Institute. His research primarily targets the association between the use of antimicrobial agents to farm animals and the emergence and spread of AMR in humans, including on how next generation sequencing can contribute to the global surveillance of AMR and other pathogens.
“When it comes to AMR, there’s also very often a case of resistant genes and resistant bacteria in all kinds of reservoirs in healthy humans that’s not become recognized by this narrow-minded focus on clinical infections at hospitals. There is all the evolution that has taken place in wildlife, in livestock transmissions, into healthy populations, and we have very limited understanding of what’s actually happening there. And all of these things are important if we really want to tackle and control the problem of AMR.”
Prof. Aarestrup emphasized the need for real-time data on occurrences of all infectious agents of AMR. The ‘next generation sequencing’ (NGS) offers advantages, including providing a universal language that enables and encourages sharing and analysis of raw data, and with capacity developing rapidly improving.
“We need to get a little bit away from this traditional way of doing research. If we could actually start generating the data more locally, and also in the front line, could it then be a way of building capacity for everybody in the world? So they can take care of exploiting the data, but also using it on a daily basis in real life.”
Prof. Aarestrup described the development of an online tool for identifying resistant and other genes, and how data can be mapped to a ‘Resfinder’ database of known genes for analysis of isolates. He then described how it’s difficult to get some samples from healthy humans, so one of the things his team are looking at is sewage. They collect and analyse it using a technology called metagenomics sequencing and surveillance, whereby a sample is taken and small fragments of DNA are randomly sequenced to find out what is in the sample: viruses, bacteria, parasites, etc. The research then used machine learning to look at the explanatory variables (World Bank data, i.e. 1,500 variables) in terms of AMR around the world, rather than focusing just on antimicrobial use (AMU).
“…. the mortality rate, a risk of maternal death, open defecation: it certainly makes sense that all this will increase transmission of AMR and would also lead to more AMR. Whereas if you invest more in water and sanitation, you would actually get less……we do conclude that abundance of AMR is mainly associated to socio-economic factors, and not so much to antimicrobial use.”
In reply to questions from participants, Prof. Aarestrup highlighted that although metagenomic sequencing may appear to be difficult, it is actually easier to standardize than other approaches, for example, it took just three days to set up sequencing in a pilot project in Tanzania.
Dr Essack introduced the second presenter Professor Thomas Van Boeckel, a spatial epidemiologist at ETH Zurich. Professor Van Boeckel’s research includes developing maps of AMR and explores economic incentives to reduce AMU in animals. The work aims to inform policy makers to address the rise of drug resistant pathogens in animals, and is pioneering the development of a platform to centralize epidemiological data on AMR.
“The first question, perhaps, is why animals….it’s worth reminding that most antibiotics on this planet are actually used in the animals that we raise for food.”
Professor Van Boeckel’s research produces maps on a global scale (now on the third iteration), with data currently available from about 40 countries. In 2017, about 90,000 tons of antimicrobials in animals were being used, with China the highest user. On current trends, by 2030, Asia will use more than two thirds of antimicrobials globally, while Africa will only represent about 6 percent of global consumption.
“In low- and middle-income countries there’s a very rapid growth of the livestock sector, there’s hardly any publicly available surveillance that is done systematically in those systems…. so it may take quite a bit of time to get there.”
Hence, alternatives to systematic surveillance systems for animals in low- and middle-income countries (LMICs) was needed. The approach of DTU was to take advantage of largely untapped sources of information from ‘point prevalent surveys’. These are done around the world by microbiologists, veterinarians, etc. to find out what resistant bacteria are present or to look for new resistant genes: on the farm, in food collected in the market, and from samples taken at slaughterhouses. Then the data is synthesized and trends identified. They found that the number of surveys on AMR in animals is increasing, including LMICs. The research then tried to build a metric to summarize trends of resistance across multiple bug-drug combinations.
“We try to be as open as possible with the data, because ultimately, this is your data: many people in low- and middle income countries do the groundwork, so we want to put your name on the map…. you can access all the data, the resistance rate for individual drugs, but we also offer the possibility to upload your own data if you want to participate in this exercise of building a stronger evidence base for antibiotic resistance in animals.”
Prof. Van Boeckel summarized some of the measures that can be taken to reduce AMR in animals: it takes time for farmers to change, so it is essential to start early in providing insights; to accompany farm-level effort by national programmes to try to eradicate animal disease, and limit the need for antimicrobials; and to substantially improve hygiene and biosecurity on farm and vaccination programmes. On a more global level, measures can include: regulations on setting targets for the reduction of AMU in animals (although given that the projected consumption in Africa will be just 6% by 2032, there would be a case to keep some countries out of a global agreement); meat consumption reduction initiatives, including the use of growth promoters; taxes on antimicrobials; and/or export/ trade restrictions.
Further issues include the role of the private sector in making data on AMU available; and the role of retailers, and of veterinarians in low- and middle-income countries, including conflicts of interest, such as often giving medical advice, but without the appropriate medical training, and conflicts of interest of vets in high-income countries.
Professor Essack summarized some of the key takeaways from the presentations and Q and A, including the importance of metagenomic surveillance, which when applied to sewage, can serve as an early warning system for detecting potential escalation of AMR and identify areas for intervention; the significant AMU in food animals and incentives that can reduce it; the importance of generating quality data and of sharing the data, to address AMR; and in ensuring that we have reliable networks that work together constructively and collaboratively to address AMR.
Policy Webinar 5.
One Health and AMR Surveillance:
Approaches and Options
Professor Sabiha Essack, Senior Implementation Research Advisor to the International Centre for Antimicrobial Resistance Solution (ICARS), introduced the first presenter Professor Frank Møller Aarestrup. Professor Aarestrup is a professor at the Technical University of Denmark and Head of Division at the National Food Institute. His research primarily targets the association between the use of antimicrobial agents to farm animals and the emergence and spread of AMR in humans, including on how next generation sequencing can contribute to the global surveillance of AMR and other pathogens.
“When it comes to AMR, there’s also very often a case of resistant genes and resistant bacteria in all kinds of reservoirs in healthy humans that’s not become recognized by this narrow-minded focus on clinical infections at hospitals. There is all the evolution that has taken place in wildlife, in livestock transmissions, into healthy populations, and we have very limited understanding of what’s actually happening there. And all of these things are important if we really want to tackle and control the problem of AMR.”
Prof. Aarestrup emphasized the need for real-time data on occurrences of all infectious agents of AMR. The ‘next generation sequencing’ (NGS) offers advantages, including providing a universal language that enables and encourages sharing and analysis of raw data, and with capacity developing rapidly improving.
“We need to get a little bit away from this traditional way of doing research. If we could actually start generating the data more locally, and also in the front line, could it then be a way of building capacity for everybody in the world? So they can take care of exploiting the data, but also using it on a daily basis in real life.”
Prof. Aarestrup described the development of an online tool for identifying resistant and other genes, and how data can be mapped to a ‘Resfinder’ database of known genes for analysis of isolates. He then described how it’s difficult to get some samples from healthy humans, so one of the things his team are looking at is sewage. They collect and analyse it using a technology called metagenomics sequencing and surveillance, whereby a sample is taken and small fragments of DNA are randomly sequenced to find out what is in the sample: viruses, bacteria, parasites, etc. The research then used machine learning to look at the explanatory variables (World Bank data, i.e. 1,500 variables) in terms of AMR around the world, rather than focusing just on antimicrobial use (AMU).
“…. the mortality rate, a risk of maternal death, open defecation: it certainly makes sense that all this will increase transmission of AMR and would also lead to more AMR. Whereas if you invest more in water and sanitation, you would actually get less……we do conclude that abundance of AMR is mainly associated to socio-economic factors, and not so much to antimicrobial use.”
In reply to questions from participants, Prof. Aarestrup highlighted that although metagenomic sequencing may appear to be difficult, it is actually easier to standardize than other approaches, for example, it took just three days to set up sequencing in a pilot project in Tanzania.
Dr Essack introduced the second presenter Professor Thomas Van Boeckel, a spatial epidemiologist at ETH Zurich. Professor Van Boeckel’s research includes developing maps of AMR and explores economic incentives to reduce AMU in animals. The work aims to inform policy makers to address the rise of drug resistant pathogens in animals, and is pioneering the development of a platform to centralize epidemiological data on AMR.
“The first question, perhaps, is why animals….it’s worth reminding that most antibiotics on this planet are actually used in the animals that we raise for food.”
Professor Van Boeckel’s research produces maps on a global scale (now on the third iteration), with data currently available from about 40 countries. In 2017, about 90,000 tons of antimicrobials in animals were being used, with China the highest user. On current trends, by 2030, Asia will use more than two thirds of antimicrobials globally, while Africa will only represent about 6 percent of global consumption.
“In low- and middle-income countries there’s a very rapid growth of the livestock sector, there’s hardly any publicly available surveillance that is done systematically in those systems…. so it may take quite a bit of time to get there.”
Hence, alternatives to systematic surveillance systems for animals in low- and middle-income countries (LMICs) was needed. The approach of DTU was to take advantage of largely untapped sources of information from ‘point prevalent surveys’. These are done around the world by microbiologists, veterinarians, etc. to find out what resistant bacteria are present or to look for new resistant genes: on the farm, in food collected in the market, and from samples taken at slaughterhouses. Then the data is synthesized and trends identified. They found that the number of surveys on AMR in animals is increasing, including LMICs. The research then tried to build a metric to summarize trends of resistance across multiple bug-drug combinations.
“We try to be as open as possible with the data, because ultimately, this is your data: many people in low- and middle income countries do the groundwork, so we want to put your name on the map…. you can access all the data, the resistance rate for individual drugs, but we also offer the possibility to upload your own data if you want to participate in this exercise of building a stronger evidence base for antibiotic resistance in animals.”
Prof. Van Boeckel summarized some of the measures that can be taken to reduce AMR in animals: it takes time for farmers to change, so it is essential to start early in providing insights; to accompany farm-level effort by national programmes to try to eradicate animal disease, and limit the need for antimicrobials; and to substantially improve hygiene and biosecurity on farm and vaccination programmes. On a more global level, measures can include: regulations on setting targets for the reduction of AMU in animals (although given that the projected consumption in Africa will be just 6% by 2032, there would be a case to keep some countries out of a global agreement); meat consumption reduction initiatives, including the use of growth promoters; taxes on antimicrobials; and/or export/ trade restrictions.
Further issues include the role of the private sector in making data on AMU available; and the role of retailers, and of veterinarians in low- and middle-income countries, including conflicts of interest, such as often giving medical advice, but without the appropriate medical training, and conflicts of interest of vets in high-income countries.
Professor Essack summarized some of the key takeaways from the presentations and Q and A, including the importance of metagenomic surveillance, which when applied to sewage, can serve as an early warning system for detecting potential escalation of AMR and identify areas for intervention; the significant AMU in food animals and incentives that can reduce it; the importance of generating quality data and of sharing the data, to address AMR; and in ensuring that we have reliable networks that work together constructively and collaboratively to address AMR.