News and Opinions  –  2023

Genomic analysis of sewage from 101 countries reveals global landscape of antimicrobial resistance

Share the article


End last year a new comprehensive study that analyses sewage samples from 234 cities in 101 countries was published. The study highlights the potential of a sewage-based global monitoring system. ReAct reached out to Professor Frank Aarestrup, co-conceiver and co-writer of the study, to learn more about the project and to hear his thoughts on monitoring and surveillance.

Scientist gets waste water for analysis. Photo: Shutterstock.

On December 2022 the Global Sewage Surveillance Project published the results of a new study analyzing 757 sewage samples from 243 cities in 101 countries between 2016-2019. As illustrated with the COVID-19 pandemic, and previously with the poliovirus, this study highlights the potential of a sewage-based global monitoring system as a valuable tool for monitoring the spreading and evolution of antimicrobial resistance (AMR). Not least, a sewage-based monitoring of Antimicrobial Resistance Genes (or ARGs), the “currency” of antimicrobial resistance, could provide further information on the source of contamination and support country-specific antibiotic stewardship measures.

ReAct interviewed Professor Frank Aarestrup, co-author of the paper, Head of Research Group at the DTU National Food Institute, and leader of the Global Sewage Surveillance Project – to learn more about the study and his thoughts.


ARGs = Antimicrobial Resistance Genes (or ARGs) are mobile genetic elements that can transfer between microorganisms and confer resistance, e.g. via horizontal gene transfer.

Resistome = The pool of antimicrobial resistance genes present in the environment.

Metagenomic = The study (by sequencing analysis) of all genetic material present in a specific environmental sample.

3 key facts from the study

  • The resistomes based on sewage-based monitoring showed geographical stability over time
  • The Sub-Saharan Africa sewage samples showed the highest ARG loads
  • Comparability of large datasets is highly connected to centralized sample processing, sequencing, and analysis

Professor Aarestrup, in simple terms, how can a global monitoring system of waste-waters help addressing the antibiotic resistance issue?

Professor Frank Aarestrup. Photo: DTU National Food Institute

– If we do not know where we are, we have no clue where we are going. A global monitoring can provide a baseline for the resistance situation in the different countries and thereby show where interventions are most important. It is also essential to identify drivers for reducing antimicrobial resistance, as well as to show the effect (or lack thereof) of interventions. Without surveillance you are flying blind.

From your monitoring data just published, African states were the most under-sampled. Why? And how can this issue be overcome?

– Correct, but compare that to other data. It is still under-sampled, but no way as much as with the clinical surveillance we currently have, where basically only the EU provides continuous data. More contacts in Africa is the way forward. I knew people in most countries in Europe from my scientific network and that could be used immediately, while it simply takes a bit more time to get to know people in other places.

How can the existing frameworks, such as national action plans, better integrate and implement a sustainable monitoring and surveillance system? What other frameworks are/should be available?

– You need surveillance and political will. Surveillance shows where you are and which way it is going. That provides politicians and authorities with a benchmark and trend. However, there needs to be a will to take action. Reducing AMU, restricting certain classes, improving hygiene etc. Surveillance data is also good because it provides a basis for discussion between countries and people and especially those who have a problem are often more prone to do something.

In your opinion, which are the most important limitations with using this type of surveillance?

– There are three main problems:

1, The sensitivity can be an issue when you use metagenomics since we are not only sequencing resistance genes but everything and thus we might miss genes that occur at low frequencies

2, Context – we simply obtain sequence reads and can not directly link the finding of a resistance gene to a given bacterial species. You can do some in silico modelling to attempt this, but in many (most) cases this can not be done.

3, The complexity of the data. We get so much information that it can become overwhelming.

What are the opportunities to make monitoring data more informative and actionable?

– I think the main opportunity is to share both summary and raw data in as close to real-time as possible from a trusted source. This could be EU, WHO or another place. Also, we need both the detailed data as well as a summary that is easier to interpret for non-scientists. We have ourselves tried this, but people with more expertise could to it better.

More from "2023"