Antimicrobials and antimicrobial resistance genes in a one-year city metabolism longitudinal study using wastewater-based epidemiology

Natalie Sims, Andrew Kannan, Elizabeth Holton, Kishore Jagadeesan, Leonardos Mageiros, Richard Standerwick, Tim Craft, Ruth Barden, Edward J. Feil, Barbara Kasprzyk-Hordern

Research output: Contribution to journalArticlepeer-review

7 Citations (SciVal)


This longitudinal study tests correlations between antimicrobial agents (AA) and corresponding antimicrobial resistance genes (ARGs) generated by a community of >100 k people inhabiting one city (Bath) over a 13 month randomised monitoring programme of community wastewater. Several AAs experienced seasonal fluctuations, such as the macrolides erythromycin and clarithromycin that were found in higher loads in winter, whilst other AA levels, including sulfamethoxazole and sulfapyridine, stayed consistent over the study period. Interestingly, and as opposed to AAs, ARGs prevalence was found to be less variable, which indicates that fluctuations in AA usage might either not directly affect ARG levels or this process spans beyond the 13-month monitoring period. However, it is important to note that weekly positive correlations between individual associated AAs and ARGs were observed where seasonal variability in AA use was reported: ermB and macrolides CLR-clarithromycin and dmCLR-N-desmethyl clarithromycin, aSPY- N-acetyl sulfapyridine and sul1, and OFX-ofloxacin and qnrS. Furthermore, ARG loads normalised to 16S rRNA (gene load per microorganism) were positively correlated to the ARG loads normalised to the human population (gene load per capita), which indicates that the abundance of microorganisms is proportional to the size of human population and that the community size, and not AA levels, is a major driver of ARG levels in wastewater. Comparison of hospital and community wastewater showed higher number of AAs and their metabolites, their frequency of occurrence and concentrations in hospital wastewater. Examples include: LZD-linezolid (used only in severe bacterial infections) and AMX-amoxicillin (widely used, also in community but with very low wastewater stability) that were found only in hospital wastewater. CIP-ciprofloxacin, SMX-sulfamethoxazole, TMP-trimethoprim, MTZ-metronidazole and macrolides were found at much higher concentrations in hospital wastewater while TET-tetracycline and OTC-oxytetracycline, as well as antiretrovirals, had an opposite trend. In contrast, comparable concentrations of resistant genes were observed in both community and hospital wastewater. This supports the hypothesis that AMR levels are more of an endemic nature, developing over time in individual communities. Both hospital and community wastewater had AAs that exceeded PNEC values (e.g. CLR-clarithromycin, CIP-ciprofloxacin). In general, though, hospital effluents had a greater number of quantifiable AAs exceeding PNECs (e.g. SMX-sulfamethoxazole, ERY-erythromycin, TMP-trimethoprim). Hospitals are therefore an important consideration in AMR surveillance as could be high risk areas for AMR.

Original languageEnglish
Article number122020
Number of pages16
JournalEnvironmental Pollution
Early online date17 Jun 2023
Publication statusPublished - 15 Sept 2023

Bibliographical note

Funding Information:
The support of Engineering and Physical Sciences Research Council (EP/P028403/1), Wessex Water Services Ltd and EPSRC Impact Acceleration Account (Project number: EP/R51164X/1, ENTRUST IAA) is greatly appreciated. The authors would like to thank EPSRC Centre for Sustainable Circular Technologies for their support. The support of Wessex Water is also greatly appreciated.

Data availability: Data is available in SI


  • Antibiotics
  • Antimicrobial resistance genes (ARGs)
  • Antimicrobials (AA)
  • Hospital effluent
  • Wastewater-based epidemiology

ASJC Scopus subject areas

  • Toxicology
  • Pollution
  • Health, Toxicology and Mutagenesis


Dive into the research topics of 'Antimicrobials and antimicrobial resistance genes in a one-year city metabolism longitudinal study using wastewater-based epidemiology'. Together they form a unique fingerprint.

Cite this