Antimicrobial resistance (AMR) is a leading cause of morbidity and mortality across the life course of humans. However, most AMR research and data presentation ignores variation by age and sex, presenting national or syndrome based resistant prevalence indices. This is despite the huge changes in infection risk, comorbidities, antibiotic and healthcare exposure that happen over the life course. Our work at a population level shows the stark importance of including age in the analysis of AMR dynamics, with different trends of proportion resistant by bacteria-antibiotic combination.

This project will be nested within an MRC CDA fellowship that has begun to address these patterns. One initial hypothesis was that resistance would increase with age – we have not found this for many single bacteria-antibiotic combinations. This PhD would test whether there is an accumulation affect with age: those bacteria that cause infections in older individuals that are resistant are resistant to more antibiotics than those in younger individuals.

Our previous research has also explored rates of resistance gene movement in a key AMR bacteria (Staphylococcus aureus)- pairing this rapid shuffling with the age patterns is a key knowledge gap for AMR.

The objectives of this project will be to:

Quantify the variation in number and type of antibiotic combinations as a patient ages and by sex, to statistically test for age and antibiotic trends
Determine the transmission and evolution rates of resistant gene movement that explain multilevel data (microbiology and ecological patterns)
Develop tools to support clinicians to account for age in empiric prescribing decision making and model potential impact on infection burden
The techniques to be used will be:

Data analysis and regression techniques
Mathematical transmission dynamic modelling to account for potential mechanisms driving the patterns by age and sex
Evolutionary mathematical models to explore resistance gene transfer building on laboratory work in S. aureus to account for ecological patterns seen.
Clinically co-designed software development
Resources:

EARS-NET isolate database with antibiograms, age and sex (3.5million isolates) across Europe for bloodstream infections
Several active hospital collaborations will provide patient level information linked to isolate resistance profiles
Mathematical modelling training and support, and computing cluster within the Centre for Mathematical Modelling of Infectious Diseases (CMMID) at LSHTM
Ongoing research on MRSA resistance movement and hence data on rates
Potential risks :

We have access to the EARS-NET data for the main fellowship but have had to anonymise countries for publication. The risk would be that patterns we find in resistant accumulation can be explained by country-level factors that we may find difficult to publish. However, we can work with the ECDC to explore publishing options.
Existing laboratory data may not provide exactly the data and hence parameters required for this project. Whilst no laboratory work is proposed in this project, there is the possibility that placement in collaborative labs could be done and the experiments (co-culture transfer of resistant work) with S. aureus are relatively cheap.
New supervisory team and hence rhythms of working have not been established. However, GK and LE work on GK’s fellowship on this topic and have a history of successful collaboration. GK and JL have been long term colleagues. All work in London so in-person meetings and discussion will support PhD supervision.
Further reading
Relevant preprints and/or open access articles:
(DOI = Digital Object Identifier)

https://doi.org/10.1371/journal.pmed.1004301
https://doi.org/10.1099/jmm.0.001724