Modelling the relationship between antibiotic consumption and resistance in heterogeneous populations
File(s)
Author(s)
Dewé, Tamsin
Type
Thesis
Abstract
Antibiotic resistance occurs when bacteria can withstand treatment with antibiotics designed to kill them or inhibit their growth, and has been identified as a major threat to global health. Mathematical and statistical models can help us understand the epidemiology and evolution of resistance, but there are important features that are not well captured by existing models. In particular, the coexistence of antibiotic-resistant and sensitive bacterial strains at intermediate frequencies is persistently observed across a wide range of bacterial and host populations, resistance mechanisms, and rates of antibiotic treatment.
In this research, an agent-based model of multi-strain bacterial populations was developed, with a novel competition mechanism that enables coexistence of resistant and sensitive bacteria across a wide range of parameters. Results from this model also demonstrated the influence of bacterial population structure, in addition to antibiotic consumption, in shaping resistance in heterogeneous populations. The model was then used to examine the influence of horizontal gene transfer on resistance dynamics, and determine the conditions for sustaining coexistence when resistance is mobile. Likelihood-free inference was used to estimate model parameters, which enabled simulations that mirrored the heterogeneity of an observed bacterial population.
As a counterpart to this model of resistance carriage in commensal organisms, a statistical model of resistance in clinical isolates was compared to antibiotic consumption data from veterinary and medical settings across Europe. Results from this analyses supported the importance of antibiotic consumption as a driver of resistance in organisms causing disease; however, they also showed that this relationship is affected by other factors. Taken together, the results of this research underline the importance of a One Health approach, and good antibiotic stewardship, in controlling antibiotic resistance. However, they also imply that more specific interventions are not necessarily transferable to different contexts.
In this research, an agent-based model of multi-strain bacterial populations was developed, with a novel competition mechanism that enables coexistence of resistant and sensitive bacteria across a wide range of parameters. Results from this model also demonstrated the influence of bacterial population structure, in addition to antibiotic consumption, in shaping resistance in heterogeneous populations. The model was then used to examine the influence of horizontal gene transfer on resistance dynamics, and determine the conditions for sustaining coexistence when resistance is mobile. Likelihood-free inference was used to estimate model parameters, which enabled simulations that mirrored the heterogeneity of an observed bacterial population.
As a counterpart to this model of resistance carriage in commensal organisms, a statistical model of resistance in clinical isolates was compared to antibiotic consumption data from veterinary and medical settings across Europe. Results from this analyses supported the importance of antibiotic consumption as a driver of resistance in organisms causing disease; however, they also showed that this relationship is affected by other factors. Taken together, the results of this research underline the importance of a One Health approach, and good antibiotic stewardship, in controlling antibiotic resistance. However, they also imply that more specific interventions are not necessarily transferable to different contexts.
Version
Open Access
Date Issued
2023-09-01
Date Awarded
2024-03-01
Copyright Statement
Creative Commons Attribution NonCommercial Licence
License URL
Advisor
Croucher, Nicholas
Fraser, Christophe
Sponsor
Wellcome Trust (London, England)
Publisher Department
School of Public Health
Publisher Institution
Imperial College London
Qualification Level
Doctoral
Qualification Name
Doctor of Philosophy (PhD)
Rights Embargo Date
2026-02-28