Modelling the spatial and temporal epidemiology of antimicrobial resistant Salmonella Typhi and Paratyphi A infections

Thesis

Enteric fever caused by Salmonella enterica Typhi and Paratyphi A infers a large burden of disease in low- and middle-income countries (LMICs). Antimicrobial resistance (AMR) reduces our ability to treat these infections, increasing morbidity and mortality.

The aim of this thesis was to improve our understanding of the spatial and temporal trends in the prevalence of AMR in enteric fever and its drivers, such as antibiotic consumption, in order to guide policy and combat the growing threat of AMR.

A systematic review was performed to collate a contemporary dataset of multidrug resistance (MDR) and fluoroquinolone non-susceptibility (FQNS) in S. Typhi and S. Paratyphi A. A meta-analysis was used to assess broad scale trends in resistance.

Estimates of antibiotic usage in LMICs, and estimates of antibiotic consumption for 204 countries were produced for the years 2000 to 2018 using statistical models. These estimates were then combined with the systematic review data and additional covariates to model the prevalence of MDR and FQNS in S. Typhi and S. Paratyphi A for each endemic country and year from 1990 to 2018.

Large national and subnational variations in antibiotic usage are identified. Antibiotic consumption was estimated to vary almost 10 fold, from 5·0 [4·8-5·3] defined daily doses (DDD) per 1,000 population per day in Philippines to 45·9 DDD/1,000/day in Greece.

The prevalence of MDR S. Typhi decreased between 1990 and 2018 in South and Southeast Asia, but increased greatly in sub-Saharan Africa. FQNS S. Typhi increased considerably in South and Southeast Asia over the study period and increased, albeit to a lesser extent, in sub-Saharan Africa. MDR S. Paratyphi A remained low throughout, but the prevalence of FQNS S. Paratyphi A increased.

This thesis identifies both high rates of antibiotic consumption and a lack of access to antibiotics in various locations. It provides detailed estimates of the prevalence of AMR in enteric fever and analyses the trends over space and time. These estimates can be used as a benchmark from which to measure future interventions and to guide policy implementation to tackle AMR.

Authors: Browne, AJ

• Type of award: DPhil
• Level of award: Doctoral
• Awarding institution: University of Oxford
• Language: English
• Keywords: Spatial modelling; Antibiotic consumption; Antimicrobial resistance; Salmonella Typhi; Machine learning