Optimising antibiotic use in humans

Abstract

During the seminar Koen Pouwels will discuss his work on improving our use of existing routinely collated and purposefully collected data to better inform decision-making around antibiotic prescribing and resistance. The talk will build on his work on using efficient analytical approaches to estimate prevalence of infectious diseases and resistance while accounting for non-representativeness in the underlying data, spatiotemporal modelling to estimate relationships between antibiotic use and resistance, causal inference methodology to estimate the impact of infections on hospital length of stay, and discrete choice experiments to inform the design of antibiotic stewardship trials. 

BIO

Koen joined Oxford Population Health's Health Economics Research Centre (HERC) as a senior researcher in January 2019. His current work focuses on infectious disease and antimicrobial resistance modelling, randomised trial designs, economic evaluations alongside trials, incorporating long-term effects into economic evaluations, and statistical approaches to address time-dependent confounding in burden of illness studies.

His work has directly informed national COVID-19 mitigation and testing strategies, vaccination policies against various infectious diseases, as well as national targets for antibiotic prescribing in primary care.

Prior to joining HERC, Koen worked at Public Health England on projects using mathematical, statistical and machine learning approaches to understand and predict the development and health-economic impact of healthcare associated infections and antimicrobial resistance. Before coming to the UK, he obtained his PhD in epidemiology at the University of Groningen where he was also involved in a number of economic evaluations of vaccinations against infectious diseases.

During the seminar Koen Pouwels will discuss his work on improving our use of existing routinely collated and purposefully collected data to better inform decision-making around antibiotic prescribing and resistance. The talk will build on his work on using efficient analytical approaches to estimate prevalence of infectious diseases and resistance while accounting for non-representativeness in the underlying data, spatiotemporal modelling to estimate relationships between antibiotic use and resistance, causal inference methodology to estimate the impact of infections on hospital length of stay, and discrete choice experiments to inform the design of antibiotic stewardship trials.