Professor Matthew Holden, University of St Andrews, presents HATUA’s mixed-methods work on AMR in East Africa, showing diverse UTI pathogens, locally circulating and imported E. coli lineages, and how patient behaviour and antibiotic access drive resistance.

Biography

Biography

Matt Holden is Professor of Pathogen Genomics at the University of St Andrews and Head of Pathogen Genomics at Public Health Scotland. With over 25 years of experience in bacterial whole genome sequencing, his work spans outbreak investigation, antimicrobial resistance surveillance, and the integration of genomics into public health. He has led major research consortia including SHAIPI, HATUA, and CARE, and has a national leadership role in implementing genomics for clinical and public health use in Scotland.

Detailed description

Antimicrobial resistance (AMR) poses a growing threat to public health globally, with unique challenges in East Africa due to diverse social, cultural, and environmental factors. The HATUA study adopts a holistic, interdisciplinary approach to investigate the drivers and burden of AMR through the lens of urinary tract infections (UTIs). Integrating clinical, microbiological, genomic, and social science data across Kenya, Uganda, and Tanzania, HATUA reveals complex treatment-seeking behaviours, widespread antibiotic access, and context-dependent resistance patterns. Genomic analysis of over 1,400 UTI isolates highlights significant diversity, with multiple dominant sequence types – including the globally disseminated Escherichia coli ST131. Phylogenetic reconstruction enables us to distinguish locally circulating E. coli lineages from more recent introductions, and to trace the origins of resistance determinants, identifying both endemic emergence and external acquisition. These insights are then contextualised through detailed patient-level behavioural data, revealing how lived experiences and healthcare interactions act as drivers of resistance. HATUA demonstrates the power of mixed-methods research in tackling AMR in resource-constrained settings.   

Getting here