Cookies on this website
We use cookies to ensure that we give you the best experience on our website. If you click 'Continue' we'll assume that you are happy to receive all cookies and you won't see this message again. Click 'Find out more' for information on how to change your cookie settings.

Background: Recent UK antibiotic stewardship policies have resulted in significant changes in primary care dispensing, but whether this has impacted antimicrobial resistance is unknown. Aim: To evaluate associations between changes in primary care dispensing and antimicrobial resistance in community-acquired urinary Escherichia coli infections. Methods: Multilevel logistic regression modelling investigating relationships between primary care practice level antibiotic dispensing for approximately 1.5 million patients in South West England and resistance in 152,704 community-acquired urinary E. coli between 2013 and 2016. Relationships presented for within and subsequent quarter drug-bug pairs, adjusted for patient age, deprivation, and rurality. Results: In line with national trends, overall antibiotic dispensing per 1000 registered patients fell 11%. Amoxicillin fell 14%, cefalexin 20%, ciprofloxacin 24%, co-amoxiclav 49% and trimethoprim 8%. Nitrofurantoin increased 7%. Antibiotic reductions were associated with reduced within quarter same-antibiotic resistance to: amoxicillin, ciprofloxacin and trimethoprim. Subsequent quarter reduced resistance was observed for trimethoprim and amoxicillin. Antibiotic dispensing reductions were associated with increased within and subsequent quarter resistance to cefalexin and co-amoxiclav. Increased nitrofurantoin dispensing was associated with reduced within and subsequent quarter trimethoprim resistance without affecting nitrofurantoin resistance. Conclusions: This evaluation of a national primary care stewardship policy on antimicrobial resistance in the community suggests both hoped-for benefits and unexpected harms. Some increase in resistance to cefalexin and co-amoxiclav could result from residual confounding. Randomised controlled trials are urgently required to investigate causality.

More information Original publication




Journal article




Plos One

Publication Date



Launching Fellow: Ashley Hammond