Background: Urinary tract infections (UTIs) are common. Antibiotic treatment is usually empirical, with the risk of under-treatment of resistant infections.
Objectives: To characterize risk factors for antibiotic-resistant community urine isolates using routine record-linked health data.
Methods: Within the NHS Scotland Infection Intelligence Platform, national surveillance patient-level data on community urine isolates (January 2012-June 2015) were linked to hospital activity and community prescribing data. Associations between age, gender, comorbidity, care home residence, previous hospitalizations, antibiotic exposure and resistant (any antibiotic) or MDR (≥1 antibiotic from ≥3 categories) urinary isolates were quantified using multivariable logistic regression.
Results: Of 40984 isolates, 28% were susceptible, 45% were resistant and 27% were MDR. Exposure to ≥ 4 different antibiotics in the prior 6 months increased MDR risk (OR 6.81, 95% CI 5.73-8.11). MDR was associated with ≥29 DDD cumulative exposure, in the prior 6 months, for any antibiotic (OR 6.54, 95% CI 5.88-7.27), nitrofurantoin (OR 8.56, 95% CI 6.56-11.18) and trimethoprim (OR 14.61, 95% CI 10.53-20.27). Associations persisted for 10-12 months for nitrofurantoin (OR 2.31, 95% CI 1.93-2.76) and trimethoprim (OR 1.81, 95% CI 1.57-2.09). Increasing age, comorbidity, previous hospitalization and care home residence were independently associated with MDR. For resistant isolates the factors were the same, but with weaker associations.
Conclusions: To our knowledge, we have demonstrated, using national capability at scale, the risk of MDR in community urine isolates for the first time and quantified the cumulative and sustained impact of antibiotic exposure. These data will inform the development of decision support tools for UTI treatment.
© The Author 2017. Published by Oxford University Press on behalf of the British Society for Antimicrobial Chemotherapy. All rights reserved. For Permissions, please email: [email protected].