Intensive antibiotic prescribing and its association with resistant urinary tract infections (UTIs) in the community: a cross-sectional ecological UK primary care-based study

Problem

Antimicrobial resistance (AMR) has been identified by the World Health Organisation (WHO) as one of the principal public health problems of the 21st century. Overuse of antibiotics is associated with antibiotic resistance in the community, with primary care practices that prescribe more antibiotics associated with higher antibiotic resistance. There is variation, however, in the distribution of antibiotic prescribing among patients in the community. Previous UK primary care-based research demonstrated that less than 10% of patients received 50% of antibiotic prescriptions, with those patients receiving at least 5 antibiotic prescriptions over 3 years.It is unclear whether variations in the distribution of antibiotics in the community affect local resistance. This community-based study aims to determine the association between (a) overall antibiotic prescribing and UTI resistance, and (b) repeat (intensive) prescribing of antibiotics to individual patients (> 4 antibiotic prescriptions per year) and UTI resistance.

Approach

A repeated cross sectional ecological analysis from 2012-2015, using antibiotic prescribing data from Lambeth DataNet (an anonymised source of coded data from all general practices in Lambeth) and urine culture results analysed by the microbiology department at Guys and St Thomas’ NHS Trust and University Hospital Lewisham. We determined the total volume of antibiotics prescribed (number of prescriptions/Average Daily Quantity (ADQ)) and the percentage of patients who received repeat (intensive) antibiotic prescriptions per Lower Superior Output Area (LSOA)/year. Following adjustment for demographic and clinical variables, LSOA level associations will be examined between overall prescribing and intensive prescribing, with same and next year UTI resistance rates, using Poisson regression modelling over 4 years.

Findings

Interim results available with analysis underway. The sample includes 294,100 patients (covering 178 LSOAs) and 553,782 antibiotic prescriptions from 2012-2015. Overall UTI annual resistance rates over the study period ranged from 64.7% to 71.5%. 8.5% of the population fulfilled the explanatory variable of intensive antibiotic prescribing (≥4 antibiotic prescriptions/year). Preliminary Poisson regression modelling for 2013 suggests that intensive antibiotic prescribing (≥4 antibiotic prescriptions/year) was independently associated with increased rates of antibiotic resistance in community UTIs. The findings of more detailed regression modelling will be available for the SAPC conference.

Consequences

Reducing AMR is a global priority. There has been very limited previous research regarding the distribution of antibiotic use and its association with resistance, with conflicting interpretations of the results. The results from this research will help determine whether efforts to reduce resistance should focus on reducing antibiotic prescribing in the community overall, or reducing intense prescribing to individual patients.