TL;DR
This study reveals that nearly 14% of community-onset urinary tract infections (UTIs) are treated with antibiotics to which the infecting bacteria are resistant. This "discordant" treatment significantly increases the risk of hospital admissions and reconsultations, highlighting a critical area for improving patient outcomes and antibiotic stewardship.
What Problem Does This Paper Address?
Urinary tract infections (UTIs) are incredibly common, leading to a substantial number of antibiotic prescriptions. While many uncomplicated UTIs might resolve on their own or with symptomatic treatment, the rise of antimicrobial resistance (AMR) poses a significant challenge. When antibiotics are necessary, using ones that the bacteria are resistant to (discordant treatment) can lead to treatment failure, more severe outcomes like pyelonephritis or sepsis, and increased healthcare burden through reconsultations. Existing research often lacks the crucial link between primary care prescribing, secondary care outcomes, and direct microbiological resistance data. This paper aims to bridge that gap by investigating the real-world impact of discordant antibiotic treatment on adverse outcomes in patients with culture-confirmed community-onset lower UTIs (COLUTI).
Methods
The researchers conducted a retrospective cohort study, meticulously linking three key data sources:
- Primary care electronic health records: This provided information on patient demographics, consultations, and antibiotic prescriptions.
- Secondary care data (SUS): This captured hospital admissions and other healthcare utilization.
- Microbiology data: This included results from urine and blood cultures, crucially providing information on bacterial identification and antibiotic susceptibility.
They defined a "discordant" treatment as an antibiotic prescribed within three days of a positive urine culture where the identified organism was resistant to that antibiotic. "Concordant" treatment meant the organism was sensitive. The primary outcome of interest was urinary infection-related hospital admission (UHA) within 30 days of the UTI episode. Secondary outcomes included reconsultation within 30 days. They used multivariable analysis to adjust for various factors, including age, sex, risk factors for complicated UTIs, previous antibiotic use, and comorbidities.
Datasets
The study utilized routinely collected data from:
- Primary care electronic health records: From General Practices in East London, managed by the Clinical Effectiveness Group (CEG).
- Secondary Uses Services (SUS) data: From Barts Health NHS Trust, serving as a comprehensive repository for secondary care data in England.
- Microbiology data: From Barts Health NHS Trust, covering urine and blood cultures.
These datasets were deterministically linked to create a comprehensive patient-level dataset. No new benchmarks were introduced; the study leverages existing, large-scale healthcare data.
Key Insights and Ideas
- The "one in seven" problem: A significant proportion of UTIs are treated with antibiotics that won't work, a finding that underscores the need for better antibiotic selection in primary care.
- Discordance is not benign: The study provides strong evidence that discordant treatment isn't just a minor inconvenience; it directly leads to worse outcomes.
- Higher risk patients are particularly vulnerable: While discordant treatment affects many, the implications for patients already at higher risk (e.g., those with CKD or diabetes) are especially concerning.
- Reconsultation is a major consequence: Beyond hospital admissions, discordant treatment dramatically increases the likelihood of patients needing to see their doctor again, adding to healthcare system strain.
Key Results
- Prevalence of Discordance: 14.1% (1686 out of 11963 episodes) of COLUTI episodes were treated with discordant antibiotics.
- Increased Hospital Admissions: Discordant treatment was associated with more than double the odds of urinary infection-related hospital admission (adjusted OR 2.31, P<0.001). Specifically, 5.2% of discordant episodes led to UHA compared to 2.1% of concordant episodes.
- Skyrocketing Reconsultations: The odds of reconsultation were over eleven times higher with discordant treatment (adjusted OR 11.25, P<0.001). 87.3% of discordant episodes resulted in reconsultation versus 38.5% of concordant episodes.
- Risk Factors for UHA: Chronic kidney disease (CKD) and diabetes mellitus (DM) were independently associated with increased odds of UHA.
- Gender Differences: Female sex was associated with reduced odds of UHA compared to male sex.
Limitations and Future Work
The authors acknowledge several limitations:
- Data representativeness: The cohort consists of patients whose urine cultures were sent from primary care, implying they are already at higher risk than the general UTI population. The findings may not be generalizable to all COLUTI cases.
- Missed healthcare encounters: The study did not capture attendances at urgent care centers or A&E that did not result in admission, potentially underestimating reconsultations.
- Misclassification risk: If patients received concordant treatment in settings not captured by the data, the adverse impact of initial discordant treatment might be underestimated.
- Pregnancy data: The Read code for UTI in pregnancy was rarely recorded, preventing its inclusion as a variable.
- Residual confounding: Despite adjustments, other unmeasured factors could influence outcomes.
- Ethnicity and resistance: Non-white patients were more likely to receive discordant treatment, possibly linked to acquired resistance from travel, an area warranting further research into health inequalities.
Future work could focus on developing and implementing rapid diagnostic tools for AST, improving risk stratification for antibiotic prescribing, and exploring the cost-effectiveness of these interventions, particularly for high-risk patient groups like those with DM, CKD, or older adults.
Summary updated