Machine Learning Model for Predicting Antibiotic Resistance in the Emergency Department in Patients With Urinary Tract Infection
Abstract
Abstract Background: Increasing antimicrobial resistance in uropathogens is a clinical challenge to emergency physicians as antibiotics should be selected before an infecting pathogen or its antibiotic resistance profile is confirmed. This single-center retrospective study evaluated patients diagnosed with upper urinary tract infection (UTI) in the emergency department (ED) between January 2020 and June 2021 to develop a predictive model for antibiotic resistance of uropathogens, using machine learning (ML) algorithmsResults: Forty variables were used to train the model for predicting resistance to ciprofloxacin and the presence of ESBL of urinary pathogens. Model performance with the XGBoost Gradient-Boosted Decision Tree (GBDT) was compared with that of empirical treatment (ET) according to effectiveness of antibiotic selected and appropriateness of selection with respect to antibiotic stewardship. Two prediction models using different decision thresholds were developed. The probability of using ineffective antibiotics in the ED was significantly lowered by 13.3% in the GDBT 0.25 using a decision threshold of 0.25 than in the ET model. Further, the rate of appropriate use of narrow-spectrum antibiotics was higher by 7.4 times in the GBDT 0.44 model than in the ET model, while the ineffectiveness level was maintained.Conclusions: An ML model is potentially useful for predicting antibiotic resistance, improving the effectiveness and appropriateness of empirical antimicrobial…
Read the paper
Related studies
- Horizontal gene transfer in the human gastrointestinal tract: potential spread of antibiotic resistance genes
- Mutation supply and relative fitness shape the genotypes of ciprofloxacin-resistant Escherichia coli
- Genomic Survey of E. coli From the Bladders of Women With and Without Lower Urinary Tract Symptoms
- Characterization of cefotaxime-resistant urinary Escherichia coli from primary care in South-West England 2017–18
- Fitness cost constrains the spectrum of marR mutations in ciprofloxacin-resistant Escherichia coli
- Emerging evidence-based innovative approaches to control catheter-associated urinary tract infection: a review
Go deeper in Evidence Hub
Evidence Hub adds a study data sheet (design, sample size, evidence level, key numbers), an AI summary and a chat grounded in the paper's text.