Urinary tract infections (UTI) are commonly diagnosed and treated in primary care. The gold standard for diagnosing a UTI is a urine culture. However, waiting for culture results delays treatment, so doctors often prescribe antibiotics while awaiting those results. Researchers modified a UTI detection algorithm developed and validated in an emergency room population to be usable in a primary care setting. The main modification was removing the requirements for microscopy since results are often not available in primary care. Researchers found that the removal of microscopy features did not severely compromise performance of the UTI detection algorithm in emergency department patients. Additionally, the algorithm performed well in the primary care sample. Results suggest that the new algorithm could be used to safely withhold antibiotics in low risk patients, thereby reducing antibiotic overuse.
Adaptation and External Validation of Pathogenic Urine Culture Prediction in Primary Care Using Machine Learning
Daniel J. Parente, MD, PhD, et al
Department of Family Medicine and Community Health University of Kansas Medical Center, Kansas City, Kansas
