Probing Automated Treatment of Urinary Tract Infections for Bias: A Case-Study Where Machine Learning Perpetuates Structural Differences and Racial Disparities


Urinary tract infections (UTI) are a common indication for antibiotic treatment worldwide. Fluoroquinolone antibiotics are widely prescribed for these infections, despite being considered ‘second-line’ treatments. The use of these therapies contributes to increased levels of antibiotic resistance. A recent paper [1] described a machine learning system to recommend the narrowest antibiotic predicted to be appropriate for an individual’s UTI. Such data-driven techniques integrated with clinical decision support may play a role in antibiotic stewardship and slow the onset of resistance.Decision making algorithms may inadvertently contain bias that should be vetted before implementation. Prior work has found unintended discriminatory practices in widely used healthcare algorithms [2]. The UTI treatment system (and the data used to develop it) was investigated for potential bias.

Proceedings of the 12th ACM Conference on Bioinformatics, Computational Biology, and Health Informatics
Vincent J. Major
Vincent J. Major
Assistant Professor

Vincent J. Major, PhD is an Assistant Professor at NYU Grossman School of Medicine working on applied machine learning for healthcare using electronic health record (EHR) data.