Methods: Using U.S. commercial insurance claims data (2006-2015), we conducted a retrospective cohort study of non-pregnant women 18-44 years who received an outpatient diagnosis of UTI with a prescription for an antibiotic with activity against common uropathogens. For each antibiotic agent, we estimated the daily cumulative risk and 95% confidence intervals (CIs) of treatment failure defined by a subsequent UTI-related antibiotic prescription since the index prescription. Propensity-score weighting accounted for patient-, geographic-, and provider-level characteristics.
Results: Among 1,100,661 eligible women, the majority received second-line fluoroquinolones (43%), first-line trimethoprim-sulfamethoxazole (28%), or first-line nitrofurantoin (24%). 7-day and 30-day treatment failure occurred in 8.4% (n=92,382) and 20.5% (n=225,746) of women, respectively. Among initiators of first-line agents, the 7-day weighted cumulative incidence estimates of treatment failure were lower for nitrofurantoin (6.0%, 95% CI, 5.9% - 6.1%) vs trimethoprim-sulfamethoxazole (8.8%, 95% CI, 8.7% - 9.0%). Among initiators of second-line agents, treatment failure did not differ between fluoroquinolones (5.0%, 95% CI, 4.9% - 5.1%), narrow-spectrum β-Lactams (5.1%, 95% CI, 4.9% - 5.4%), or broad-spectrum β-Lactams (5.3%, 95% CI, 4.9% - 5.7%). Among initiators of non-guideline recommended β-Lactams, treatment failure was 9.6% (95% CI, 9.0% - 10.3%). Results were similar for 30-day treatment failure, with the exception of lower risk for fluoroquinolones compared to other second-line agents.
Conclusion: The risk of treatment failure differs widely by antibiotic agent, with substantial differences between two first-line agents. Understanding the effectiveness of antibiotic therapy is critical to guide clinical decision making, reduce suboptimal antibiotic prescribing, and prevent antibiotic resistance and other adverse events.
A. M. Butler,
M. J. Durkin, None
V. R. Dharnidharka, None
M. A. Olsen, None