2951. Nursing home clinician characteristics associated with simulated urinary tract infection treatment decision-making

Abstract

Abstract Background Antibiotic overuse and misuse are common in nursing homes (NHs). Urinary tract infection (UTI) is the most common indication for antibiotics in NHs. We conducted a clinical vignette experiment to identify clinician-level characteristics associated with suboptimal UTI treatment decision-making among a sample of NH clinicians. Methods Six clinical vignettes, two designed to assess clinicians’ treatment threshold and four designed to assess antibiotic choice decision-making, were created through an iterative process. NH clinicians were recruited nationally via professional organizations from December 2021 to April 2022. Clinical vignettes were shared using a web-based survey tool (Qualtrics) during which participants were asked if they would initiate antibiotics and which one. Participants provided demographic information, including degree, specialty training and NH experience. A multivariable logistic regression model with the decision to initiate antibiotic therapy (Yes/No) as the dependent variable and a multivariable ordinal regression model with number of vignettes for which a fluoroquinolone (FQ) was empirically chosen as the dependent variable were constructed in R. Results A total of 298 NH clinicians, whose characteristics are detailed in Table 1, responded to the survey. Despite national guidelines not recommending treatment for asymptomatic bacteriuiria (ASB), 10.4% of participants chose to initiate antibiotics when presented with the ASB vignette. Physicians (vs. APPs) and clinicians with ≥ 10 years NH experience were less likely to initiate antibiotics for ASB (Table 2). Empiric treatment was withheld for simple cystitis by 38.9% of clinicians, with APPs more likely to withhold antibiotics than physicians (Table 2). Across the four cases assessing empiric antibiotic choice, 38.3% of clinicians chose a FQ in at least one clinical case (Table 1). Clinicians with more years of NH-practice experience were more likely to prescribe a FQ (aOR 2.41, 95% CI 1.47-3.95, p< 0.01, Table 2). Respondent Characteristics Abbreviations: Advanced Practice Provider (APP); Asymptomatic bacteriuria (ASB); Fluoroquinolone (FQ), Nursing Home (NH). Effect of clinician credentials, training, and experience on antibiotic prescribing for urinary tract infection Abbreviations: * reference; Adjusted odds ratio (aOR); Confidence interval (CI); Asymptomatic bacteriuria (ASB); Advanced practitioner (APP) Conclusion Our results indicate that specific clinician characteristics are associated with suboptimal antibiotic prescribing. This suggests that tailoring stewardship interventions to clinician-specific factors may augment ongoing efforts to improve antibiotic use in NHs. Disclosures Lindsay Taylor, MD, MS, Merck: Grant/Research Support Robin Jump, MD, PhD, Merck: Grant/Research Support|Pfizer: Advisor/Consultant|Pfizer: Grant/Research Support