A single center observational study on emergency department clinician non-adherence to clinical practice guidelines for treatment of uncomplicated urinary tract infections

Abstract Background The Emergency Department (ED) is a frequent site of antibiotic use; poor adherence with evidence-based guidelines and broad-spectrum antibiotic overuse is common. Our objective was to determine rates and predictors of inappropriate antimicrobial use in patients with uncomplicated urinary tract infections (UTI) compared to the 2010 International Clinical Practice Guidelines (ICPG). Methods A single center, prospective, observational study of patients with uncomplicated UTI presenting to an urban ED between September 2012 and February 2014 that examined ED physician adherence to ICPG when treating uncomplicated UTIs. Clinician-directed antibiotic treatment was compared to the ICPG using a standardized case definition for non-adherence. Binomial confidence intervals and student’s t-tests were performed to evaluate differences in demographic characteristics and management between patients with pyelonephritis versus cystitis. Regression models were used to analyze the significance of various predictors to non-adherent treatment. Results 103 cases met the inclusion and exclusion criteria, with 63.1 % receiving non-adherent treatment, most commonly use of a fluoroquinolone (FQ) in cases with cystitis (97.6 %). In cases with pyelonephritis, inappropriate antibiotic choice (39.1 %) and no initial IV antibiotic for pyelonephritis (39.1 %) where recommended were the most common characterizations of non-adherence. Overall, cases of cystitis were no more/less likely to receive non-adherent treatment than cases of pyelonephritis (OR 0.9, 95 % confidence interval 0.4–2.2, P = 0.90). In multivariable analysis, patients more likely to receive non-adherent treatment included those without a recent history of a UTI (OR 3.8, 95 % CI 1.3–11.4, P = 0.02) and cystitis cases with back or abdominal pain only (OR 11.4, 95 % CI 2.1–63.0, P = 0.01). Conclusions Patients with cystitis with back or abdominal pain only were most likely to receive non-adherent treatment, potentially suggesting diagnostic inaccuracy. Physician education on evidence-based guidelines regarding the treatment of uncomplicated UTI will decrease broad-spectrum use and drug resistance in uropathogens