|
Back to 2014 Annual Meeting Abstracts
Decision and Cost Analysis for Treatment of Uncomplicated E. coli Cystitis
Michal Ursiny, BS1, Cornell T. Cummings, MD 2, Brian H. Eisner, MD 3.
1University of Vermont College of Medicine, Burlington, VT, 2Washington University, St. Louis, MO, 3Massachusetts General Hospital, Boston, MA.
BACKGROUND: Fluoroquinolones are costly and resistance to these effective medications is increasing. The purpose of this study was to se hospital local hospital antibiograms and decision analysis to optimize the cost- effective treatment of uncomplicated E. coli urinary tract infections. The objective was to determine the optimum first-line therapy.
METHODS: Formal requests for antibiograms were made to all Massachusetts Hospitals registered with the Massachusetts Deparment of Public Health. Antibiotic sensitivity profiles were obtained from 18 Massachusetts hospitals to account for hospital to hospital variation. A decision analysis model was created that utilized costs of antibiotic regimens and integrated them with sensitivity data from local hospitals. Expected value calculations were used to determine which antibiotic was cost- effective in treatment and prevention of complications. Pharmaceutical and secondary treatment costs were obtained both from institutional billing, the Red Book, and the available English literature.
RESULTS: Antibiograms for e. coli sensitivity were analyzed from 18 Massachusetts hospitals. When evaluated collectively, median antibiotic resistance was as follows: 23% (TMP/SMX), 22% (Ciprofloxacin), 20% (Levofloxacin), 5% (Nitrofurantoin). Accounting for both cost of treatment and respective antibiotic resistance, the median expected values (with ranges), surrogates of cost-effectiveness, are the following based on the decision analysis: Nitrofurantoin = $60 ($53 - $72), TMP/SMX = $72 ($60 - $82), Ciprofloxacin = $81 ($67 - $95), Levofloxacin for $125 ($113 - $141). Nitrofurantoin is the most cost- effective antibiotic when fluoroquinolone resistance is >13.1% and TMP/SMX resistance is >18%. If TMP/SMX resistance is < 14.8%, it is cost-effective regardless of other antibiotic resistance values. For 9 hospitals with complete antibiotic resistance panels available, nitrofurantoin was the most cost-effective for 8/9 and TMP/SMX was the most cost effective for 1/9.
CONCLUSIONS: These findings suggest that fluoroquinolones may not be the best choice (both from cost and resistance perspective) for uncomplicated cystitis. Use of individual hospital antibiograms may enable physicians to select the most cost- effective antibiotics for the population of their specific institution.
Back to 2014 Annual Meeting Abstracts
|
|
