Performance of the TREAT decision support system in an environment with low prevalence of resistant pathogens

Abstract number: O399

Zalounina A., Kofoed K., Paul M., Lisby G., Leibovici L., Andreassen S., Andersen O.

Objectives: The decision support system for antibiotic treatment TREAT [1] has been shown to improve appropriateness of antibiotic therapy and reduce cost in regions with intermediate or high prevalence of resistant bacterial strains. The purpose of this study was to explore if TREAT can achieve similar improvements in Denmark, which has a low prevalence of resistant bacterial strains.

Methods: A retrospective trial of TREAT has been performed at Copenhagen University, Hvidovre Hospital. The system was calibrated by local data (e.g., distribution of pathogens, resistance to antibiotics, various administrative factors). The study was based on a database with detailed clinical data on adult patients with suspicion of moderate to severe infections, collected in 2005–2006. The data included risk factors for infections and pathogens, clinical and microbiological data, physicians chosen therapy and results for susceptibility tests of isolated pathogens. TREAT was tested empirically, i.e. in all cases (with exception of 2) the morphology/identity of the isolate was neither known to the physician nor to the system. Coverage (defined as the percentage of antibiotic treatment matching the susceptibility of isolated pathogen) and cost of treatment obtained by TREAT were compared to clinical practice.

Results: Out of 171 patients in the database, 161 fulfilled the inclusion criteria previously applied in clinical trials of TREAT. Significant clinical isolates were found in 65 (40%) cases, among them 25 isolates from the blood. Coverage achieved by TREAT in 65 patients with significant clinical isolates was 86%, while coverage achieved by the first attending clinical physician was 66% (OR 3.2, 95%CI 1.3–7.6, p = 0.009). The mean costs (in Euro) per episode for TREAT were 76 for direct expenses for antibiotics and administration, 96 for side effects, 310 for future resistance; and in clinical practice 54, 126 and 289, respectively.

Conclusion: Coverage achieved by TREAT was significantly higher than coverage achieved by the physician. The costs obtained by the system are lower in regard to side effects, and higher for direct expenses and the pressure for future resistance. These results suggest that TREAT can markedly improve appropriateness of antibiotic therapy and reduce cost for side effects in regions with low prevalence of resistant pathogens.


1. The TREAT Working Group. J Antimicrob Chemother 2006; 58: 1238–45.

Session Details

Date: 19/04/2008
Time: 00:00-00:00
Session name: 18th European Congress of Clinical Microbiology and Infectious Diseases
Location: Barcelona, Spain
Presentation type:
Back to top