500. Surveillance for hospital-acquired VRE and MRPA infections using electronic algorithms
Session: Poster Session: Hospital-acquired and Transplant Infections
Friday, October 30, 2009: 12:00 AM
Room: Poster Hall A
Background: Automated surveillance’s strengths lie in reliability and timeliness. We previously built an electronic algorithm (EA) on an information model framework to detect methicillin-resistant Staphylococcus aureus healthcare-associated infections (HAI). We wanted to test its accuracy when adapted for other target organisms.
Methods: We extracted patient census, microbiology, laboratory, vitals, procedure and pharmacy data from the SLC VAMC data warehouse from 10/1/04-10/1/08. By adding modules to our EA’s framework, we adapted it to define HAI due to Vancomycin-resistant Enterococcus (VRE) and multiply-resistant Pseudomonas aeruginosa (MRPA; resistance to ≥3 Pseudomonas spectrum classes of antibiotics).The EA used rules that considered the temporal relations among the data collected. We compared EA to physician chart review (CR) using NHSN (National Healthcare Safety Network) HAI criteria. EA defined episodes as positive cultures within 2 weeks of each other. Each culture queued for chart review, but the unit of analysis was an episode. Sensitivity, specificity and positive predictive value were calculated.
Results: The algorithm was readily adapted to VRE and MRPA due to its modular framework. The EA identified 17 and 10 cases of HAI for VRE and MRPA, respectively. Compared with CR, the sensitivity and specificity were 73.3% and 68.4% for VRE, and 85.7% and 80% for MRPA, respectively. The positive predictive value of EA was 64.7% for VRE and 60% for MRPA.
Conclusion: We demonstrate the generalizability of the EA at the organism level and plan on expanding to other VA centers with further testing. Although more work will be needed to improve performance, EA may eventually be an effective screening tool prior to manual review for surveillance.
Makoto Jones, MD, MS, Patricia Nechodom, MPH, IDEAS Center, Salt Lake City, UT, Michael Rubin, MD, PhD, Matthew Samore, MD, FSHEA, University of Utah School of Medicine, Salt Lake City, UT and  M. M. Jones, None..
M. H. Samore, None..
P. Nechodom, None..
M. Rubin, None.