385. Development and Validation of a Simple and Easy-to-Employ Electronic Algorithm for Identifying Clinical MRSA Infection
Session: Poster Abstract Session: MRSA, MSSA, Enterococci
Thursday, October 3, 2013
Room: The Moscone Center: Poster Hall C
Background:  With growing demands to track and publicly report and compare infection rates, efforts to utilize automated surveillance systems are increasing.

Objectives: We sought to develop and validate a simple algorithm for identifying patients with clinical MRSA infection using microbiologic and antimicrobial prescribing data. We also estimated resource savings.

Methods:  Patients receiving care at any of 5 acute care VA medical centers who had a positive culture for MRSA either as an inpatient or outpatient were eligible. A random selection of positive cultures was reviewed manually, and patients were classified as having a clinical infection based on standard definitions. In our electronic algorithm, patients were defined as having a clinical MRSA infection if they had a positive sterile-site culture or a positive non-sterile site culture with receipt of MRSA-active antibiotics +/- 5 days of the culture date. The optimal electronic definition was determined using test characteristics. Time and cost for chart review was calculated.

Results:  246 unique non-sterile site cultures were included. Of these, 168 represented infection as determined by clinical review.  The sensitivity and specificity of the electronic algorithm varied from 43.4%-95.8% (sensitivity) and 34.6%-84.6% (specificity), depending upon the combination of antimicrobials included in the definition.  The optimal electronic algorithm included clindamycin, daptomycin, doxycycline, linezolid, trimethoprim-sulfamethoxazole, and vancomycin and classified 178 patients as clinical MRSA infection. On multivariable analysis, predictors of algorithm failure included outpatient status at the time of culture collection (OR 0.23, 95% CI 0.10-0.56) and respiratory culture (OR 0.29, 95% CI 0.13-0.65). The median cost was calculated to be $2.43 per chart given 4.6 minutes review-time per chart.

Conclusion:  Our simple electronic algorithm for detecting clinical MRSA infections has excellent sensitivity and good specificity. In settings with a robust electronic medical record, implementation of this electronic system may streamline and standardize surveillance and reporting efforts. Almost 20 hours of review time per 250 charts could be re-allocated towards prevention activities.

Westyn Branch-Elliman, MD, MMSc1,2,3, Judith Strymish, MD3,4 and Kalpana Gupta, MD, MPH4,5, (1)Divisions of Infectious Diseases and Infection Control, Beth Israel Deaconess Medical Center, Boston, MA, (2)Infectious Diseases, VA Boston HCS, West Roxbury, MA, (3)Medicine, Harvard Medical School, Boston, MA, (4)VA Boston HCS, West Roxbury, MA, (5)Department of Medicine/Boston University School of Medicine, Boston, MA


W. Branch-Elliman, None

J. Strymish, None

K. Gupta, None

Findings in the abstracts are embargoed until 12:01 a.m. PST, Oct. 2nd with the exception of research findings presented at the IDWeek press conferences.