Comparing the Performance of Clinical History of MRSA vs. Active Surveillance at Individual and Hospital Ward Levels

Jones MM, Samore MH, Rubin MA, Leecaster M. Comparing the Performance of Clinical History of MRSA vs. Active Surveillance at Individual and Hospital Ward Levels. Presented at: Infectious Diseases Society of America Annual Meeting; 2012 Oct 23; San Diego, CA.

Background: A clinical history of MRSA is widely used to identify patients colonized with MRSA and direct patient isolation; however, its correlation to universal active surveillance data at the individual and hospital ward levels are unclear. Objective: To compare the performance of clinical history vs. active surveillance for predicting MRSA carriage and its impact on colonization pressure estimates. Methods: Microbiology, MRSA surveillance, and admission data from VHA Patient Care Services were analyzed on the VA Informatics and Computing Infrastructure. We analyzed 1,145,101 acute care admissions to wards performing universal active surveillance at 153 hospitals from 10/07-12/09. A history of clinical MRSA in the past year was compared to each surveillance result and sensitivity (SN), specificity (SP), and positive and negative predictive values (PPV and NPV) were calculated. Two estimates of colonization pressure, defined as the ratio of colonized patients to total patients each day on intensive care or other acute-care unit types, were made using clinical history and surveillance data. Daily ward colonization pressures were compared using Pearson's correlation coefficient in the intensive care unit and other acute-care settings. Results: The SN of the clinical history was 22.5%, SP 97.0%, PPV 50.3%, and NPV 90.4%. The mean VA station level colonization pressure was 10.1% calculated by clinical history and 16.9% using surveillance data, with a difference of 6.8%. The correlation of colonization pressure estimates was 0.73 in the ICU and 0.84 in other settings. Conclusion: The SN and PPV of the clinical culture rule are poor compared to universal active surveillance. Over one third of MRSA colonized patients would not be isolated if only the clinical history is used. However, the correlations between the estimates may be sufficiently high to allow clinical history to serve as a surrogate for carriage in certain research settings. Subject Category: N. Hospital-acquired and surgical infections, infection control, and health outcomes including general public health and health services research Speakers: Makoto Jones, MD , Internal Medicine, University of Utah School of Medicine Division of Infectious Diseases, Salt Lake City, UT Michael Rubin, MD , Infectious Diseases, University of Utah, Salt Lake City, UT Tom Greene, PhD , Internal Medicine, IDEAS Center, Salt Lake City, UT Molly Leecaster, PhD , Clinical Epidemiology, Salt Lake City VA Healthcare System, Salt Lake City, UT Joshua Spuhl , Clinical Epidemiology, IDEAS Center, Salt Lake City, UT Christopher Nielson, MD , Patient Care Services, Veterans Healthcare System, Reno, NV Martin Evans, MD , University of Kentucky School of Medicine, Lexington, KY Rajiv Jain, MD , VA Pittsburgh Healthcare System, Pittsburgh, PA Matthew Samore, MD , University of Utah School of Medicine, Salt Lake City, UT