1292. Claims-Based Surveillance Improves Detection of Surgical Site Infections Following Hysterectomy and Colorectal Surgery
Session: Oral Abstract Session: Surgical Site Infections (SSI)
Saturday, October 20, 2012: 8:45 AM
Room: SDCC 32 AB

Background: Surveillance for surgical site infections (SSI) is labor-intensive and subject to bias. Administrative codes have been successfully used to improve capture of SSI following coronary artery bypass grafting and hip arthroplasty. We evaluated the use of claims data to identify SSI following hysterectomy and colorectal surgery.

Methods: Retrospective cohort study at two academic medical centers. Medical records were reviewed for patients identified as having SSIs through routine surveillance following procedures performed between 7/1/2003 and 6/30/2005 plus a random sample of patients from this same period without known SSIs to assess for SSI using CDC criteria and for discharge codes suggestive of SSI. Review results were used to select code combinations with relatively high sensitivity and positive predictive value.

Results: Among 832 hysterectomies, we confirmed SSI for 14 of 18 patients with SSI identified by routine surveillance and for 9 of 344 randomly sampled patients without previously identified SSI. Extrapolating these results to the population yielded an overall SSI rate of 3.0% vs 1.7% based on routine surveillance and a deep or organ/space (D-OS) SSI rate of 1.4% vs 1.0%. Sensitivity was superior using claims-based methods (Figure). Using codes to trigger chart review would identify 1 SSI out of every 2 charts reviewed and 1 D-OS SSI out of every 4 charts.

Among 2,782 colorectal surgeries, we confirmed SSI for 63 of 72 patients with SSI identified by routine surveillance and for 28 of 305 randomly sampled patients without previously identified SSI. Extrapolating these results to the population yielded an overall SSI rate of 9.1% vs 2.3% based on routine surveillance and a D-OS SSI rate of 3.7% vs 1.2%. Sensitivity was superior using claims-based methods (Figure). Using codes to trigger chart review would identify 1 SSI out of every 3 charts reviewed and 1 D-OS SSI out of every 6 charts.

\s

Conclusion: Claims-based screening identified SSI missed by routine surveillance following hysterectomy and colorectal surgery. This method could be used to identify patients likely to have SSIs for targeted review to improve and standardize detection of SSI. Additional research is needed to evaluate the generalizability of the performance of these codes.

Alyssa R. Letourneau, MD1, Michael S. Calderwood, MD, MPH1,2, Susan S. Huang, MD, MPH, FIDSA3, Dale Bratzler, DO, MPH4,5, Allen Ma, PhD5, Richard Platt, MD, MS, FSHEA1 and Deborah Yokoe, MD, MPH, FSHEA6, (1)Department of Population Medicine, Harvard Medical School and Harvard Pilgrim Health Care Institute, Boston, MA, (2)Infectious Diseases, Brigham and Women's Hospital, Boston, MA, (3)Department of Medicine, University of California Irvine, Irvine, CA, (4)College of Public Health, University of Oklahoma Health Sciences Center, Oklahoma City, OK, (5)Oklahoma Foundation for Medical Quality, Oklahoma City, OK, (6)Brigham and Women's Hospital, Boston, MA

Disclosures:

A. R. Letourneau, None

M. S. Calderwood, None

S. S. Huang, None

D. Bratzler, None

A. Ma, None

R. Platt, None

D. Yokoe, None

Findings in the abstracts are embargoed until 12:01 a.m. PST, Oct. 17th with the exception of research findings presented at the IDWeek press conferences.