1088. Automated Surveillance for Central venous catheter Related Bloodstream Infections in a Korean hospital
Session: Poster Abstract Session: Surveillance of HAIs: Implementation and National Perspectives
Friday, October 4, 2013
Room: The Moscone Center: Poster Hall C

Background:

Manual surveillance for central venous catheter related bloodstream infection (CRBSI)s is labor intensive and time-consuming. In addition, it has often interobserver differences. To solve these problems, we made an automated surveillance program using existing databases with patient-level variables and microbiology data in the hospital`s electronic healthcare record system. We report our project to evaluate electronic surveillance of CRBSI.

Methods:

All patients who have been in ICUs at Gangnam Severance Hospital between January 1, 2011, and December 31, 2011, were evaluated. The charts of patients who had positive blood culture were reviewed using the NHSN definitions. We made true CRBSI lists from chart reviewing. The hospital has reported nosocomial infections of intensive care units to Korean nosocomial infection surveillance system (KONIS). Past manual surveillance results were obtained from KONIS. We created electronic surveillance algorithms by translating existing manual surveillance practices into automated electronic methods. The results of automated surveillance were obtained from that algorithm. We compared these 3 lists.

Results:

There were 35 CRBSI cases in manual surveillance. According to chart review, 4 were not true CRBSI. Nineteen cases were existed in automated surveillance. Eleven cases were not found because data including body temperature, systolic blood pressure and catheter insertion were missed. One case was excluded because same pathogen was identified on sputum culture. There were 55 true CRBSI cases. There were 54 CRBSI cases in automated surveillance. According to chart review, 35 cases were true CRBSIs. Twelve cases were not found because data were missed. Eight cases were excluded because same pathogen was identified on other cultures. If all of data were collected properly, automated surveillance would find forty-seven true CRBSIs. It would be better than manual surveillance.

Conclusion:

In this study, automated surveillance with accurate information has good performance. If data are recorded appropriately, automated surveillance would replace manual surveillance totally in the future.

 

Ji Hyun Yoon1, Soowan Park2, Nan-Hyoung Cho3, Mina Kim3, Sung Joon Jin1 and Young Goo Song1,3, (1)Departement of Internal Medicine, Gangnam Severance Hospital Yonsei University College of Medicine, Seoul, South Korea, (2)Department of Medical & Informatics, Gangnam Severance Hospital Yonsei University College of Medicine, Seoul, South Korea, (3)Department of Infection Control, Gangnam Severance Hospital Yonsei University College of Medicine, Seoul, South Korea

Disclosures:

J. H. Yoon, None

S. Park, None

N. H. Cho, None

M. Kim, None

S. J. Jin, None

Y. G. Song, 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.