1072. Comparison of NHSN-Defined Central Venous Catheter Day Counts with a Method that Accounts for Concurrent Catheters
Session: Poster Abstract Session: Surveillance of HAIs: Evaluating National Strategy
Friday, October 4, 2013
Room: The Moscone Center: Poster Hall C
  • Central Lines Days Talbot IDWeek 2013 Graph Version.pdf (65.2 kB)
  • Background: The current National Healthcare Safety Network (NHSN) definition for central venous catheter (CVC) days does not account for the presence of multiple concurrent CVCs, which may indicate an increased risk for developing a central line-associated bloodstream infection (CLABSI).  We used an electronic surveillance system to compare CVC day counts calculated using the NHSN definition (≥1 CVC in place = 1 CVC day, “conventional method”) to counts that accounted for concurrent CVCs (x concurrent CVCs in place = x CVC days, “unique method”). 

    Methods: For all inpatient units at an academic medical center, CVC days using CVCs documented in place at midnight each day for the years 2010 and 2011 were calculated via electronic nursing documentation.  Differences in conventional counts and unique counts were assessed, including comparisons of intensive care units (ICUs) vs. non-ICUs and adult vs. pediatric units.  CLABSI events were determined by trained infection preventionists using NHSN definitions.  Units with an annual CLABSI rate greater than zero were ranked into CLABSI rate decile strata, and the impact of accounting for multiple CVCs on unit rankings was assessed.

    Results: For all units combined, mean monthly CVC day counts increased using the unique method by 8.5% (conventional method: 7093.9 [95% CI 6942.0-7245.8] CVC days vs. unique method: 7693.5 [95% CI 7522.5-7864.6] CVC days for the study period). The percent change in CVC days was greater among ICUs (12.7% increase) compared with non-ICUs (4.6%), and among pediatric units (9.5%) compared to adult units (7.9%). The pediatric ICU was impacted the greatest, with total CVC days increasing 23.8%.  CLABSI rates were mildly impacted, with 81.3% of units with at least 1 CLABSI event during 2010 having no change in the 2010 CLABSI rate decile and 82.6% of units with at least 1 CLABSI event during 2011 having no change in the 2011 CLABSI rate decile compared to decile rankings of rates calculated using the unique CVC day count.  All decile ranking changes were within 1 decile (e.g. a change from decile rank 5 to 6). 

    Conclusion: Changing to a CVC day definition that accounts for multiple concurrent CVCs increased device day counts and mildly impacted CLABSI rate rankings. ICUs were particularly affected, but the degree of impact varied by ICU type.

    Thomas Talbot, MD, MPH, FIDSA, FSHEA1,2, James Johnson, MD MPH1,2, Rachel Hayes, RN PhD3 and Theodore Anders, MBA4, (1)Medicine, Vanderbilt University School of Medicine, Nashville, TN, (2)Infection Control and Prevention, Vanderbilt University Medical Center, Nashville, TN, (3)Surgical Sciences, Vanderbilt University School of Medicine, Nashville, TN, (4)Informatics Center, Vanderbilt University Medical Center, Nashville, TN


    T. Talbot, Joint Commission Resources: Consultant, Consulting fee
    Community Health Systems: Consultant, Consulting fee

    J. Johnson, None

    R. Hayes, None

    T. Anders, 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.