Quantifying excess patient risk for CLABSIs attributed to multiple central line use and consequential flaws in current performance measurement in the CMS Hospital Inpatient Quality Reporting Program Open Access

Dube, William (Spring 2019)

Permanent URL: https://etd.library.emory.edu/concern/etds/7p88ch58z?locale=en


Although relatively rare, central-line associated blood-stream infections (CLABSI) are an important healthcare associated infection (HAI). As part of the goal to eliminate HAI, they are included in NHSN reporting, and are a performance measure used to determine hospital re-imbursement from the Center for Medicare and Medicaid Services (CMS). Current NHSN methodology for counting central-line days counts a patient with one line the same as a patient with two or more. This fails to adjust for the additional risk of developing a CLABSI from multiple lines.

           We conducted a large, multi-facility retrospective cohort study, including over 52,000 patient admissions and over 61,000 central venous catheters (CVC) from January 2012 to December 2017. We analyzed these data using a propensity score matched logistic regression to compare CLABSI risk in patients with multiple concurrent lines and those with only one line, adjusting for other risk factors.

           We found that having any concurrent CVC periods in an admission was associated with a 65% increase in the risk of developing a CLABSI (when using NHSN central-line days). If central-line days are counted as we recommend, the risk of concurrence becomes insignificant, because it is accounted for by the central-line day metric. In these data, the risk for each additional NHSN CVC day was 6%. In contrast, the per additional day risk with our recommended method was 3%.

           These findings suggest that in order to make CLABSI risk adjustment fairer for facilities who are likely to have patients with more than one CVC at a time, NHSN central-line reporting methodology needs to be changed. 1 patient with n lines should contribute n central-line days.

Table of Contents

Table of Contents

Introduction. 1

Methods 3

Study Population. 3

CVC Characteristics and Presence of Multiple Catheters 3

Propensity Score Matching. 4

Statistical Analyses 4

Results. 6

Full Dataset 6

Matched analysis. 6

Model Results 7

Discussion. 8

References 10

Tables & Figures. 11

Supplement. 24

About this Master's Thesis

Rights statement
  • Permission granted by the author to include this thesis or dissertation in this repository. All rights reserved by the author. Please contact the author for information regarding the reproduction and use of this thesis or dissertation.
Subfield / Discipline
  • English
Research Field
Committee Chair / Thesis Advisor
Last modified

Primary PDF

Supplemental Files