Program Schedule

Excess Length of Stay Attributable to Clostridium difficile Infection (CDI) in the Acute Care Setting: A Multi-State Model

Session: Poster Abstract Session: Clostridium difficile Infection: Epidemiology, Presentation, Treatment
Saturday, October 11, 2014
Room: The Pennsylvania Convention Center: IDExpo Hall BC
  • CDI LOS IDWEEK 2014.pdf (472.9 kB)
  • Background: Evaluation of the economic impact of interventions designed to reduce the incidence of Clostridium difficile infection (CDI) depends on accurate estimates of healthcare resource use attributable to infection.    Studies that do not account for the timing of infection overestimate attributable length of stay (LOS).  The purpose of this study was to evaluate the excess LOS due to CDI using a multi-state model 

    Methods: We conducted a retrospective cohort study of all patients hospitalized on a general medical, surgical, or intensive care unit within the US Department of Veterans Affairs (VA) health care system between January 1, 2005 and December 31, 2012.  A diagnosis of CDI was based on a positive laboratory result by enzyme immunoassay or cytotoxin test.  A multi-state approach, implemented through the etmpackage in R, was used to estimate the excess LOS attributable to CDI while accounting for the timing of infection.  Confidence intervals were estimated using bootstrapping techniques.  A composite outcome of discharge or death was used.  Estimates from the multistate model were compared to traditional estimates, including crude comparisons, ordinary least squares (OLS), and a generalized linear model (GLM) with a log link and gamma distribution

    Results: During the study period, 4.2 million Veterans were followed up for nearly 22 million patient-days of observation.  CDIs were observed in 43,661 (1%) of hospitalizations.  The median LOS among patients with and without CDI was 10 (IQR: 17) and 3 (IQR: 4) days, respectively, for a crude difference of 7 days (p<.0001).  OLS regression returned an estimated excess LOS of 14.3 (95%CI: 14.0 – 14.6) days.  The estimated increase in LOS from the GLM model was 7.0 (95% CI: 6.9 – 7.1) days.  When accounting for the timing of infection in a multi-state model, the attributable LOS was estimated to be 2.28 (95% CI: 2.14 – 2.41) days

    Conclusion: CDI significantly contributes to LOS in the acute care setting but the magnitude of its estimated impact is substantially smaller with methods that account for the time-varying nature of infection. Further efforts to control for time-varying confounding will refine these estimates. Estimates of CDI costs should be based on more accurate measures of attributable LOS to avoid overstating the benefit of interventions to reduce CDI

    Vanessa Stevens, PhD1,2, Karim Khader, PhD2, Richard E. Nelson, PhD2, Makoto Jones, MD, MS3, Michael Rubin, MD, PhD4 and Matthew Samore, MD5, (1)University of Utah College of Pharmacy, Salt Lake City, UT, (2)Ideas Center, VA Salt Lake City Health Care System, Salt Lake City, UT, (3)Internal Medicine, University of Utah School of Medicine Division of Epidemiology, Salt Lake City, UT, (4)Internal Medicine, University of Utah School of Medicine, Salt Lake City, UT, (5)University of Utah School of Medicine, Division of Epidemiology, Salt Lake City, UT


    V. Stevens, None

    K. Khader, None

    R. E. Nelson, Roche: Consultant and Grant Investigator, Consulting fee, Research grant and Research support

    M. Jones, None

    M. Rubin, None

    M. Samore, None

    Findings in the abstracts are embargoed until 12:01 a.m. EDT, Oct. 8th with the exception of research findings presented at the IDWeek press conferences.

    Sponsoring Societies:

    © 2014, All Rights Reserved.

    Follow IDWeek