6018.0: Thursday, October 25, 2001 - 8:30 AM

Abstract #29203

Understanding the relationship between patient, process and organizational factors and primary central line associated bloodstream infections (BSI): The EPIC study

Barbara I. Braun, PhD, Department of Research, Joint Commission on Accreditation of Healthcare Organizations, 1 Renaissance Blvd, Oakbrook Terrace, IL 60181, 630-792-5928, bbraun@jcaho.org, Stephen Kritchevsky, PhD, Preventive Medicine, University of Tennessee, Memphis, 66 N. Pauline Ave, Suite 633, Memphis, TN 38105, Bryan Simmons, MD, Quality Improvement, Methodist Health Systems, 188 Bellevue, Suite 408, Memphis, TX 38104, Lynn Steele, RN, CIC, MS, Special Studies, Centers for Disease Control and Prevention, 1600 Clifton Road, Atlanta, GA 30333, Steve Solomon, MD, Division of Healthcare Quality Promotion, Centers for Disease Control and Prevention, 1600 Clifton Road, Atlanta, GA 30333, Cheryl Richards, BS, RHIA, Division of Research, Joint Commission on Accreditation of Healthcare Organizations, 1 Renaissance Blvd, Oakbrook Terrace, IL 60181, and Edward Wong, MD, Infectious Disease, VAMC, 1201 Broad Rock Blvd, Richmond, VA 23249.

As hospitals strive to lower blood stream infection (BSI) rates, it is important to understand which factors affecting rates are amenable to change within the organization. The Evaluation of Processes and Indicators in Infection Control Study (EPIC) investigated the relationship between patient factors, processes of care, organizational characteristics, and patient outcome (BSI) as determined by the CDC's National Nosocomial Infection Surveillance System (NNIS) protocol. Patient and practitioner information was obtained from a random sample of approx. 60 central line insertions per hospital in 55 sites over a 13-month period including both domestic (n=41) and international (n=14) participants. The total number of BSIs across all sites was 582. Infection rates varied from 0.55 to 12.4 infections per 1000 line-days (median=3.7). Poisson regression was utilized for a hospital level analysis of overall infection rate with hospital level factors (e.g. % practitioners with less experience, percent sampled patients with known risk factors). Logistic regression was used in a patient level analysis including the 2970 sampled patients to identify factors associated with onset of infection (n=114) in this cohort. Several factors amenable to change were identified in the preliminary analyses. For example, the hospital level analysis showed that mask use and inserter experience inversely correlated with infection rates, while the patient based analysis strongly underscored the importance of timely central line removal. Overall, the findings highlight the fact that identification of underlying risk factors and care-related predictive factors may depend on the level of analysis used.

Learning Objectives: At the conclusion of the session, the participant (learner) in this session will be able to:

  • 1. List 3 factors associated with increased patient risk for developing a BSI.
  • 2. List 3 factors related to practitioner and organizational performance associated with infection rates.
  • 3. Understand how the level of analysis affects identification of factors.

    • Keywords: Infectious Diseases, Quality Improvement

    Presenting author's disclosure statement:
    Organization/institution whose products or services will be discussed: None
    I do not have any significant financial interest/arrangement or affiliation with any organization/institution whose products or services are being discussed in this session.

    The 129th Annual Meeting of APHA