Prevention of Catheter-Related Bloodstream Infections: Looking to the Department of Veterans Affairs Health Care System for Guidance

Abstract

The recent well-documented furor over care provided to active military personnel at Walter Reed Hospital and to veterans at Department of Veterans Affairs (VA) hospitals throughout the United States provides an interesting backdrop for reading the article by Krein et al in this issue of Mayo Clinic Proceedings. Their survey demonstrated that VA hospitals were more likely than non-VA hospitals to use 2 appropriate quality measures to prevent catheterrelated bloodstream infections (CR-BSIs). The evidence-based measures surveyed include implementation of maximal barrier precautions and the use of chlorhexidine gluconate for skin antisepsis when inserting a central line. In addition, VA hospitals were significantly more likely to comply with a composite approach to CRBSIs. Regrettably, 16% of VA and 29% of non-VA hospitals did not use maximal sterile barrier precautions when inserting central lines. Furthermore, although almost all VA hospitals used chlorhexidine gluconate for line insertion antisepsis, 31% of non-VA hospitals did not use such disinfection. Routine line changes were avoided at most VA and nonVA hospitals. The VA hospitals were also more likely to have a hospital epidemiologist and a certified infection control professional as part of their infection control program. Admittedly, the research by Krein et al is based on a survey and not on direct observation; however, their data are consistent with those of recent studies documenting improvements in health care quality throughout the VA health care system. Indeed, many of the centralized and standardized processes of care and practices used for infection control by the VA health care system have real merit, including centralized medical records, a common formulary, and a systematic quality improvement program. Clearly, a standardized approach to preventing CR-BSIs offers many advantages. Pronovost et al showed remarkably low infection rates in many hospitals and significant improvement in CR-BSI rates after implementation of several routine practices in a state-wide demonstration project in Michigan. As pointed out in a recent editorial, CRBSIs can be prevented and the ability of hospitals to do so has become an important quality measure for the Joint Commission on Accreditation of Healthcare Organizations as well as for the Centers for Medicare and Medicaid Services. Moreover, CR-BSIs are beginning to be publicly reported, and comparative data will be available. In fact, recently in Boston one hospital CEO published his own hospital’s rates on his blog and encouraged others to do so as well. However, it is unclear if he reported the data using central catheter line days or patient days in the intensive care unit as the denominator, illustrating the problem with reporting information in a nonstandardized fashion. The denominator used in reporting CR-BSIs should be the number of central catheter line days. Comparative data must provide standardized definitions for the cases included in the numerator as well as the denominator and use patient severity of illness indicators to adjust for confounding factors. Now that the steps necessary to reduce CRBSIs are widely known, institutions should follow the lead of the VA health care system and adopt the recommended practices so as to reduce this preventable form of infection.