Pronovost et al 2006, NEJM

Journal Club 002

Authors: Tonya Littlejohn and Chris Nickson

Pronovost P, Needham D, Berenholtz S, Sinopoli D, Chu H, Cosgrove S, Sexton B, Hyzy R, Welsh R, Roth G, Bander J, Kepros J, Goeschel C. An intervention to decrease catheter-related bloodstream infections in the ICU. N Engl J Med. 2006 Dec 28;355(26):2725-32. Erratum in: N Engl J Med. 2007 Jun 21;356(25):2660. PubMed PMID: 17192537. [Full Text] [Supplementary Appendix]


Can a unit-based patient safety program involving the introduction of 5 evidence-based procedures, education, a central line cart and use of a checklist reduce the rates of CLABSI in the ICU setting?



  • multicenter prospective cohort study
  • 108 Michigan ICUs (5 were technically out-of-state), 103 ICUs in 67 hospitals reported data representing 85% of all ICU beds in Michigan
  • 1981 ICU-months and 375,757 catheter-days were studied


  • all patients in the participating ICUs


  • 5 evidence-based procedures recommended by the CDC:
    1. hand washing
    2. using full-barrier precautions during the insertion of central venous catheters
    3. cleaning the skin with chlorhexidine
    4. avoiding the femoral site if possible
    5. and removing unnecessary catheters
  • To implement these procedures the following were done:
    1. Clinicians were educated about practices to control infection and harm resulting from CRBSI
    2. a central-line cart with necessary supplies
    3. a checklist was used to ensure adherence to infection-control practices
    4. providers were stopped (in nonemergency situations) if these practices were not being followed
    5. the removal of catheters was discussed at daily rounds
    6. teams received feedback regarding the number and rates of CRBSI  at monthly and quarterly meetings, respectively
  • Also introduced were a unit-based safety program, use of the daily goals and an intervention to decrease VAP
  • Nursing and medical team leaders were appointed in each hospital and given instruction in the science of patient safety


  • pre-existing usual practice
    • only 13 of the 67 hospitals (19%) included chlorhexidine in the central-line kits used in the ICUs


  • Primary outcome: decrease in the CRBSI rate at 3 months
  • Secondary outcome: decrease in the CRBSI rate over 18 months
    • The median rate of infection per 1000 catheter-days decreased from 2.7 at baseline to 0 (P≤0.002) throughout all periods after implementation of the study intervention
    • incidence-rate ratios (i.e. baseline = 1.0) decreased from 0.62 (95% CI, 0.46 to 0.85) at 0 to 3 months to 0.15 (95% CI, 0.07 to 0.32) at 16 to 18 months of follow-up
  • the intervention was modestly more effective in small hospitals, with an incidence-rate ratio of 0.97 (95% CI, 0.96 to 0.99; P<0.001) for each 100-bed decrease in the size of the hospital


  • Some clarifications:
    • NNIS definition of CRBSI was used
    • the study included PICC lines
    • use of multiple lines in one patient was counted as 1 catheter-day
    • quarterly rate of infection was calculated as the number of infections per 1000 catheter-days for each 3-month period
    • total number of catheter-days changed little during the study
  • Information that was not available/ assessed:
    • no data on causative organisms (e.g. contaminant or true infection?)
    • compliance with the study intervention was not assessed
    • change in antibiotic prescribing
  • Potential biases and confounders
    • uncertain if confounding infection control practices were also implemented
    • potential measurement bias from under-reporting CRBSIs by ICUs
    • not all ICUs provided baseline data. It was calculated that data was missing for 113 of 2216 potential ICU-months (5%). However, sensitivity analysis found no difference if only those ICUs with complete/ baseline data were used
    • may have been subject to the Hawthrone effect (improvement occurred because it was being studied, rather due to any specific therapy)
  • This study design can only demonstrate an association, not causation. However, a subsequent cluster-RCT by Marsteller et al (2012) supported the findings from this study.
  • Do all components of the intervention need to be performed? Could similar gains be achieved with a less complex, less costly intervention?
  • a more recent meta-analysis by Marik suggests that femoral CVCs do not have increased rates of CLABSI


This evidence-based bundled intervention was associated with a large and sustained reduction (up to 66%) in rates of catheter-related bloodstream infection that was maintained throughout the 18-month study period.

The benefits from the intervention are likely to be real, despite this study being unable to establish causation as it is an observational study. The benefits are not simply due to the existence of a checklist, there was also extensive education, infrastructure and adoption of a patient safety culture.

One comment to “Pronovost et al 2006, NEJM”
  1. Pingback: The LITFL Review 139 - LITFL

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.