ISSUE: Catheter-related bloodstream infections (CRBSIs) affect over 200,000 patients
per year in the United States, increasing mortality risk 4%–35% and costing an estimated
$6,000–$40,000 per infection. Strict adherence to evidence-based practices has been
shown to reduce nosocomial CRBSI.
PROJECT: We studied the reduction of CRBSI using a modified Institute for Healthcare
Improvement framework and attention to campaign strategies (military, political, and
marketing) at intensive care units (ICUs) from four healthcare systems representing
tertiary referral, moderate-sized community, and small rural facilities. We implemented
use of maximal sterile barriers (sterile large drape, gown and gloves, mask, cap)
and chlorhexidine prep. CRBSI rates were calculated using Centers for Disease Control
and Prevention (CDC) definitions.
RESULTS: All sites benchmarked infection rates against National Nosocomial Infection
Surveillance (NNIS) project, which confirmed practitioners' views that their low infection
rate was “acceptable.” All sites were able to achieve the stretch goal of reducing
CRBSI by 50% (CRBSI 1.7 to 0.4/ 1000 line days, p < 0.05.) There was eradication of
all line infections in 50% of the ICUs. Initially, use of chlorhexidine prep and sterile
large drapes was initially uncommon (40% and 20%, respectively) but increased to over
90%. Forcing functions such as replacement of iodophor with chlorhexidine in the central
line kits, removal of all other available iodophor, creating of a “full barrier” accessory
pack, measurement of compliance using a checklist at line insertion, and friendly
competition among the hospitals facilitated adherence to best practices. Sharing learning
tools and group cooperation with product representatives reduced the costs of interventions.
LESSONS LEARNED: Despite local competition among the hospitals, project leaders freely
shared achievements and problem-solving strategies to overcome barriers such as data
collection, contents and organization of central line carts, and staff and physician
cooperation. Benchmarking data unexpectedly establishes a lower limit, creating the
perception that no improvement is needed. It was shown that even the “acceptable”
low infection rates at these four hospitals could be further lowered by continued
adherence to best practices. This project created a system where ICUs can create a
new community standard of patient safety and implementation of evidence-based practices
even after project completion.