Data-driven Tools for CLABSI Reduction

Central line associated bloodstream infections (CLABSIs) are hospital-acquired complications that contribute to increased patient morbidity and longer lengths of stay. Lucille Packard Children’s Hospital (LPCH) has previously reported 2.2 CLABSIs per 1000 line-days, which is higher than the national average of 1.5 CLABSIs per 1000 line-days. One of the leading strategies for reducing CLABSI rates is improving compliance with central line bundle elements – a standard of care for dressing changes, antiseptic washes, sterile line accesses etc.

Our high level objective was to use data to reduce CLABSIs by changing institutional practices and individual behaviors.

Specific aims:

  1. Collect, understand and process EHR central line data
  2. Build a suite of analytics and visualization tools of central line data
  3. Collaborate with IS to develop a dashboard for bundle compliance

Key conclusions:

Re-evaluation of hospital wide line-days according to NHSN guidelines reveals that the CLABSI rate per 1000 line days at LPCH has historically been overreported. We calculated a rate of 1.8 CLABSIs per 1000 line days, compared to the previously-quoted rate of 2.2. The updated rate still exceeds the national average and has been increasing steadily over the last 3 years.

Bundle compliance varies significantly between departments. In many cases, failure to do 1-2 specific elements is leading to a low overall rate. This can help target future education campaigns.

Next steps:

  1. Predictive model for line duration: Can we use patient-level data to predict how long a line should remain in, with a view to drive target-based care.
  2. Patient-level risk modeling: Can we use EHR data to risk-stratify patients in terms of future CLABSI risk, in order to identify patients at highest infection risk.

Simran Mirchandani

MS student in Management Science & Engineering

Martin Seneviratne, MD

MS student in Biomedical Informatics

Augustine Chemparathy

Undergraduate student in Bioengineering

Ron Li, MD

Clinical Informatics Fellow

Andrew Ward

PhD student in Electrical Engineering