September 25-27, 2013 - O'Reilly StrataRx Conference

My Folder

Stanford Hospital and Clinics’ John Shepard will be presenting at this event on the topic “Clinical Algorithms at Work: Shifting from Surveillance to Improving Clinical Outcomes” on Friday, Sept 27 at 2:20 pm (EDT).

O’Reilly’s Strata Rx Conference brings you a rare opportunity to take a fresh and deep look at improving healthcare by acquiring, analyzing, and applying big data.

Clinical Algorithms at Work: Shifting from Surveillance to Improving Clinical Outcomes

Learn how Stanford Hospitals and Clinics used a clinical effectiveness model and late-binding™ data warehouse to help reduce central line associated bloodstream infections (CLABSI) and catheter-associated urinary tract infections (CA-UTI). A multidisciplinary team of clinicians, preventionists, clinical informaticists and IT specialists created data visualization, analytic algorithms, and automated data reporting systems to increase the efficiency of CLABSI and CA-UTI surveillance. The algorithm provided inclusion and exclusion criteria, utilizing National Healthcare Safety Network (NHSN) definitions, for patients with a possible CLABSI or CA-UTI as well as an algorithm for calculating central line days.
Results

Utilizing the hospital’s electronic health record (EHR) system, the interdisciplinary team successfully developed prototype CLABSI and CA-UTI surveillance systems. For six months, the results from the surveillance system were validated by trained infection preventionists by comparing the electronic results to results gathered through complete chart review over the time period. There was an average of 90% reduction in surveillance requirements by utilizing the algorithms. The result? Clinicians can focus on interventions instead of chart abstraction. In addition, the system produced a near real time reporting dashboard which eliminated the need for the staff to produce surveillance reports along with compliance results related to the department’s interventions.

Lesson Learned

Key success factors include executive sponsorship and interdisciplinary team buy-in, late-binding™ data architecture and algorithm implementation design.