P0140 - Optimizing and Sustaining Timely Intravenous Access for Children Treated for Suspected Septic Shock in a Pediatric Emergency Department

Background: In 2007, we launched our pediatric emergency department program for rapid recognition and escalated management of suspected septic shock. We encountered large variation in our 2-element bundle (timely antibiotic and intravenous fluid [IVF] administration) compliance. Following order set implementation in 2011, we achieved median 96% compliance to our 2-element bundle, however noted wide month-to-month variation over the ensuing 2 years.

Methodology: In 2014, we began troubleshooting the problem using standard quality improvement methodologies including Pareto analysis. We identified intravenous (IV) access delay as a significant barrier (45%) to achieving 2-element bundle goal adherence. We expanded the sepsis workgroup to include pharmacy and a trauma charge nurse (TCN) as only TCNs were trained to use a rapid infuser at that time. A brainstorming session led to a process map and fishbone diagram to identify high leverage points. We created a Key Driver Diagram focused on reducing IV access delays. We implemented a multi-part Code Sepsis PDSA in April 2015.
The primary outcome was to reduce IV access delays as a barrier to goal compliance by 50% within 6 months. Secondary outcomes were to improve adherence to the 2-element bundle. Balancing measures included over utilizing the IV team (30% estimated pre-implementation use).

Discussion: Following implementation of a multi-part Code Sepsis PDSA, we observed sustained improvements in IV access time and subsequent adherence with the 2-element bundle (see graphs). IV access delays accounted for only 9% of barriers to goal compliance. IV team responded to < 15% of all Code Sepsis activations. Incidentally, our hospital changed to Cerner in October 2017 and although the screen and order set were embedded in the electronic health record (EHR), we noted an increase in time to IV access (median line shift). Following detailed review and process mapping, it was determined that the delay was due to inaccurate date entry for IV placement time. During a code sepsis activation, the nurse can be at the bedside for an extensive time, after which estimated intervention times are entered. A brainstorming session resulted in the creation and placement of a Code Sepsis laminated poster in all rooms allowing for real-time documentation of data that, following acute bedside care, can be entered in the EHR. We observed a subsequent and sustained drop in IV access time thereafter.
An unanticipated balancing measure involved our TCNs. Three trauma activation conflicts with Code Sepsis activations resulted in scheduling change of 2 TCNs per shift and training of all nurses in rapid infuser use. No further conflicts have been identified.


Conclusions: Iterative changes including a multipart Code Sepsis PDSA resulted in sustained improvements in time to IV access. Reduction in IV access time was associated with increased adherence to the 2-element bundle.