Striving to Save 21 Million Neurons Through Quality Improvement

Meeting: Hospital Medicine 2012, April 1-4, San Diego, Calif.

Abstract number: 97749


Approximately two million neurons are lost every minute during an ischemic stroke. University of Colorado has a median time of 71 minutes for door to tissue plasminogen activator (t–PA) administration. The American Stroke Association (ASA) sets a goal of 60 minutes or less from initial Emergency Department (ED) triage to intravenous thrombolysis for patients with ischemic stroke. That difference corresponds to the loss of approximately 20.9 million neurons.


We hypothesize that applying quality improvement principles, including data–driven process maps and plan–do–study–act (PDSA) cycles, will allow for identification of inefficiencies in the current t–PA administration process amenable to intervention.


Baseline data from March 2010 to March 2011 were analyzed. All patients presenting to the emergency department (ED) with signs and symptoms consistent with ischemic stroke were included. The primary outcome is median time to t–PA administration. Secondary outcomes include time to stroke alert activation, commuted–tomography (CT) scanning, stroke team arrival to patient bedside, and interpretation of CT images. The median time to t–PA was 71 minutes and 19% of patients received t–PA in under 60 minutes. Stroke alerts in which t–PA was not administered were slower at every measured step. For t–PA alerts, the median time to CT, CT results, and t–PA was 11 minutes, 43 minutes, and 71 minutes, respectively while non–t–PA cases had a median time to CT of 60 minutes. No patient received t–PA beyond 3 hours and 10 minutes of symptom onset. Based on this data, several targeted interventions were initiated July 15, 2011: (1) stroke alert criteria revision to only include patients who are t–PA candidates in order to increase stroke alert urgency, (2) stroke alert protocol revision to streamline efficiency in the process and allow events after non–contrast CT to occur in parallel rather than in sequence, (3) pharmacy to be present or available at all alerts to reduce time from t–PA order to administration, and (4) real time data collection to allow data–driven modifications of processes in subsequent PDSA cycles. Revised stroke alert criteria and streamlined activation process were implemented using in–service educational forums.


This project demonstrates that quality improvement strategies, specifically process mapping and PDSA cycles, can be applied to stroke alert procedures in an academic medical center in order to successfully identify inefficiencies and design interventions to address them. Ongoing data collection will be used to refine the process map, identify strategies of further improvement, and generate new interventions in subsequent PDSA cycles.

To cite this abstract:

Graves A, Tad–Y D, Levin D, Harry E, Cumbler E, Glasheen J, Sweigart J, Reichelt K, Price L, Hudak M. Striving to Save 21 Million Neurons Through Quality Improvement. Abstract published at Hospital Medicine 2012, April 1-4, San Diego, Calif. Abstract 97749. Journal of Hospital Medicine. 2012; 7 (suppl 2). Accessed May 26, 2019.

« Back to Hospital Medicine 2012, April 1-4, San Diego, Calif.