First Steps and Measures to Reduce Sepsis Mortality

University of Rochester/Strong Health
Rochester, New York, USA

Team
Michael Apostolakos, MD, FCCP Associate Professor of Medicine and Director, Adult Critical Care
Isabelle Michaud, MD, Senior Instructor of Medicine, Critical Care Attending
Mary Wicks, RN, MSN, Nurse Manager, Medical Intensive Care Unit Barry Evans, RN, MSN, Adult Critical Care Data Coordinator Timothy Kehl, RN, Nurse Leader, Medical Intensive Care Unit
Lucille Nelson, RN, Care Coordinator, Medical Intensive Care Unit

Aim
The aim of implementing the Sepsis Bundle is to improve survival of septic patients admitted to the Medical Intensive Care Unit and decrease overall mortality associated with sepsis by 25 percent.

Measures
Note: This Improvement Report from University of Rochester/Strong Health discusses the use of a previous iteration of the Sepsis Bundle concept, which differed from the present Sepsis Bundle in terms of some of the timing and measurement issues.
 
4-Hour Non-Shock
  • 2 hour diagnosis
  • Lactate
  • 1 hour antibiotic
  • All items done
  • Survivor
  • Percent mortality
  • Percent bundled
 
24-Hour Non-Shock
  • Glucose <150
  • Peak pressure <30
  • Xigris considered
  • All items done
  • Survivor
  • Percent mortality
  • Percent bundled
 
4-Hour Septic Shock
  • Fluid resuscitation
  • 1 hour antibiotic
  • CVP
  • Vasopressors
  • Inotropes
  • All Items Done
  • Survivor
  • Percent mortality
  • Percent bundled
 
24-Hour Septic Shock
  • Glucose <150
  • Peak pressure <30
  • Xigris considered
  • Cortisol stimulation test
  • All items done
  • Survivor
  • Percent mortality
  • Percent bundled
 
Changes
We began working on the Surviving Sepsis Campaign shortly after the practice guidelines were published in February 2004.  The Institute for Healthcare Improvement had formulated an initial Sepsis Bundle concept, which differed from the present Sepsis Bundle in terms of some of the timing and measurement issues, but was otherwise similar to the present formulation.  We developed three sets of measurements that constitute an assessment of our baseline functioning with regard to treating sepsis.  Alterations include the following:
 
  • Changed the data collection method from real-time collection to retrospective chart review after two months.  Review of retrospective diagnosis related group (DRG) coding summaries of discharged patients revealed missed opportunities to collect sepsis data.
  • Added survivor and mortality percentages in three bundle categories to allow for the capture of data from sepsis patients who were admitted to the intensive care unit and did not survive to the 24-hour collection time. 
  • Included a cortisol stimulation test in the 24-hour septic shock bundle to assess the appropriateness of steroid treatment for patients with severe sepsis.


Results

 

graph_percentcasescompliantwiththreeelements.jpg

 

graph_diagnosisseveresepsiswithin2hours.jpg

 

graph_serumlactateorsurrogatemeasuermentreliability.jpg

 

 

graph_reliablilityantibioticsadministered1hour.jpg Summary of Results / Lessons Learned / Next Steps
The results obtained thus far represent our baseline data that will continue to be tracked with the project improvement process.
 
Lessons Learned
  • Institute system-wide change, including the emergency department, inpatient units, and intensive care units.
  • Educate care providers because understanding is essential to success.
  • Communicate among staff, departments, and practitioners.
  • Reduce variation when possible because diagnostic tests and the treatment of sepsis varies depending on individual physician preferences.
  • Standardize the treatment of sepsis to promote intra-hospital continuity of care and improved outcomes.
 
Barriers
  • Difficulty capturing data during the initial 4-hour window (now 6 hours in the new Sepsis Bundle) because patients are frequently not admitted to the intensive care unit until 6 to 8 hours following diagnosis of sepsis.
  • Diagnosis related groups (DRG) coding excludes “urosepsis,” a frequent but inappropriately used clinical term for sepsis due to urinary tract infection.
  • Presently, not all aspects of the bundle can be completed in the emergency department and patients spend excessive time in the emergency department.
  • Lack of “champions” in the emergency department to promote the adoption of change.
 
Next Steps
  • Implement a uniform sepsis order set based on the Sepsis Bundle.
  • Expand provider education:
    • Plan Critical Care Clinical Rounds Day
      • Raise awareness of Sepsis Bundle initiative
      • Raise awareness of Early Goal Directed Therapy for treating septic shock
      • Promote and develop resident and nurse “champions” in the emergency department
  • Implement the revised Sepsis Bundle in future data collection.
  • Conspicuously post the graphs and sepsis bundle data for involved staff and physicians.



Contact Information

Barry Evans, RN, MSN, Adult Critical Care Data Coordinator
University of Rochester/Strong Health
Barry_Evans@urmc.rochester.edu

 

Average Content Rating
(0 user)
Please login to rate or comment on this content.
User Comments
Show More Comments...