Penn Medicine




Diabetic ketoacidosis (DKA) is a serious complication of diabetes that occurs when a patient’s body produces high levels of blood acids called ketones.  There are approximately 200-300 admissions per year for DKA at the Hospital at the University of Pennsylvania (HUP) alone.  

Although an evidence-based guideline for the management of hyperglycemic emergencies existed at the start of this project, it was not being utilized. 

With this and 133 other pathways in the pipeline at Penn Medicine, we partnered with a team led by Ilona Lorincz, MD, and Nikhil Mull, MD, to explore how to translate best practices into action at the bedside successfully. 


First, the team shadowed patients as they came into the emergency department (ED) and collected feedback from providers as they utilized multiple paper iterations of the pathway, which led to more specific inclusion criteria and user-friendly design of the pathway.

They then sought to deliver the redesigned guidelines in context to physicians by texting them action-driven protocols the moment a patient fulfilled criteria for care.

During observations and interviews, a common theme emerged.   Although the current convention was that DKA patients should be seen in the medical intensive care unit (MICU), members of the care team felt that the MICU was a high cost and over-resourced setting for what they described as "some of their least sick patients."

To test their assumption that DKA patients could be cared for effectively outside of the MICU, the team piloted a targeted intervention unit where they could pull in resources as needed.

The temporary unit was staffed with floor nurses, on-call intensive care unit (ICU) nurses, and a range of other resources.  In just the first case, the team identified unnecessary resources such as the ICU-level nurse.  

They also discovered the importance of a huddle with the floor pharmacist for a safer transfer of care and generated the evidence needed in order to accept DKA patients to the floor.


The decreased variation in care, appropriate triage based on the level of care required, and avoidance of the ICU for the mildest cases led to savings of a half a day of length of stay (LOS) for all DKA patients and cut ICU stays in half.  These outcomes enabled more patients to go directly home from the ED and freed up MICU beds for sicker patients.

The iterative, bedside approach identified changes to the pathway to make the evidence actionable to providers – and the safe testing space secured the buy-in they needed from stakeholders to move treatment for DKA patients away from the MICU.  DKA patients are now accepted by the floor as standard care.

The team's insights and approach to refining and optimizing pathways also serves as a roadmap for other clinical teams working to implement evidence-based guidelines. 


3400 Civic Center Boulevard
14th Floor, South Pavilion
Philadelphia, PA 19104 

Work with us