Optimizing the efficiency of GI endoscopy units is a business mandate and a quality imperative for hospital-based ambulatory surgical centers.  However, many units are inadequately equipped to address this priority because of the complexity of the clinical environment and the lack of robust tools to measure and improve operational efficiency.  This is especially true for clinical labor management, which comprises the largest cost center for endoscopy practices. Systems engineering methodologies such as discrete event simulation (DES) work to harmonize complex systems using mathematical modeling and analytic techniques and represent a unique opportunity for service innovation in healthcare.  DES has been well established and validated as a business intelligence tool in the manufacturing sector and has been shown to be effective in a number of healthcare settings because it has the key advantage that it can guide quality improvement activities without disrupting the clinical care of patients.

This project has two main aims. The first is to leverage the Duke EMR to develop an efficiency and productivity dashboard for GI endoscopy to support data-driven management. The second is to use simulation modeling to guide systems redesign activities in the endoscopy unit.   These aims are further illustrated in Figure 1 below:

 

Figure 1. Specific Aims for Project

 

As described in the Institute of Medicine and National Academy of Engineering’s reporton Systems Engineering in Health Care, very little talent has been focused on improving or optimizing the productivity of the health care system.  The proposed project will address this understudied area and represents an innovative operations-research partnership to tackle complex health system challenges.  Additional, this project advances the notion of a learning health care system within Duke Medicine by applying data-driven systems engineering methods to facilitate continuous operational improvement.  This model, if successful, can be replicated in other service lines within Duke as well as with outside of Duke.
 
Lessons Learned
Through careful attention to the implementation process during the first year of this project, we learned valuable lessons for Duke Health System.  
 
Key among these were:
  • Limited bandwidth for optimization within Maestro team leads to resource challenges and capacity constraints in accessing data for process improvement
  • Inconsistent EPIC workflows between practice sites complicates data collection
  • Operational leads within DUHS are focused on day-to-day tasks with little “buffer” for innovation
  • DUHS is operationally complex with many “touches” required to make a decision.  As a result, it is difficult to be nimble and create a true learning health system
 
Successes
Challenges notwithstanding, this project met our expectations for productivity and institutional impact in the first year.  DIHI funding allowed us to explore the use of simulation modeling as aguide for systems redesign and resulted in:
  • Operational Improvement – Creation of time tracking capabilities within Maestro Care; New scheduling and patient engagement strategies; Created infrastructure for novel applications of health IT (further described below)
  • Organizational Learning – New scheduling policies and guidelines; streamlining of organizational accountability
  •  Academic Output – Abstract submission for major GI meeting; preliminary data for successful VA Career Development Award Application.
 
Next Steps
We believe that the first year of this project has created an infrastructure and organizational momentum that should be leveraged to fully realize the potential impact of this project. Extending this project for another year would yield a number of benefits
 
  • The impact of the implementation activities will not be seen for several months because of the lag in procedure scheduling.  Extending the project will allow us to quantify this impact using our endoscopy efficiency metrics.  We believe that we will be able to demonstrate significant financial returns to the health system.
  • There remain two important operational questions in the unit that could not be addressed in year 1 due to the challenges in implementing the changes discussed above.  Extending the project will allow us to analyze the ability to maximally utilize procedure rooms with a fixed physician staff number and also explore the concept of cellularization.  The latter has been successfully employed in manufacturing to break complex processes into more simple, manageable cells.  We believe this same approach may improve the operational efficiency of the 2H endoscopy unit
  • There are other endoscopy units in the Duke Health System that could be similarly analyzed with the approach we describe above, especially since the time data is now available across the Duke enterprise. Extending the project will allow us to analyze similar operational questions in the Brier Creek Ambulatory Surgery Center, the Duke Raleigh Endoscopy Unit and the Endoscopy unit at Duke Regional Hospital.  These enterprise-wide analysis will help identify best practices, standardize these practices across sites, and further improve the operational efficiency of Duke endoscopy units
  • Using the time data now available, it is possible to develop patient engagement interventions to improve patient and family satisfaction.  For example, instead of having a status board in the waiting room outside the endoscopy unit, we would like to develop a mobile application that will allow family members to track the status of patients anywhere in the hospital.  This would allow them to travel more freely and feel more control over the procedure process.  This application could be expanded to include way-finding within the hospital.  The application also has potential for dissemination to other procedural-based service lines in the health system, including the OR,Catheterization Laboratories and Interventional Radiology.