Rapid Change with a Clinical Performance Group: Define, Design, Implement
Historically it takes 17 years for research to become practice in healthcare. Banner Health, a large healthcare organization that includes 24 facilities in seven states, has developed a model to cut this time down to months rather than years. Clinical Performance Groups (CPGs) are formed to provide focused, accountable leadership for improving patient outcomes across Banner Health by insuring that best clinical practices are in use. Each CPG focuses on a specialty and includes a medical director, clinical nursing director, and a process director. The medical and clinical directors have expertise in the specialty area as well as leadership experience within the organization. The process director is an industrial engineer who may or may not have healthcare experience but who has experience with development of metrics, designing a robust process and implementation. Banner Health has several CPGs including obstetrics, critical care, and hospital medicine. This article will describe how this model has been used successfully in obstetrics to reduce Banner Health’s rate of elective deliveries at less than 39 weeks gestation, an initiative supported by most obstetrical professional organizations such as the American Congress of Obstetricians and Gynecologists (2009), as well as national quality organizations including the March of Dimes (2010) and The Joint Commission (2012).
Define the Problem
The first step was to identify which project to focus on. We convened a group of clinicians, physicians, and nurses, from all 19 hospitals within the Banner Health system that provide obstetrical care and performed an affinity exercise. This process—listing every problem that might need to be addressed—yielded 126 potential projects. We then ranked projects by rating them on the clinical impact (available evidence, improving patient outcome, reducing waste, and standardizing care), the level of effort required (time, culture, and workflow changes) and the financial impact (reduction in costs, increase revenue). Reducing elective deliveries at less than 39 weeks scored 270 out of 300 points. Research indicates neonates who are allowed to complete 39 weeks gestation have fewer neonatal intensive care admissions, fewer immediate respiratory and feeding problems, as well as reduced morbidity within the first year of life (Melamed, 2009; Clark et al., 2010).
An analysis of current practice indicated few of our facilities had policies regarding elective deliveries at less than 39 weeks and those that did were not consistently following them. It became clear to the CPG that this was an area that could be positively impacted in a short period of time. A Clinical Practice Statement was written stating that it would be an expectation at Banner Health that “Elective singleton inductions or cesarean sections will only be performed on patients who are 39 weeks gestation or greater.” It was reviewed by key stakeholders in the organization including subject matter experts. The two teams the CPG rely upon for direction and support from a clinical perspective are the Clinical Consensus Group (comprised of practicing physicians and nurses in a specialty area) and the Clinical Discipline Group (comprised of nursing leadership responsible for operations). These groups became the change champions in their facility as we moved forward with the project. The project was also reviewed and approved by the chief medical officers and chief nursing officers from every facility. The define phase took approximately 1 month.
Design the Process
Once the clinical practice was accepted by leadership, we assembled a team to design the process. Banner Health is a healthcare system with large, urban facilities and small, critical access, rural facilities. The delivery rate throughout the organization ranges from 60 babies to 7,000 babies per facility annually. This wide range of services poses its own challenges when trying to develop standardized approaches to care. Therefore, the design team was made up of a cross section of facilities and included a variety of roles such as physicians, nurses, and support staff.
The team agreed the best way to reduce elective deliveries was to focus on the time of scheduling labor inductions and cesarean sections. A process flow was proposed whereby the physician asking to schedule induction or cesarean section was asked to provide the gestational age on the day of admission, the indication for the delivery and, if less than 39 weeks, supporting documentation for the indication. The facility OB chair would review the supporting documentation before the scheduled case could be confirmed. Realizing that not all elective deliveries occur as scheduled cases, a process was developed where the nursing staff could access the OB chair for approval before admitting patients who presented to the hospital for an elective delivery at less than 39 weeks gestation.
In order to support this workflow, the design team identified changes that were needed in our electronic systems. The system used for scheduling these procedures did not have fields for documenting gestational age or indication for the procedure. The electronic medical record allowed users to select both induction and augmentation, terms with very different meanings, which were being used interchangeably and were skewing our data. Changes were made to both systems to support the agreed upon workflow.
The final work for the design team was to develop education and communication plans for implementation. The communication plan was twofold: internal and external. Internally, we needed to communicate the process change to the medical providers and their office staff as well as the hospital staff. In addition, many of the physicians reported that it was often the patient’s desire to deliver prior to 39 weeks for reasons of convenience such as availability of child care for her other children. Banner Health partnered with the March of Dimes to get the message out to the public through television news stories, newspapers, and the Internet in all of our markets. Once it appeared on the Internet, several ”mommy” blogs picked up the story, and it spread across the county. The design phase of this project took approximately 10 weeks.
Implementing change across 19 hospitals is challenging. At each facility, an implementation lead was identified who was responsible for identifying a team to implement, educate, and monitor the change. They were given an implementation toolkit that included a draft email they could send to their physicians and nursing staff, a presentation to use at staff meetings, a 3-minute change-of-shift statement, and access to our reference library. Implementation was set to start at every facility on the same day. Weekly meetings occurred with the leads to monitor their preparation and identify any barriers they may have encountered and strategies were developed to deal with these. The meetings continued post-implementation to provide ongoing system support to ensure a standard model was followed, consistent communication was occurring, and to help monitor the change. The implementation phase took 3 weeks to prepare with one day to go live across the system.
A system dashboard was developed for monitoring all perinatal projects. Data is extracted from the electronic medical record and updated nightly to allow near real-time information. The dashboard is designed to review and compare measures at the system, facility, or unit levels and allows managers with secure access to drill down to the patient and provider level. Department managers are encouraged to use this tool to help drive the process on their unit. In addition, during this project, the facility leads performed chart audits on any neonate delivered between 37 and 39 weeks to ensure compliance to the established criteria. Cases that did not meet criteria were sent to the quality department for peer review. Post-implementation, Banner Health has observed a downward trend in total deliveries at less than 39 weeks, decreasing by 22% over 6 months, which represents 2,600 more babies per year being delivered at term.
The Clinical Performance Group is a robust model for defining, designing, and implementing change rapidly in an organization. During 2011, the CPG for Obstetrics was able to complete five separate clinical changes and has begun work on two more in 2012 (see sidebar) Organized timelines and a mix of large and small projects ensure a steady stream of change for the OB departments. The CPG focuses its time on developing processes including clinical evidence, implementation planning, and good in-process monitoring and outcome measures. Utilizing a one-day, system-wide implementation strategy creates productive competition among the hospitals. This focus has allowed Banner Health to reduce neonatal intensive care days, labor time, primary cesarean section rate, and to eliminate unnecessary procedures. The CPG projects have saved more than $1.7 million in material costs alone. Facilities have begun to expect change and even bring ideas of their own for system wide projects. Overall, the use of dedicated resources to focus on define, design, and implement has been a game changer at Banner Health.
Decrease elective deliveries at less than 39 weeks gestation
Use of birthing ball to reduce labor time
Reduction in use of indwelling catheters for laboring patients with epidurals
Elimination of adhesion barriers in cesarean sections based on practice-based evidence
Standard postpartum hemorrhage/massive transfusion protocol
Develop communication page to link two separate EMRs
Establish standard for timely documentation in EMR
Barbara LaBranche is the clinical performance director for obstetrics at Banner Health and is responsible for improving outcomes and standardizing perinatal care in 19 facilities. LaBranche may be reached at
Michael Parris serves as the process director for the OB Clinical Performance Group at Banner Health, creating and implementing innovative approaches to care which provide patients with a highly coordinated, safe, and reliable experience. Parris may be reached at
LaBranche and Parris presented a session on this case study at ASQ’s Quality Institute for Healthcare (http://wcqi.asq.org/2012), May 21-23, 2012, in Anaheim, Calif.