Mayo Clinic's move to a single, enterprise-wide EHR

By Shannon Counselman, MBA, unit manager, Mayo Clinic; Emily Hamilton, MA, principal health systems engineer, Mayo Clinic; Sonia Martindale Mathern, MHA, MBA, senior principal health systems engineer, Mayo Clinic; and Kassandra Young, MS, senior health systems engineer, Mayo Clinic, Rochester, Minn.

Mayo Clinic is a multi-campus, not-for-profit healthcare organization committed to clinical practice, education and research, with three large academic medical center campuses in Jacksonville, Fla., Scottsdale and Phoenix, Ariz. and Rochester, Minn. The Mayo Clinic Health System is comprised of more than 50 clinics and hospitals throughout Minnesota, Wisconsin, and Iowa. Due to the organization’s size and wide geographic distribution, standardization has been challenging, especially with regard to the medical record, systems used and related processes. Historically, Mayo Clinic in Arizona and Florida, and the Mayo Clinic Health System used different versions of one vended medical record, while Mayo Clinic in Rochester used a completely different vendor.  

Having three different EHRs and hundreds of ancillary clinical and revenue cycle management (RCM) systems has been a barrier to practice convergence and standardization efforts. When Mayo Clinic decided to implement a single EHR across the enterprise, leadership saw an opportunity to achieve practice convergence, making this effort the largest project in the organization’s history. The single integrated EHR and revenue cycle management system will replace 287 legacy systems. More than 1,400 future state workflows were created, and more than 51,000 end users are affected by this change. The implementation of this large-scale change required four unique implementations over the course of two years.

To ensure a user-centered implementation experience, Mayo Clinic leveraged a three-part implementation approach: user-centered workflow design, user experience review and reengineering, and change management support. This article outlines key components of the process and illustrates the process in action using a brief case study.

User-centered workflow design

Care was taken to develop workflows that account for how system and end users work together. The goal of the workflow is to identify system configuration, assist in development of test scripts, guide design of training curriculum, provide insight in potential challenges to user experience and help the change management team assess process changes for the end users.

When available, the engineers started with a nonspecific vendor-supplied workflow. If none were available, the engineers collected data to create a new workflow. Standards and templates were developed, ensuring consistency between engineers. Collaborative build sessions allowed the end users to provide input regarding how the system would work, and base workflows were redesigned to incorporate institution-specific decisions.

Next, workflows were reviewed with the build teams, and any new design or additional system configuration identified during the build was documented. To finalize the workflow, end users provided a final sign-off on the system design. During review sessions, the end users identified gaps from current state to future state.

The overall goal was to create workflows representing the enterprise, but because of structural differences, divergent workflows were occasionally needed. These differences — typically resulting from different physical layouts or personal assignments — were captured within the workflow.

Once the workflows received final sign-off from the end users, a final copy was saved to a centralized location for the project. This allowed the user experience, testing, training and change management teams easy access to the documents. Furthermore, centralized storage of the workflows facilitated a smooth transition as work assignments changed and engineers left or joined the project team.

User experience review and reengineering   

The user experience (UX) team focused on refining the system to meet a high proportion of users’ expectations rather than forcing users to change their behaviors. To achieve this goal, team members shadowed and observed multiple users in many key roles, learning what was beneficial and what was cumbersome about their current processes.  Workshops were conducted to gather users’ input on what an ideal state could look like for those highest priority ideas. This allowed Mayo to consider user experience at the onset of the project rather than after implementation.

Next, vendor representatives helped to target key roles and workflows with the highest volume and impact.  Multiple interviews were conducted with members of the system build teams, including the workflow engineers, to ask questions focused around key pieces of functionality that teams felt were exciting or worrisome. This feedback was compared to the findings from UX research to identify commonalities for additional study.

System functionality was tested by two methodologies: usability labs and in-person experiences.  Usability labs consisted of highly structured testing sessions where users would employ the new functionality while teams of approximately five people observed the user’s interactions, documented comments and tracked accuracy for each task. Software was utilized to capture screen navigations, mouse clicks and keystrokes.  In-person experiences focused on showcasing portions of functionality and gathering feedback but often focused more on informal interview techniques and qualitative data capture.

Upon completion of a study, data was synthesized and presented to the project to be incorporated via changes to the build and the workflow. Development requests were supplied to the vendor for consideration with future releases. Points for training sessions or change management communications were disseminated to the appropriate teams.

Change management support

The third facet of the user-centered implementation approach was a robust change management strategy. The change management team was charged with addressing the people side of change, seeking to increase user knowledge of the new system and the anticipated workflow changes. The team had two primary guiding principles. The first principle was to use a structured change management approach and awareness, desire, knowledge, ability and reinforcement (ADKAR) framework — commonly used for successful change implementation — to guide our efforts including communication efforts.  The second principle was to leverage existing operational structures for identifying and engaging stakeholders at all levels. Operational leaders were expected to be actively engaged and sharing information across their staff.

Monthly communication networks were designed to deliver high-level project information, change management lessons using the ADKAR model, and specific topics focused on changes applying to a broad audience. The monthly forums were set up based on an end user’s primary work role or location, with inpatient, outpatient, revenue cycle, patient access and research forums. The forum attendees were expected to take the information shared at the forums back to their colleagues to assist with spreading the messages across their departments.

Change management lessons were built based on the structured ADKAR methodology. Forum attendees learned the change management tools to use with their colleagues when discussing the transition to a new EHR/RCM. The specific topics focused on broad changes affecting various end users, such as medication reconciliation, the new scheduling/registration models and the new dictation tools.

Impact analysis sessions were designed to deliver specific changes to a specific department’s end users. The sessions highlighted any gaps between current and future state workflows, primarily from a process and role perspective. Changes were categorized by level, type and extent of impact and shared with site operational leaders, who were expected to coordinate efforts at the local level to address the gaps identified.

Case study: Patient check-in kiosks

The user-centered implementation approach can be illustrated using the example of the patient check-in kiosks. Kiosks move some tasks away from the front desk staff to the patient, freeing time for other work. Using a vendor-supplied workflow as a starting point, the workflow engineering gathered feedback from scheduling, registration, revenue cycle and practice leaders. During sessions with operational owners called collaborative build, stakeholder negotiated which tasks a patient should be able to complete on the kiosk, including:

• Taking patient photographs that would file to the medical record
• Answering the Medicare Secondary Payer Questionnaire (MSPQ)
• Reviewing demographics, allergies, and medications

The group also determined which features would not be available, such as answering lengthy questionnaires and printing receipts and appointment itineraries. The application team validated that the end user requests were technically feasible, and the stakeholder group provided final sign-off on the workflows.

After the process decisions were configured in the system, the user experience team organized sessions with volunteer patients, simulating appointment check-in with actual kiosks. The user experience team also evaluated how the kiosk fits into Mayo Clinic’s overall patient strategy, including the patient portal and the electronic check-in workflow. The team provided a comprehensive list of recommendations to the build team and operational owners. Some recommendations were immediately actionable, such as changes to on-screen instructions during the patient photograph process. Others were added to a list for future enhancements, including changes to patient navigation of the MSPQ questionnaire.

In parallel, kiosk change management activities were conducted. Through operational process change review via the Operation Impact Assessment, it was validated that kiosks and tablets were a major change for Mayo Clinic. News of the change was shared through monthly change management forums, giving staff a chance to ask questions and clarify areas of concern. Key issues were delivered to site operational leaders, allowing them to coordinate with local registration leadership to plan for additional just-in-time training and site-specific readiness activities.

Conclusion

While the magnitude of the change was greater than anything Mayo Clinic has ever undertaken, the collaboration of the workflow, user experience, and change management teams enhanced the success of the overall project. Post-implementation change management surveys focused on end-user perceptions showed positive results (See Figure 1).


The user-centered workflow design provided a foundation for other critical project activities, and allowed the user experience team to refine processes and take them to the next level. In total, the user experience team made 143 Epic development recommendations and 208 configuration recommendations. The change management team ensured that end users were fully prepared for the coming changes. Each aspect of the approach built upon the previous dimension, ensuring a quality final product while minimizing negative impact to the end users.