When Colorado-based integrated health system UCHealth set out to reduce the burden their Epic EHR imposed on clinicians, it started by prioritizing improvements that affected the largest number of clinicians possible, along with distributing EHR tip sheets and holding training sessions.
None of these approaches had the kind of impact they had hoped to see. To increase their flexibility and responsiveness, UCHealth’s IT leaders decided to try a new approach. They went with the optimization “Sprint” model, a team-based intervention which focused on solving problems one clinic at a time. The IT team rolled out this approach within six of the health system’s 400 clinics.
The Sprint intervention they designed involved three components: 1) training clinicians to use existing EHR features more effectively, 2) redesigning the multidisciplinary workflow the clinic and 3) building new specialty-specific EHR tools. The underlying purpose of the Sprint process was quality improvement, so team members focused more on continuous process improvement then the data collection.
The Sprint project team included one FTE primary physician informaticist, a nurse informaticist, a project manager, 4 EHR trainers, and 4 EHR analysts.
UCHealth carried out Sprints, which relied on Agile software development methods, between January 2016 and July 2017. The Sprint leaders met with 90, 60 and 30 days before each Sprint to map out what needed to happen.
As part of the process, some Sprint team members met both individually and in small groups with nonclinician staffers to observe multidisciplinary workflow and patient flow within each clinic. Once they finished the observation process, trainers taught relevant EHR best practices to staffers, and each clinic’s manager redesigned common workflows to see that clinicians and staff did important tasks the same way.
Meanwhile, on the development side of the project, IT leaders began building new EHR tools and fixing existing tools.
First, the managers had to approve requests for new EHR tools. Once that approval had been obtained, the Sprint manager put the requests in a chart whose headings included Backlog, To Do, In Process and Parking Lot. These items were reprioritized each day as the process evolved. Project leaders, after it was updated, put the chart in a location where clinic staffers see the status of all requests, including Parking Lot items that weren’t going to be addressed during the Sprint. The EHR analysts also handled requests to fix broken items such as a smart link that did not route users to the right data and incorrect routing of patient portal messages.
Ultimately, a total of 220 clinicians participated in the Sprint process, including 65% who attended at least three training sessions and 19% who participated in more than 10 sessions.
As measured by the Net Promoter Score, satisfaction with the Sprint was +52 (with -100 the worst score and +100 the best). Clinician satisfaction with the EHR increased from -15 to +12. Meanwhile, physicians’ perceptions of their own burnout levels fell. Positive narrative feedback on Sprints was almost uniformly positive, with one clinician, for example, writing that “Your team spirit is a model for us and for any team,” and another stating that “A related outcome…has been the improvement in clinicians’ morale and attitudes.”
When analyzing the costs and benefits of this approach, the researchers determined that the Sprint team of 11 cost $1.2 million annually which, with a full schedule, can accomplish 17 two-week Sprints annually, engaging 30 clinicians per Sprint.
Meanwhile, using assumptions from the AMA STEPS forward physician burnout calendar they calculated that Sprint interventions can reduce burnout by 20% or 2.5 physicians. With the cost of replacing a physician estimated at $1 million, these initiatives could save UCHealth $2.5 million per year.