Spectrum Health is a not-for-profit health system based in West Michigan, offering a full continuum of care through the Spectrum Health Hospital Group, the Spectrum Health Medical Group, and Priority Health, a health plan provider. The system is comprised of twelve hospitals, 180 ambulatory and service sites, 1,300 physicians and advanced practice providers, and more than 600,000 health plan members. Spectrum Health serves the health of our community with 23,000 employees. In support of a large user base that provides direct patient care, the Information Services Service Desk processes over 26,600 incidents, 49,600 requests, and 60,100 change orders per year.
What was the situation before the launch of the service improvement initiative?
Spectrum Health had two IS request systems: Request for Service (RFS) and Secure Access Request (SAR). The RFS “system” was a Word document that was emailed to the help desk for hardware or software purchases. Help desk staff would then take the RFS document and create a ticket in the ticketing system. The SAR system was a long list (twenty-three pages, if printed) of expanding and collapsing applications that colleagues could request. There were a number of problems with these request systems:
- Colleagues didn’t know which system to use and often chose the wrong system and needed to start over with their requests.
- Colleagues couldn’t see the status of their requests, which generated calls to the help desk and resubmissions for the same items.
- Colleagues were overwhelmed by the list of applications and couldn’t find the applications they needed (and if they could, they didn’t know which access level to request).
What was the improvement strategy?
In addition to enlisting a process improvement coach to help us change the way we looked at problems, we took three critical actions:
- We observed our colleagues using the request tools where they worked. We realized that we needed to present information in a role-specific manner (i.e., a nurse has different needs than a financial analyst).
- We learned to ask better questions. Our VP was ready to shut down our service catalog project, so we asked him what he wanted. “A request system like Amazon,” he said. So, we identified the features and concepts that make Amazon easy to use, and we looked for ways to implement those features and concepts in our tool.
- We implemented Agile practices. Our VP had run out of patience with our speed of delivery. Implementing Agile enabled us to break complex problems down into manageable pieces that we could focus on in two-week sprints.
Which processes and/or tools had to be implemented, modified, or leveraged to support the improvement strategy?
We implemented several new processes to support our new request system, including:
- For automation purposes, every service must be a configuration item in the CMDB, so service owners are now required to verify the accuracy of their service configuration items so that services can be fulfilled correctly.
- Though no longer responsible for creating and routing tickets for RFS requests, the help desk adopted a new process for routing “Service Not Found” tickets to the appropriate team for fulfillment.
- Asset management populates all orderable software and hardware in the CMDB, with pictures, costs, and descriptions.
- Identity management updates and maintains user-friendly descriptions on Active Directory groups.
We also made significant changes to the way we used our CA tool:
- We modified CA Service Catalog to use the data we mined from our HR system and other key systems, like Cerner, Epic, and Lawson. We also modified the user interface to display the user’s current services and make suggestions based on what others in the same department and/or position were assigned/using.
- We implemented CA Service Desk’s CMDB to inventory and track services and equipment.
- We implemented CA Process Automation to run scripts for:
- Verifying appropriate budget and process approval
- Creating, updating, transferring, and closing tickets
- Fulfilling requests like adding/removing users in Active Directory groups and adding users to CA Client Automation groups
- Routing tickets for manual fulfillment
- Sending status updates and confirmations
What organizational changes (cultural, structural, or political) had to be implemented or modified to support the improvement strategy?
The change from the SAR and RFS was a huge cultural change for us. The SAR was supported by the Identity and Access Management (IDM) team in Information Services. So, early on, there was a transition of knowledge from IDM to the IS Service Management team. There was some trepidation around what the new catalog would mean to both teams. It was critical for the two teams to work together. This was accomplished by bringing the two teams together early in the Agile process and keeping all work very transparent.
What were some of the lessons learned?
Make sure you know which problem(s) you’re trying to solve. We had several false starts, but when we started using A3 thinking, we were able to identify the root problem we needed to solve.
Keep your stakeholders involved. Scrum reviews after each sprint kept our stakeholders involved and a part of the process along the way.
Break up the problem(s) so you can prioritize and solve them iteratively. We were much more effective using Scrum methods to help us organize the overwhelming amount of work we needed to complete. Intense focus on smaller pieces minimized context switching and helped us gain momentum.
Keep improving along the way. At the beginning of each sprint, we held a postmortem on the previous sprint. We identified things we could do better and experiment we wanted to try in the next cycle. Through this process, we were able to continuously improve throughout the project.