Christopher Roth, M.D., associate professor of radiology, vice chair, information technology and clinical informatics, and director of imaging informatics strategy at Duke University Medical Center, and Don Dennison, healthcare IT consultant, discussed how to build the right team to ensure enterprise imaging success
As healthcare enterprise consolidation continues to expand, bringing closer integration between the clinical and operational sides of the medical industry to streamline patient care, radiology plays an increasingly important central role. Medical imaging and data exchange are being performed in departments and settings outside of the traditional radiology suite, necessitating an enterprise imaging approach to data management. While the right technology is essential to collect and exchange data, just as important is having the right personnel to manage the data and keep the lines of communication open. At the 2019 Society for Imaging Informatics in Medicine (SIIM) annual meeting, Christopher Roth, M.D., associate professor of radiology, vice chair, information technology and clinical informatics, and director of imaging informatics strategy at Duke University Medical Center, and Don Dennison, healthcare IT consultant, discussed how to build the right team to ensure enterprise imaging success. The following is part of an interview with ITN Associate Editor Jeff Zagoudis.
Jeff Zagoudis: Why does the team structure need to change for consolidated enterprises, and what are some of the things that make
Don Dennison: Historically within a single hospital there have been some classic roles that took care of not only the system but the at-the-elbow end user training and data quality tasks. As these enterprises get consolidated through merger and acquisition (M&A) and affiliation, the systems get replaced by a large shared system. The tasks involved in that are different, and at certain scale there are new things to be considered — the overall architecture of all of the imaging systems integrated with the information systems, like the electronic health record (EHR) and other things. You need a different type of skillset and you need to build a team to handle the day-to-day triaging of basic password resets all the way up to significant new technology changes.
The human resources that are involved are as critically important as the information and the information technology (IT) systems. We have to create career paths for these professionals and professional development opportunities like coming to SIIM and learning these skills because technology is evolving, as well as what’s expected out of the enterprise and what IT is supposed to deliver. The expectations are rising and costs can get out of control if you don’t manage that very important human resource.
Christopher Roth: Technology has gotten so complicated, and workflows have gotten so complicated in the EHR world that what it’s sometimes missing is a degree of coordination between all of the people who need to know — on both sides of an enterprise consolidation or an M&A — what is expected.
One of the things we’ve done [at Duke] that has been really successful is to create a coordinator position, who was the team lead for our EHR imaging platform. She provides the holistic perspective to others around the enterprise, both in operations on the clinical side and also on the technical side. She was a technologist years ago, transitioned into IT and now she’s the coordinator of all these different pieces of imaging that are happening in ICUs, surgical suites and in radiology. She functions to a degree like a physician’s assistant. This position is an important one because you can interface between imaging on the IT side and various hospital services. Plus, if there’s a new consolidation coming in she’s going to be one of the first people that group faces when they come and see how we do IT at Duke. It requires a different set of skills — a degree of soft skills, marketing, customer service, but also a terrific technical background to be able to share, “This is how we do it, this is why we do it.”
This is the model we’ve been deploying not just in imaging but also in other areas at Duke — in labs, cardiology, research and other places — so that there’s the same degree of coordination that we’ve built in imaging across the other areas of our institution.
JZ: Are there prescribed roles you need to fill when you’re building this type of team or does it depend on the institution?
DD: It really depends on the size of the institution. For example, at a small scale you may have one person who’s responsible for the system management, vendor management and data quality control on a daily basis, and managing things like storage. As you scale up you start to see very specialized roles — somebody’s looking at the network, someone’s looking at storage, someone’s looking at the interface engine for integrations, the EHR team. That’s where the role that Chris describes is incredibly important because I think if we all just kind of check the boxes of those deep technical skills but no one’s creating that human interface, or is sitting above all of that and seeing where the gaps are. In other words, everyone’s doing their job perfectly but the end product isn’t solving the problem we need it to solve and delivering the mission, and end users are dissatisfied.
CR: That’s why we very deliberately chose an EHR function person to understand the big picture of our enterprise. I think you can have operations be dissatisfied when different requests are coming in and you’re providing reactionary responses back to solve that little problem. But if you’re really going to address some of the governance issues I suspect many groups around the country have, you need to have somebody who has a really close relationship with those who are designing the strategy and running the governances who can push back on some of those requests and say, “Yeah, that’s not a bad idea, but we do it this way and here’s three reasons why.”
It becomes easier then for them to not hear “no,” but instead, “Here’s another way you might consider working.” It’s less threatening to those in operations, and it actually becomes helpful for the technology side too, because [they are not] then building System A, B, C and D that all do similar things. You’ve got somebody who is aware of all the other systems that are in place and they can say, “You know, we already have a system to do that and a workflow to do that that’s already being used over here. You might go talk to those docs.”