It’s fair to say that the leaders at the University of Texas MD Anderson Center, a Houston-based cancer treatment and research center, didn’t go the traditional route when it came to picking out an EMR.
As Sherry Preston, R.N., a business analyst manager at MD Anderson explains it, the cancer center deals with chronically ill patients that keep coming back and have cyclical treatment. Unfortunately, when the center began its process of looking for an EMR, it didn’t find a vendor that could not only provide it with a great combined inpatient and outpatient system, but one that could meet all of its research requirements.
“We had to try and convince a lot of vendors to do a lot of customization that they wouldn’t be able to resell on the acute care market,” Preston says.
MD Anderson trudged through three vendor implementation attempts before eventually choosing to self-develop its EMR. The idea, Preston says, began in radiology to support digital imaging informatics on the computer. What began as a radiology application kept getting add-ons that would allow the practitioner to pull up multiple components (i.e. the lab, x-rays, history). Eventually due to its convenience, it became entrenched in the workflow, where the idea was proposed to expand it even further.
“The decision was made, right or wrong, at the time, to enhance the application further into the workflow for the inpatient/outpatient and for the research,” Preston says. “Being able to go into one place, pull up one patient and have their results available, it became a valuable tool for the clinicians. So it speaks for itself to how it got absorbed into the workflow.”
The Five-Year Project
Naturally, turning this application, ClinicStation, into a usable EMR for the cancer center didn’t just happen overnight. It was approximately a five-year project until M.D. Anderson’s EMR was meaningful use certified.
Over the course of a year and a half, the team moved the application from a Visual Basic platform to a .net, for future development. It also had to develop software methodology, because as Preston says, “We essentially became a software factory. We couldn’t be a liaison between the stakeholder and the vendor, we were the vendor.”
MD Anderson recruited a number of clinical experts – nurses, pharmacy technicians, lab technicians, x-ray technicians – to be business analysts, rather than train business analysts on the clinical side. “They didn’t know what a use case was or a UI [user interface] spec. So we had to incorporate some business analyst education and tools to help them and we had to do it fairly quickly since we were launching on a fairly substantial project with some high expectations from the institution,” says Preston.
Because the cancer center obviously couldn’t shut down while the EMR was in development, one of the important tools MD Anderson leveraged to help the clinical business analysts was software that helps visualize clinical applications, created by the El Segundo, Calif.-based vendor iRise. The tool, which simulates how an application will work, was overwhelmingly approved by the stakeholder groups, who could easily discuss alterations that could then be brought to the development team.
“For me, it was like a light bulb, we knew this was the way we’d do it because they [the practitioners] can see it and they can get answers right then and there. Not after three or four meetings,” Preston says.
From a timing perspective, the simulation tool was essential because it allowed for quick changes on the fly. In fact, Preston says, this is how it got adopted in the first place, as MD Anderson needed a medication reconciliation component to the EMR for a regulatory compliance requirement in a short amount of time (four months to build it, train practitioners, test it, and deploy it). Not only did the component get built, but she says it’s still in use today.
From that launching point, the simulation eventually became a requirement for all business analysts to use as a tool when developing ClinicStation. She says using the visualization software as a tool, to act as a go-between both the development team and the clinicians, is where MD Anderson got the most value.
There were challenges that Preston and her team of business analysts encountered along the way. For one, the decision to turn clinical experts into business analysts, without training, meant there would be a lot of learning on the fly, especially when it came to the initial medication reconciliation component. She credits O'Dell Hutchison, who was the business analyst champion, for helping train the business analysts on the iRise simulation tool and bringing together the group through lunch-and-learn sessions.
“They were committed together to learn it, and say things like, ‘Oh I did it this way and it was a lot faster,’” Preston says. “They had that specific time to learn and struggle through it as a group.”
The Next Steps
The journey of MD Anderson’s EMR is not yet complete. Currently, the center is in the request for proposal stage with various vendors, looking to invest in a product that can provide a platform for its pharmacy, billing and financial, as well as various clinical documentation needs. This upcoming project will happen over three years, she says.