“Our Job Is to Make It Happen”: Marc Probst & Nick Szymanski on Migrating Legacy Data

At its core, it seems like a simple idea: ensuring that care providers are able to go into a patient’s record and access the data they need.

It is simple — in theory. In practice, however, it’s anything but. In fact, it’s an extremely daunting challenge for organizations that have data stored in legacy systems (which, let’s face it, is a sizeable percentage).

“It’s still very difficult to access that data,” said Marc Probst, who knows a few things about the trials of data accessibility, having amassed three decades of healthcare IT leadership (including 16 as CIO at Intermountain Healthcare). During a recent webinar panel discussion, Probst shared his thoughts on Application Decommissioning Challenges, along with Nick Szymanski, CIO at Richmond University Medical Center (RUMC), and Kamal Patel, CIO at Ellkay.

The timing was fortuitous, as both organizations are in the midst of a major migration. Intermountain, a 24-hospital system based in Utah, is transitioning from the homegrown systems that have been in place for 40 years, to an integrated platform.

“We’ve had to aggregate a lot of data,” Probst said, noting that the systems were built to be able to produce discrete data that could be used for analytics. “It was very forward-thinking, but we realized we weren’t going to be able to achieve Meaningful Use — at least, not without a massive investment.”

They were also created with a very unique architecture and operating system, which can make it even more difficult. “The way the database was designed is completely different than what we see today,” said Patel. And that means “having to learn a new programming language and have separate architectures around extraction.”

It’s a situation Ellkay has dealt with before, and therefore, can provide expertise in synchronizing systems so that providers can access the data they need.

And that, Patel noted, means doing some reverse engineering to extract and normalize the data. “Within each health system, the various departments have their own processes. Understanding these nuances is the most challenging part.”

Probst agreed, adding that although the ultimate goal is to have data that is “useful, usable and understood,” the skill-level needed to navigate homegrown systems is beyond what most organizations have. “Healthcare has been historically focused on vendors and partners that do a lot of the technical work for us, and so we haven’t built the same level of internal capabilities,” he said.

Szymanski found himself in a similar situation; when it came to decommissioning the ER system as part of an organization-wide initiative, he needed a partner with both technical expertise and familiarity. “Otherwise, you spend a lot of time internally trying to make sense of the data,” which isn’t where he wants to focus his time and energy. “It’s our jobs as CIOs to show the value in moving to an integrated system — what that looks like and what it brings when you’re able to do that, especially with everyone wanting to be data-driven.”

Working with Ellkay has enabled Szymanski to do just that. Of course, not every organization has the resources for outside expertise. But no matter what course is taken, there are best practices that can be applied to move toward the ultimate goal of seamless data access.

• Understand your needs. This, said Szymanski, can be done by asking what data needs to be accessed, how it is being accessed, and for how long does it need to be available? Once those questions have been addressed, leaders can decide whether to engage with a vendor.
• Choose the right partner. As with any initiative, identifying the right partner is critical. “We knew we were going to need outside help,” said Szymanski. “But beyond that, we wanted a partner who has done this type of project before and is familiar with the data.”
• Plan ahead. “Good planning makes all the difference,” said Probst, particularly with something as complex as moving data from legacy systems. “We underestimate the level of training that end users need to go through. They’re used to doing the same thing in the systems we have, and so, when we introduce a different workflow or a different piece of data or a different chart, if you don’t go through the additional training, we’re creating a disruption.”
• Communicate, communicate, communicate. In a best-of-breed model, a given department may operate at a high level when they have their tool of choice, but the organization as a whole will likely operate at a mediocre level, noted Szymanski. “But if you can communicate the benefits of moving to what the rest of the house is using and show that the organization as a whole is going to improve, that can make a big difference. If you can communicate that effectively and clearly, most of the time people will understand.”

Last but not least, don’t lose sight of the overall goal of improving usability, and consequently, improving patient care. “If there’s a way of looking at the data that’s more appropriate for the users, then my job is to make that happen,” said Probst. “They’re the ones making a difference in people’s lives; I’m the one providing the tools so they can do that.”

To view the archive of the webinar — Application Decommissioning Strategies That Save Money & Streamline Workflow (Sponsored by Ellkay) — click here.