The road to PACS perfection is paved with distractions and pitfalls, Paul Chang, MD, FSIIM, says. Chang is professor of radiology, vice chair of radiology informatics, and medical director of enterprise imaging at University of Chicago Medical Center in Illinois. During the 2010 Dwyer Lecture, “The Role of Imaging Informatics in the Next Generation of EMR/EHR,” presented June 4, 2010, at the meeting of the Society for Imaging Informatics in Medicine in Minneapolis, Minnesota, Chang used the experiences of other industries to debunk common PACS myths—and to illuminate next-generation requirements for imaging.
“We need to address the real threats and challenges facing not only radiology, but health care in general. We can learn a heck of a lot more from outside this industry. We’re 10 years behind. We need to spend some time outside our own zone to see how to leverage IT better.”
—Paul Chang, MD
The first PACS myth that Chang cites is the consternation over PACS-driven versus RIS-driven workflow. “This is a meaningless concept,” Chang says. He brings up the example of Amazon, where 25 systems (ordering, inventory, billing, credit card, shipping, and so on) work together in a way that appears seamless to the end user: “It is a workflow engine optimized to address a workflow requirement,” he says. “The fact that we concentrate on PACS-driven workflow versus RIS-driven workflow is such a dangerous oversimplification of what we really need to accomplish: helping our patients.”
Instead, Chang calls for an Amazon-type workflow configured for all potential users—radiologists and others alike—that works invisibly and without interruption.
“The engine needs to be constantly reevaluated and reestablished, and will be constantly changed,” he cautions. Chang recommends that those in his position refuse to swallow lines from vendors about which radiology system drives workflow. “Next time, tell them that doesn’t make any sense,” he says. “Say, ‘My world is much more complicated than that. Yours is as well. Give me the tools that will allow me to orchestrate true integration.’”
Chang’s second PACS myth is the importance of dashboards. “Don’t get me wrong—we need business intelligence,” he says, “but the fact that we’re so proud of our dashboards? No one outside of medicine does dashboards anymore.” Instead, Chang calls for systems that not only alert users to a problem, but can actually highlight steps toward fixing it. “Dashboards are an important first step, but we need to swallow our pride and spend a day at Wal-Mart or Amazon and see how grown-ups handle business intelligence,” he says. “We can leapfrog from their experience.”
His third example of a PACS myth is the idea that one size fits all—that health-care organizations can successfully employ the same solutions in use at their peer facilities without customization. “We are the only industry that does this,” he says. “Vendors love it. It’s easy for vendors, but it’s not how the rest of the world works.”
Again, Chang uses the example of tech-savvy industries to illustrate the issue at hand. “When you go to Netflix and Amazon, their solutions are not the same,” he says. “They orchestrate and define a unique experience optimized for their needs. We don’t do that. We bend to the will of a vendor’s offerings.” Chang advocates an approach where organizations are unafraid to demand innovation from their IT vendors. “Somehow, we were tricked into believing that our problems are so hard to solve that we all need the same solution,” he says.
How can health-care IT grow out of its adolescence and into maturity? “I think there are three requirements,” Chang says. “We need much richer interoperability; we may want to tweak how we use Integrating the Healthcare Enterprise. We have to go even further with workflow and make it much more optimized. We have to be synergistic with a completely different worldview: that information will not be provider-centered. How do we deal with a world where data persist and are represented for patients? How do we synergize a personal health record with a modern-day electronic health record?”
To achieve these three goals, Chang says, health IT will have to reengineer itself as a united front, collaborating on solutions while demanding that vendors provide more innovation. “We need an informatics community, and we need our vendors to provide us with the tools we need, encouraging us to lower the barriers to becoming true value providers,” he says.