It’s 1:00am on a Saturday. A patient comes into the emergency room complaining of cough and severe chest pain—hallmark symptoms of pneumonia. The attending physician calls for a chest X-ray. After making a preliminary read of the film, and seeing airspace opacity, the physician says that pneumonia is likely despite the lack of fever. But no radiologist will be available to review the X-ray and confirm the diagnosis until business hours the following day. The patient is sent home with oral antibiotics and instructions to drink plenty of fluids and rest. It’s a common enough story. One that can happen in any emergency department, any night of the week.
But what if something more insidious is lurking in that patient’s lungs? A small mass or nodule that might be overlooked during a preliminary read—but would definitely be caught by a more exacting radiology review? If such a nodule was discovered, how easy would it be for the radiology department to get back in touch with the emergency department (ED), as well as the patient in question, to make sure that the right diagnosis—and, consequently, the right care—will be provided?
That’s exactly the issue that Roper St. Francis, a growing health system in the Carolina Lowcountry with three hospitals and more than 90 facilities and physician’s offices, was facing on a routine basis. As films made their way from the ED to radiology and back again, clinicians were faced with a manual and labor-intensive workflow to communicate reads of plain films—a vital piece of clinical information required for accurate diagnosis and treatment.
“If we found something on an ED plain film that we thought was not seen, we had no direct way of finding out what the ED preliminary report was,” says William Crymes, M.D., a radiologist at Roper St. Francis and a Physician Champion for Radiology Solutions. “Typically, what we’d have to do is call one of the four emergency departments, ask to speak to the charge nurse, get the charge nurse to look up the patient’s chart and then see what the emergency physician did and what he might have noted about the plain film in the chart. It took us a very long way to get what was really a simple amount of information.”
A long way to get a simple amount of information. As important clinical information travels from department to department, whether it’s a chest X-ray, a mammogram, or other type of image, many clinicians have to deal with disparate healthcare information technology (IT) systems. Too often, these systems create awkward and time-consuming workarounds to get clinicians the patient data they need, when they need it. And that’s before a potential diagnostic discrepancy comes into play. Successful health systems need ways to circumnavigate these clumsy workarounds—new, automated clinical workflows that allow them to overcome frustrating communication gaps, improving patient care and lowering healthcare costs as they go.
Using Workflow Intelligence to Bridge the Gaps
Qualitative Intelligence Communication System (QICS)** is a flexible, customizable workflow engine that ties various IT systems together to help provide an optimized workflow even in the face of the kind of communication gaps Roper St. Francis was seeing between its emergency and radiology departments.
“We needed to improve the very manual and labor-intensive form of communication between the ER and our radiology team,” says Amy Alexander, Director of Imaging Informatics at Roper St. Francis. “The challenge was to provide a meaningful tool that would integrate across our electronic health record (EHR) and picture archiving and communication systems (PACS).”
Joan Wherley, the Service Line Director for Radiology at Roper St. Francis says they were looking for a tool that would allow them to integrate across the existing systems in ED and radiology—reducing any duplication of documentation and truly enhancing the workflow for clinicians. The idea was to improve upon existing procedures, not add more complicated steps to an already clunky process.
“We needed to find a vendor that could provide a full closed-loop system. So if a discrepancy was identified, it would be reported back to ED and then ED could then say, ‘Yes, I see the discrepancy and I’ve acted on it,’ without requiring any duplication of documentation,” she says. “And we found that workflow solution allowed clinicians the ability to pass documentation between the systems in the ED, notify the ED when a discrepancy was found, and then give the ED options to acknowledge that discrepancy and notate how they followed up to close that result in a way that would let radiology know but also populate their ED record as well.”
With the implementation of the workflow engine, that patient with the suspected pneumonia and confirmed lung nodule wouldn’t get lost in the communication shuffle. When the radiologist does his over-read, a menu pops up on the PACS monitor, where he or she can agree with the preliminary read—but then add information in cases where the patient requires follow-up.
“This allows us to deal with both critical and incidental findings,” says Dr. Crymes. “We have the preliminary report right there. We can find out what the emergency physician said. It’s very simple. The ED doctors don’t have to go into more complicated reads. That’s our job. And if we think there is a discrepancy, we can add the discrepancy to our report with just a click. That then notifies the emergency department that they need to follow-up—and then lets us know when they’ve actually done it.”
Clinician Input, Clinician Control
The key to bridging any communication gap is to receive input directly from clinicians and other key users on how to make the system both useful and usable. At Roper St. Francis, the information technology team looked to both the emergency department clinicians and radiologists to ensure they were creating a help rather than a hindrance. Not the easiest task considering ED and radiology were working with two different systems—and any solution would have to seamlessly interact with both those spaces.
“Our clinicians provided a tremendous amount of input—which was important because the workflow engine is so customizable. They worked with us to create defined categories of how quickly a notification had to take place, what constitutes a clinically significant discrepancy and so on,” says Wherley. “These different physician groups came together to agree on those definitions and worked through the process so that they could make sure to minimize the documentation required, while understanding how the findings would be reported. It took a lot of agreement and understanding between those groups to pull that off in a way that we could meet all those needs.” Wherley states that having a physician champion, from both the ED and radiology sides of the fence, is crucial to reaching that understanding. Alexander adds that having a documented sign-off on the system’s requirements was also extremely helpful.
“At first, it’s hard because the ED doesn’t know why it’s being asked to work on a radiology system and radiology isn’t sure why they are working on something that isn’t necessarily radiology. But once both groups realized that the system wasn’t one or the other, but rather ‘our’ system, that helped,” says Alexander. “But having documentation that references materials that everyone agreed on, to go back to is incredibly valuable. People can see those panels, those menus, and those workflows and it helps them stay focused on why everyone decided to do it that way originally. It was an indispensable resource as we continued to roll the workflow solution out—and helped get buy-in from the clinical staff.”
Reaping the Benefits
Now that the workflow engine is up and running, Roper St. Francis is able to carefully follow ED/radiology discrepancies. Physicians both in the ED and radiology are seeing great value in being able to manage the discrepancy workflow in a more optimized manner. In fact, one of the ED physician groups went from an 11% report completion rate to 85% in only one month’s time. But Wherley also believes that the workflow solution’s implementation will also provide instrumental data for the health system’s overall quality assurance management.
“Prior to the implementation of the workflow engine, we had no way to quantify our discrepancy rates. As it turns out, there aren’t a lot of national benchmarks to compare ourselves to. There’s not that much information or peer-reviewed studies to create a national benchmark,” says Wherley. “But now we at least know how we are doing internally. We’ve built into our quality performance metrics, for both ED and radiology, performance goals based on their discrepancy information. Now we can actually have a quality program related to these discrepancies and track how we’re doing.”
Alexander says that as the technology matures, Roper St. Francis will be able to look closely at discrepancies and better understand where they come from—and, reduce potentially costly mistakes in the future.
“We may see that there are a lot of discrepancies, for example, coming from a specific body part like skull films or spine films. And, if that’s the case, we can focus targeted education back to the EDs, so they can understand what types of discrepancies occur and where they are occurring. We are going to learn a lot,” says Alexander. “And it’s going to be very interesting to see how we can decrease the rate of discrepancy—and how that will ultimately impact the quality of patient care.”
**QICS is now Conserus Workflow Intelligence™
