July Feature: A New View of Imaging
Imaging departments were some of the first to go digital in most healthcare organizations, through the installation of picture archiving and communication systems (PACS). Changes in PACS have been slow and steady for radiologists. Until lately, that is.
As hospitals have increasingly moved to electronic health record systems, there's been growing pressure to integrate images into EHRs and other patient-centric systems so physicians can access images while using digital records.
Industry consolidation also portends change for imaging systems, as integrated delivery systems grow through acquisition and merger. As organizations seek critical mass, these new partners are likely to have different imaging systems and different ways of handling the metadata associated with those images. Building an enterprise approach to imaging is a new but growing imperative.
Some of the biggest changes are being brought about by shifts in how healthcare organizations will be paid for services. The rapid movement toward value-based reimbursement systems is creating unforeseen pressures on imaging departments at provider organizations, forcing IT executives to begin radically restructuring how those departments will operate in the future.
"This change is going to happen," says Kevin McEnery, MD, a professor in the department of diagnostic radiology at MD Anderson Cancer Center in Houston. "We're changing to a value-based system, and so the focus of what we do will change."
Shift in emphasis
Imaging departments and radiologists have brought revenue into provider organizations by doing radiological studies and interpreting them as quickly as possible-more exams and reports translated into more revenue. In addition, as departments brought in new technologies or improved capabilities of existing modalities, they increased revenues for providers.
As organizations now expect to be reimbursed for the value of the care they provide to patients, imaging operations no longer will be viewed as revenue producers, but as cost centers.
"Everything we've done involves creating more 'stuff,'" McEnery says. "Our systems have to change because the game is changing. We're going to be asked to achieve outcomes at the lowest cost. Healthcare will be patient-centered; we will be asked where we can add value to patient care. We will have to stop worrying about the volume and revenue focus."
Imaging systems will have to be able to help radiologists add value to the entire patient encounter, McEnery adds. That's a change from the past, when imaging systems added value by helping radiologists gain access to images, view them efficiently, and then create and associate reports with them, facilitated by effective workflows.
New imaging platforms will be centered on patients, use common data definitions, encompass all kinds of patient data and be accessible to all parties involved in patient care, and the resulting reports and findings will need to impact patient care, McEnery says.
As healthcare organizations increasingly use EHRs, imaging systems will have to integrate diagnostic imaging and specialist reports into organizations' patient record systems. This will present large technology challenges for providers, which currently find it difficult to include images in EHRs, or easily and quickly locate images and results in other departmental systems.
And, as healthcare organizations consolidate and merge, they're wrestling with finding the best approach to handling consolidation of imaging operations, says Steven Horii, professor of radiology and clinical director of medical informatics at the University of Pennsylvania Medical Center. Some consolidating organizations are taking a federated approach, allowing providers with existing imaging systems to continue using them, but using middleware to enable entities to communicate with each other.
Other organizations are adopting enterprise imaging systems that hold all images from a network and centralize storage. Both the federated and enterprise approaches pose problems for organizations, Horii says.
The problems don't lie in the ability to view the radiological images, which have long been based on the DICOM standard. In place for years, DICOM enables radiologists to view images from different vendors' acquisition systems on one imaging workstation.
Rather, as consolidation occurs, providers face issues integrating workflows from different organizations or using different terminology, says Eric Rice, chief technology officer of Mach7 Technologies, a Burlington, Vt.-based enterprise imaging system vendor.
Consolidating healthcare organizations plan to increase the efficiency of imaging operations and eliminate unnecessary expenses, such as doing away with duplicative examinations.
"We're seeing more traction around the idea of a universal worklist, helping to drive efficiencies and load-balance work by different radiologists across disparate sites," Rice says. In addition, enterprise systems can optimize image transmission so the images are available when radiologists are ready to view them.
That's important because radiological studies are massive, and they take time to send electronically. For instance, as imaging enables studies that provide finer resolution or more images per study, many providers are wrestling with their ability to handle large quantities of data. For example, dual-energy CT scans can produce 4,000 images, each consuming hundreds of megabytes.
Optimizing the EHR
The importance of images as part of the electronic health record has grown, as more healthcare organizations have implemented EHR systems, says Kim Garriott, principal consultant for healthcare strategies at Logicalis Healthcare Solutions, a New York-based consultancy.
"We're moving toward enterprise imaging as healthcare organizations are looking for ways to optimize their electronic health record systems," Garriott says. "Images are a large component of clinical information; clinical decisions are being made based on this information and it's often not widely available to the whole clinical team."
Additionally, the types of images available to clinicians is growing, because most end users use smartphones to take pictures or videos of patients and their conditions, Garriott says. Those images are not DICOM-standardized, and are difficult to incorporate into existing imaging-based systems, she explains.
Imaging professionals are vexed by the challenges of incorporating non-diagnostic images, such as smartphone photos of patient conditions or treatments. Not only do these images typically not meet DICOM standards, but they also are not yet of diagnostic quality useful to clinicians in recording or conveying information, says Rasu Shrestha, chief innovation officer for the University of Pittsburgh Medical Center.
It's easier to incorporate images from radiological systems into EHRs through the use of application programming interfaces (APIs), but barriers remain, Garriott says. "It's raising questions of how to create access to images at an encounter level, within the EHR, and how we make those [images] available within the context of a clinical note [outside of a radiological report]."
Data governance is growing in importance because of the variety of images that can be associated with EHRs, as well as variety in descriptions and naming conventions that can make it difficult to navigate images, she adds.
Vendor-neutral archives (VNAs) are growing in popularity because they are able to store images from a variety of vendors and modalities, and provide some ability to maintain current information. Considered to be a step beyond traditional PACS, VNAs themselves face challenges. For example, radiological images are easily copied and forwarded, and they may reside in many places within an organization. As a result, it's difficult to determine where the latest version of an image, with its associated report or annotation, may reside, says Louis Lannum, director of enterprise imaging at the Cleveland Clinic.
Some recent research suggests that complex and extensive image databases might enable expanded abilities to use image storage to facilitate research and comparative studies.
Brian Bialecki, a technology architect who conducts research for the American College of Radiology, has found that digital storage now offered by commercial vendors gives the ability to store objects and can enable users to customize it, which provides opportunities to add annotation fields. "That started our wheels turning," he says. "When you buy storage in that way, you get a value-add."
Using object storage with extensible metadata can enable an organization to associate information with images, and that means researchers can conduct quick searches of factors or conditions associated with images, Bialecki says. "We're creating so much information for doctors, they have a hard time parsing through it," he adds. "I can ask more complex questions around this approach," like querying for CT scans of 50-year-old diabetics complaining of frequent headaches. "Or if a patient has 50 head CTs, I may only want to find the ones where he complained of a headache. This storage is smarter than any other storage I've used in the past."
Still, the largest challenges to consolidating systems will likely be human factors, says Don Dennison, who heads Don K. Dennison Solutions, a Waterloo, Ontario-based consultancy. While imaging standards are widely adopted and mature, "where you run into challenges is when you have 10 hospitals, and each of them may have a different name for a procedure. It's hard to configure systems to look for 10 equivalents. We have a good handle on portability of the data, but it's definitely not plug and play."
A variety of forces will cause evolution in imaging technology, Dennison believes, and he expects that development of EHRs and VNAs will eventually take over much of the functionality of PACS.
Beyond that, radiologists' roles are changing, as healthcare systems focus on optimizing patient care vs. volume of services. "This will change the entire way people in the imaging profession think about their jobs," Dennison says. "The question will become, 'How can I make the patient healthier doing the least amount of imaging?'
"The other thing we need to avoid is the notion that all imaging should be banned," he adds. "We just want to eliminate things we don't really need to do, based on prior evidence."