Ideas in Hospital-Based Imaging
Residents Can Do It On Call
New research reveals that well-trained residents can read studies after hours with no clinically significant negative outcome for the patient.
Each new generation of radiologists must be trained to read images, but to really develop these skills requires reading many exams. Training can—and often does—occur “on the job,” independently but with supervision, which benefits the education process. However, educators must balance those needs with the needs of the patient, who should be given the best care possible.
New research shows that well-trained residents can competently read studies after hours with no clinically significant negative outcome for the patient. Physicians and educators can therefore be confident with programs that take this approach. “We want to safely educate [residents] on the front lines and have them take care of patients without putting the patients at risk,” said Richard B. Ruchman, MD, chief of radiology and director of the radiology residency program at Monmouth Medical Center in Long Branch, NJ.
|Richard Ruchman, MD (foreground) explains radiologic findings to an attending physician.
Photo by JERRY CASCIANO
Ruchman spearheaded the study, which appeared in the American Journal of Roentgenology* in September 2007. (For details, see Ruchman RD, Jaeger J, Wiggins EF III, et al. Preliminary radiology resident interpretations versus final attending radiologist interpretations and the impact on patient care in a community hospital. (AJR Am J Roentgenol. 2007;189:523-526.)
At Monmouth, the residency program is continuously measuring quality and tracking the accuracy of residency reports given to physicians. In 2001, the team began collecting data related to on-call resident reading. “[During the on-call shift] residents would write down their findings. In the morning, the attendings would review the studies and look for discrepancies. When appropriate, reports were corrected and resubmitted to the referring physicians,” said Ruchman.
When the stats were analyzed 3 years later to examine safety, data on nearly 12,000 cases had been gathered—every study ordered after hours. “No study previously had covered the volume or breadth of imaging exams,” said Ruchman.
Discrepancies were defined as major or minor. Major discrepancies had the potential to significantly impact patient care, often requiring verbal notice to the referring physician; minor ones were judged to have no direct impact. Ruchman illustrates by comparing a missed diagnosis of appendicitis (major) against a simple renal cyst (minor). Professionals from the general medicine department determined clinical impact based on charts to help categorize discrepancies without bias.
Data analysis found a major discrepancy rate of 2.62% and a minor rate of 21.37%. Ruchman compares this to the attending misinterpretation rate of 2.1% to indicate that residents do not perform significantly worse than attendings. The most major discrepancies involved abdominal and chest examinations. The most commonly missed diagnoses were acute appendicitis followed by pulmonary embolism. Ruchman attributes this to the nature of the exams.
“These are very subtle diagnoses. In the case of appendicitis, physicians are looking for a tiny structure in the mass of bowel,” said Ruchman. The study found 81% of discrepancies related to appendicitis were false negatives and suggests that may be due to “inherent limitations based on the quality of the examination and the presence of equivocal findings.”
But these cases had few clinically significant medical consequences. A significant negative clinical effect was found in 0.3%. The researchers therefore concluded that there “is no detrimental effect on the quality of patient care from relying on preliminary interpretation made by radiology residents.”
This fact was found to hold true for properly trained first-year residents as well. Monmouth’s residents must complete a battery of tests before they are able to take call. The Accreditation Council for Graduate Medical Education (ACGME of Chicago) currently mandates that only second-year residents and older can read films. “But we found that first-year residents, who were well trained and supervised, could read properly. This supports a change in the rules,” noted Ruchman.
Whether a change in the rules is effected or not, Ruchman points out that what is important about the study is that for hospitals in which there is a radiology residency, referring physicians can have confidence in a well-trained resident.
On Top of the Teleradiology Trend
Through a recent deal made with clinical products developer Visage Imaging Inc, medical imaging service company 3DR Laboratories will offer advanced reading and postprocessing services to hospitals and other health care organizations.
Specifically, the agreement allows 3DR to integrate Visage’s Thin Client Server and PACS Server into its product portfolio, thereby enabling it to offer internal and remote advanced 2D, 3D, and 4D services to health care customers. Visage, a subsidiary of Mercury Computer Systems Inc, will receive a percentage of the fees paid by 3DR’s clients to use its technology, a cost that varies based on amount of usage and type of equipment operated.
Essentially, Visage will provide terminals and systems for 3DR on a private label basis, and 3DR will use the equipment to offer a variety of bundled services, from workflow consulting to setting up high-speed connectivity.
In doing so, the companies hope to provide a time-efficient solution for hospitals that want to ease the pressure on busy radiologic technologists and save on expensive software and hardware.
“Together, our combined technologies will free hospitals and radiology centers to outsource 3D reconstructions of CT and MRI scans, while still allowing staff the ability to further manipulate and distribute studies off of any PC platform,” said Robert L. Falk, 3DR founder and medical director. “This approach virtually eliminates the need for costly, on-site ‘fat’ workstations.”
Rather than being bound to a specific workstation, post-processing and distribution are available directly within the PACS workflow anywhere in the hospital or, via the Internet or WAN, anywhere outside the facility.
3DR senior managing director David Ferguson said his company helps relieve radiologic technologists of their burden to perform their own rendering of DICOM files. These health care workers may not be qualified to execute this task, as it requires a whole set of skills for which they are not trained, Ferguson said.
“The Visage Imaging system makes it easy for us to take those DICOM files,” Ferguson said. “Instead of putting them in an expensive workstation at the hospital, [technologists or physicians] can load them into a relatively inexpensive server that remains in their network.”
Once the customer places the scanned files into the server, 3DR technicians—”super techs,” Ferguson lightly ad-ded—can log on from the central laboratory and perform the rendering for a low variable cost without the hospital having to create or spend any major capital budget. Because of 3DR’s deal with Visage Imaging, scans remain in the Thin Client Server, which 3DR can access from its computers.
Not only does outsourcing benefit the hosptal, Ferguson said, but it also is a service to patients, who will receive the full attention they deserve from medical staff.
Ferguson added that teleradiology is gaining more and more steam in the hospital industry. 3DR has embraced this phenomenon, he said, and it has seized the opportunity to allow its health care customers to reap its benefits.
“The movement from fat, stand-alone postprocessing workstations to shared, multiuser, thin-client servers is a definite trend in the industry, and we just decided to get ahead of the curve,” Ferguson said.
Wishard Implements Veriphy Critical Test Result Management
An article in the July/August issue of the Journal of the American Medical Informatics Association examined the effectiveness of an electronic medical record (EMR) system for critical test result notification at the Veterans Affairs Medical Center in Houston. The results were disheartening: providers failed to acknowledge receipt of more than one third of transmitted alerts, and in 4% of these cases, the imaging study was completely lost to follow-up 4 weeks after the date of study.
Himanshu Shah, MD, medical director of the radiology department at Wishard Hospital, Indianapolis, understands firsthand the root of the problem. Until recently, Wishard was using a manual notification system for critical test results. “From a radiologist’s standpoint, especially in a hospital, we do everything from scratch in the reading room,” Shah said. Hearing him delineate the legacy process is almost painful: “The first step is to look at the requisition, decipher the name, page the doctor. Sometimes they’ll call back and say, ‘I don’t know about this patient.’ Wrong Dr Smith? Wrong resident? Wrong attending? What if the right person is on vacation? We can spend a lot of time just trying to find the right people and notify them of the results.”
Shah explains that at Wishard, flaws in the critical result notification process became particularly worrying after The Joint Commission released new guidelines for documenting and tracking result reporting. Steven M. Scott, vice president of facilities and ancillary services at Wishard, was also concerned. “We were having problems with the quality piece—ensuring that the radiologists did communicate results to attendings, and that I could verify a year later that this really happened.”
Scott heard about the Veriphy CTRM system from Vocada Inc, Dallas, at a conference. Thomas P. Kuster, director of imaging services at Wishard, was delighted, and not least because until that point there was no way to accurately measure compliance. “Only about 4% of the procedures are considered critical values that need to be reported,” Kuster pointed out. “So we couldn’t just do random audits, because the odds of a critical value being audited are so low. We were showing a false 100% compliant tracking methodology.”
Scott said that with Veriphy, confirmation of compliance is rapid and accurate. “That’s what makes the system so powerful for us,” he said. “It’s not some retrospective study. Should we discover that certain physicians or departments have a problem complying with our requirements, we can deal with that in real time.”
So what’s the new process, now that the Veriphy result management system is in place? It’s not hard to explain, according to Shah. “It’s a one-stop process,” he said. “Just press a button and we’re done. The system keeps track of things from there.” Kuster and team have set windows during which they expect the physician to call back; if the time limit isn’t met, Vocada staff automatically go to work tracking the physician down.
“You have a 100% likelihood of getting the right results to the right person,” Shah said. “Some days it probably lops an hour off the radiologists’ time. It’s not just allowing us to read more, but allowing us to read more efficiently. We can get the images and the reporting done in under an hour.”