Radiology discrepancies plaguing ED managers
Growing number of incidents seen in recent years
A patient came into the ED with acute onset of headache. A computed axial tomography (CAT) scan of the brain was performed, and a wet reading (an initial review of the study without definitive interpretation) in radiology reported intracranial hematoma. The patient went to the OR for emergency neurosurgery.
The next morning, the final reading was normal on a CAT scan.
Unfortunately, this is not an isolated case, laments Kevin Klauer, DO, FACEP, director of quality and clinical education at Emergency Medicine Physicians (EMP), a physicians group based in Canton, OH. "How would you like to have a craniotomy for no reason?" he poses.
Klauer says he has noticed a change in approach from his radiology colleagues in the last few years — largely in the way CAT scans and special studies are read at night. In the past, there was one reading, he adds. "Whether in the night or during the day, the results were sent to their home or they’d come in and review them, but they’d give us one reading," Klauer explains. "It’s only in the last few years that they have switched to a wet reading approach."
Some observers say that’s because today it is generally the least experienced, general radiologists who work at night, and special studies often have to be over-read by a subspecialist.
Whatever the reason, the practice is unacceptable, asserts Gregory L. Henry, MD, FACEP, clinical professor in the department of emergency medicine at the University of Michigan Medical School, risk management consultant at Emergency Physicians Medical Group, and chief executive officer at Medical Practice Risk Assessment, all in Ann Arbor. "If they are going to charge for it, they have to do it right; I don’t get a second chance," he says. The patients need the definitive reading when they are there, Henry emphasizes. "What good does it do patients to get a report the next day that they have a possible fracture?" he asks.
For whatever reason, two standards have evolved for radiological readings: one for daytime, the other for night, Klauer says.
"The difficulty is, we are making a patient management decision based on the first reading," he explains. These types of studies are of substantial consequence, Klauer says. "If you’re doing a CAT scan of the brain, you’re looking for bleeding or strokes; in the chest, you may be looking for a pulmonary embolism; in the abdomen, a possible appendicitis," he says.
Ironically, radiological misreads apparently only increase liability for the ED physician. "We’ve found that many hospitals do not hold radiologists to the same [liability] standards as they hold us," Klauer notes.
Andrew S. Kaufman, JD, a partner in the New York City law firm of Kaufman Borgeest & Ryan, says, "My experience is that it is difficult for the emergency room doctor to escape liability for his misread on the ground that he is not a trained radiologist, particularly where a delay in notifying the patient once a formal read is performed jeopardized the outcome."
Henry contends that, given the technology available today, there should never be any wet readings. "I think the days of a first reading and a second reading should be gone," he adds, noting that images can now be digitalized and sent anywhere in the world without sacrificing quality. "It makes no difference if you are down the hall or sending the image to China — there’s essentially no delay wherever I send it," Henry says.
Kaufman agrees. "In this day and age of telecommunications and digital radiology, hospitals are starting to have access to formal reads 24 hours a day by simply e-mailing the images to the radiologist’s home. This will, no doubt, become the standard of care some day — and probably sooner than you think."
Of course, we live in a far from perfect world, and digital imaging is not yet a universal reality. So what can ED managers do to help offset the challenge of radiological discrepancies?
You need an airtight radiology discrepancy process that deals on your end and on radiology’s end, Klauer suggests. "There are many hospitals that have no process improvement plan in place for radiologists, because they provide the gold standard, but what if they’re wrong?" he asks. "You have to make sure their errors are tracked as well, so they can learn from them."
At several sites, Klauer has implemented a process that requires every wet reading performed on a special study (computed tomography, nuclear studies, and so forth) to be done in writing. "It has to be part of the permanent record," he insists. "If you do a wet reading, you have to accept responsibility for both readings."
At every site where Klauer works, there also is a site quality director position. These individuals have many responsibilities, but one of their primary ones is to manage the X-ray discrepancy process. In addition, he says, any discrepancy is reported to the ED. "It’s logged; the chart is reviewed by the physician on duty or the quality director — in real time, whether or not there is a change in patient management documented," Klauer explains. "We dictate a note in the medical record about the event, and we contact the patient."
The first step toward solving the problem is awareness, he says. "If people don’t know there’s been an error, they won’t change," Klauer observes. "If a radiologist knows he’s missed appendicitis a number of times, he will get better." (Henry recommends playing hardball to solve the problem; see box, below.)
|
Fed up with discrepancies? Let’s play hardball!’ If radiological discrepancies are plaguing your ED, you might want to start getting down and dirty — in the political sense, recommends Gregory L. Henry, MD, FACEP, clinical professor in the department of emergency medicine at the University of Michigan Medical School, risk management consultant for the Emergency Physicians Medical Group, and chief executive officer at Medical Practice Risk Assessment, all in Ann Arbor. "You should realize that No. 1, you are the radiologists’ largest client," he advises. "Sit down with them, outline the problem, and say, You want to keep your largest client happy, don’t you?’" If that doesn’t work, take the problem higher up; most hospitals contract with radiologists, Henry says. "If you think there’s any sympathy on the part of neurosurgeons for radiology, for example, you’re wrong," he says. "They would back you up and demand correct service." When radiology’s contract comes up for review, you want to be part of the committee making the decision, Henry points out. "The real way to handle this is not one guy to another at 2 in the morning, but in the light of day at the executive committee meeting," he says. Put the focus on the service component that is not being met, Henry suggests. "It doesn’t matter how bright the radiologist is, or how good they are at reading films," he points out. "If they don’t come in to read film, for example, from Saturday until Monday, I’m sorry, but that is inadequate service." Pointing out such problems to a hospital administrator or chief of staff is like presenting them with found money through reduced liability, avoiding unnecessary procedures, return visits from patients whose films were originally read as normal, etc., Henry adds. You should have no hesitation about bringing such problems to their attention, he advises. |
For more on radiological discrepancies, contact:
- Gregory L. Henry, MD, FACEP, Risk Management Consultant, Emergency Physicians Medical Group, 1850 Washtenaw Ave., Ann Arbor MI 48104. Phone: (734) 995-3764. Fax: (734) 995-2913. E-mail: [email protected].
- Andrew S. Kaufman, JD, Partner, Kaufman Borgeest & Ryan, 99 Park Ave., 19th Floor, New York, NY 10016. Phone: (212) 980-9600.
- Kevin Klauer, DO, FACEP, Director of Quality and Clinical Education, Emergency Medicine Physicians, 4535 Dressler Road, Canton, OH 44718. Phone: (330) 493-4443. E-mail: [email protected].
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content