Imaging hand-offs: Tips to help prevent medication errors
By Bonnie Darves
When U.S. Pharmacopeia (USP) earlier this year issued a report citing high rates of harm associated with medication errors in hospital radiological services, it sparked protest from the radiology community.
The USP's "MEDMARX Data Report" found that while a larger number of hospital medication errors occur in intensive care units (ICUs), harmful medication errors are proportionally higher—12% vs. 3.3%—in radiological and cardiac catheterization services than in ICUs.
Radiologists quickly pointed out that the USP report combines radiology, nuclear medicine and cardiac catheterization--and that the actual number of errors, 2,030 over the five-year study period, reflects a tiny percentage of the more than 2.5 billion imaging procedures conducted during that time. They also stressed that many of the medications involved in those errors are prescribed or administered by internists and other test-ordering physicians.
"There is nothing in this study to support claims of systemic medication handling problems inherent [in] or unique to radiology facilities," said James Borgstede, MD, chair of the American College of Radiology's (ACR) board of chancellors, in challenging the report's methodology.
No matter how the controversy pans out, experts say internists may well be part of the problem--and an even bigger part of the solution. Many of the errors were linked to patient hand-offs and communication lapses regarding patients' allergies, current medications and medical conditions, said the report's primary author John Santell, RPh, director of educational program initiatives for the USP's Center for the Advancement of Patient Safety.
'The transition of patients from their primary hospital bed to the radiology exam room opens up the risk of error.'
—John Santell, RPh
"One of the report's key findings shows that there was a general breakdown in communications across all phases of the medication-use process," Mr. Santell said. "Simply put, the transition of patients from their primary hospital bed to the radiology exam room opens up the risk of error." (See "Medication error fast facts.")
The USP report highlighted these common errors: failure to note contrast-agent allergies or to avoid drug-drug interactions; erroneous switching of infusion rates for IV medications following procedures or incorrect programming and operation of IV pumps; and improper dosing of contrast agents or sedating medications, among others.
Another major cause of errors was overlooking the growing number of drugs in certain classes, known as "offenders," that can interact with contrast agents—such as diabetes drugs.
Internist Russell Jenkins, MD, medical director for the Institute for Safe Medication Practices (ISMP), an error-prevention nonprofit in Huntingdon Valley, Pa., said ISMP has uncovered a number of such problems. Although most radiology technicians know, for example, that metformin and contrast agents shouldn't mix, some might not know all metformin-containing drugs, citing Avandamet and Metaglip as examples.
And even institutions that clearly identify potentially problematic medications often have insufficient protocols, he added. Many protocols call for withholding metformin for 48 hours after a contrast agent has been administered, but stop there. The practice standard, he said, is to draw serum creatinine at 48 hours and then decide whether it's safe to restart the drug.
The issue then becomes: Who's responsible for ordering that lab and following up? "The difficulty you run into is that radiology doesn't want to be involved in ordering the creatinine because it's not their primary responsibility," Dr. Jenkins said, pointing out that insurers usually won't cover such a test ordered by radiology. "But someone has to order it and follow it up and that's where the internist has to get involved."
Patients need to know they can't restart the metformin product until the internist calls to say it's OK—and "that's a real hole in the system where internists can have a tremendous impact on," he added. "It doesn't cost a lot of money to tell people about metformin and have policies drawn up about [it]."
Missing allergy information
Gregory A. Maynard, ACP Member, director of hospital medicine at University of California, San Diego (UCSD), knows just how problematic it is to get consistent allergy information. Even before the USP report came out, UCSD was trying to devise a system to make sure any hospital unit that dispenses contrast dye would have access to the same patient-allergy information.
"Allergy information was being entered in three different places—patient registration, the pharmacy database and the inpatient medical record—that really didn't 'talk' to each other," he said. "We tried to address that by making sure allergy info is entered in only one place."
UCSD is also implementing medication barcoding in radiology and moving toward making the medication-administration record available on the wards and, ultimately, in every other area of the hospital in which updated data would be needed.
"We've had problems with patient identification and medication reconciliation in general," Dr. Maynard said, "so these [projects] are actually a continuation of efforts being undertaken in a lot of different areas—not specifically radiology."
Hospitalists and general internists at Oregon Health & Science University in Portland, Ore., are tackling protocols from another standpoint: prophylaxis of at-risk patients who will receive contrast agents. Reviews of patient records and outcomes showed that contrast-agent prophylaxis was "all over the map," said Sima S. Desai, FACP, section chief of hospital medicine.
They're now working to establish standardized protocols in the hope of avoiding problems. "Our hospitalists are working with radiology to standardize the 'who, what and when' of prophylaxis for patients receiving contrast agents."
To help design protocols and avoid errors, experts say internists should work toward the following targets:
Providing complete information. Too often, radiologists receive scant or incomplete details about why the patient has been sent to radiology or why the ordered test is needed.
"Internists can't be expected to know all of the [test] ordering nuances, but they can indicate why the test has been ordered," said patient safety specialist Gail Nielson, clinical performance improvement education administrator for the 10-hospital Iowa Health System headquartered in Des Moines. An order will come in for "an ankle," for example, but won't say whether a bone fracture or a mass is suspected.
Internists should also provide any information radiology might need to achieve the best images or in the event of something going wrong—not just on known or possible contrast-dye allergies, but also on medications or health conditions that could affect treatment if a complication arises. "Referring physicians know the patient better than radiology ever will," she said. "They're in the best position to forward the critical information the radiologist might need."
Pushing for joint meetings. First agree on what information needs to be exchanged, said UCSD's Dr. Maynard. "I don't think most organizations have really done that."
Physicians should jointly decide on what should be communicated when patients return to the unit. Medications, including anticoagulants, may get "turned off" when a patient is in radiology, he noted, and "it's not clear whether to restart them or what triggers that." While smaller institutions might hammer out such issues in the pharmacy and therapeutics committee, Dr. Maynard suggested that larger institutions give them overall hospital committee attention, so any decision reached will have "some teeth."
Creating up-to-date medication-reconciliation records. New technology can help solve an old problem: non-matching medication lists, said Russell J. Cucina, ACP Member, a hospitalist and medical informatics specialist at University of California, San Francisco. He and his colleagues are formalizing just such a program.
"Instead of the medication list being tucked in a shadow chart in the outpatient clinic or written in the most current progress note in an inpatient's chart, have the most current med list ... as structured data in an [electronic medical record]," he said. That way, everyone—the primary care physician, hospitalist and radiologist—is working off the same list, he explained.
Establishing tight protocols for high-risk problems. These include known drug-drug interactions, high-risk medications and dealing with patients on IV medications. According to the ISMP's Dr. Jenkins, many institutions don't have policies requiring clear labeling of multiple IV tubes—an issue that surfaced in the USP report when tubes were inadvertently switched while being restarted after an imaging procedure.
He recommended that internists push to get syringes—not just containers—labeled for medications such as xylocaine and saline, used in conjunction with certain imaging and diagnostic procedures.
Identifying all patients at risk for contrast-agent allergies or reactions. That isn't always easy, but it's best to err on the side of caution, said David W. Bates, FACP, chief of general medicine at Brigham and Women's Hospital in Boston. "The key thing ... is to sort out who should get low-osmolar contrast agents if your institution doesn't use one [routinely]" because of associated expense, he said.
Appropriate risk stratification can help internists "make the case" when the more expensive agent is indicated, he added. Patients at relatively higher risk for contrast-related nephropathy include not only those with renal insufficiency but also patients with heart failure or diabetes.
Outpatient 'devil's advocate'
Although the USP report focused on inpatient errors, community internists can—and should—consider outpatients going for imaging studies or procedures to be at risk for medication errors as well, Dr. Jenkins cautioned. He supports carrying over the practice of allergy "ID-banding"—used in some hospitals for inpatient care—into areas where outpatients may show up for diagnostic testing. The patient bracelets can be used to display information and alert personnel to allergies or other issues.
"If you give patients conscious sedation before a procedure, they may not be able to tell you who they are or that they have an allergy to the medication you're about to give them," he said. "There's still a risk, and there's no reason why we can't have a banding system for outpatients, too." Recognizing that internists might object to that level of documentation if the test is basic or if contrast agents won't be used, Dr. Jenkins pointed to the importance of having key information available to anyone treating the patient.
"One can always take the devil's advocate position that something could happen—such as a fall—while the patient is in X-ray," he said. At that point, details on allergies, medications or medical history could become critical.
"Of course, 99% of the time everything is going to be just fine, even with an invasive procedure," Dr. Jenkins said. "But when it isn't, you don't want everyone suddenly scrambling for information."
Bonnie Darves is a freelance writer in Lake Oswego, Ore.
The information included herein should never be used as a substitute for clinical judgment and does not represent an official position of ACP.
Internist Archives Quick Links
Fenway Guide to Lesbian, Gay, Bisexual, and Transgender Health, 2nd Edition
This new edition reflects recent clinical and social changes and continues to present the important issues facing practitioners and their LGBT patients. Read more about the Guide. Also see ACP’s recent policy position paper on LGBT health disparities.
Join Us in Washington, DC for the Most Comprehensive Meeting in Internal Medicine
Register now and enjoy:
Discounted rates, the best national faculty, a wealth of clinical and practice management topics and hands-on sessions! Learn more about the meeting.