Reduce Risks from Missed Radiological Abnormalities

Liability risk is "substantial"

The cervical spine x-rays of a motor vehicle accident victim with a chief complaint of neck pain appear normal, but fractures are later picked up by a computed tomography (CT) scan. A patient's small subarachnoid hemorrhage wasn't picked up on the initial read by either the ED attending or the radiology resident.

How do you ensure that missed abnormal results such as these receive the appropriate follow-up in your ED? If this doesn't occur, you could be held liable for the patient's bad outcome.

"This departmental interface represents a substantial liability risk to ED physicians," says John Burton, MD, residency program director for the Department of Emergency Medicine at Albany (NY) Medical Center. "The relationship is typically one where there are two separate systems—the emergency department and radiology. These must be merged to assure an efficient, clear, and safe medical systems-based practice."

Your ED also needs a consistent system for communicating with the patient's primary care or referral physician. "Effective communication systems increase the number of practitioners that review the studies and findings," says Burton. "This reduces the probability of significant misses, or changes in care due to failures or oversights in the index ED visit."

Anyone reviewing the studies should document the elements considered, the decisions made, and the individuals involved, says Burton. He notes that missed radiological findings is an area where "delay to diagnosis" is frequently alleged in case reviews or suits. "Failure to treat secondary to the missed diagnosis or delay to diagnosis is a common element of these cases in quality reviews or plaintiff suits," says Burton.

Follow System to Letter

Clinton MacKinney, MD, MS, a board-certified family physician delivering emergency medicine services in rural Minnesota, says that in his opinion, "a system that works every time, all the time is key. It seems to me that liability risk is mitigated if the system is followed to the letter."

"Community standard would suggest that ER providers occasionally misdiagnose imaging studies. We are not radiologists," adds MacKinney. "Thus, risk occurs not simply with misdiagnosis, but with missed communication of imaging findings. Risk occurs when communication between the radiologist and the ED, and between the ED and the patient, breaks down, for whatever reason."

MacKinney says that your ED's system should include:

• Documented patient advice, at the time of the imaging, that the ED provider read is preliminary only. "This should give the date and time that the image will be reviewed, and state that if there is any significant discrepancy, the patient will be notified," says MacKinney. "Also, there should be documentation in the patient's discharge instructions about the radiologist overread system."

• A tracking system, whether paper or electronic, in which the radiologist can review the preliminary read and comment about discrepancy.

• The discrepancy form forwarded to the ED provider ASAP, noting initial read, over read, and patient contact information.

• A clear chain of command stating "who does what and when" in this process.

• Documentation of patient contact, with date and time along with the follow-up plan.

• A quality assurance system. "Do a periodic review to ensure that the process is working as planned: Every patient is advised of a discrepancy and appropriate follow-up occurs," says MacKinney.

Here are "must haves" for your ED system, according to Pete Steckl, MD, FACEP, director of risk management for Emerginet, an Atlanta, GA-based emergency medicine management group:

• A mechanism for relaying ED physician x-ray interpretations to the radiologist, such that discrepancies between readings can be identified. "A fail-safe mechanism also should be present for the radiologist if no ED reading appears on the form, in which case all significant findings would be called to the ED," says Steckl.

• A radiology discrepancy log present in the ED. This should document radiologist/ED practitioner discrepancies, actions in response such as call backs to patient and/or primary care physicians, date of occurrence, date of practitioner notification by radiology, and date of contact with the patient or the primary care physician.

• Availability of the original chart to the ED practitioner, who is informed of the discrepancy for clinical context at the time of patient notification.

• A mechanism for contemporaneous documentation of patient/practitioner contact within the chart.

• Processes to assure timely contact in the case of no answer or wrong number.

• Patient contact performed by licensed personnel.

Steckl says that a "best case scenario" is a process that assesses personal responsibility to one physician for each discrepancy, as opposed to passing off responsibility from shift to shift. "The more people involved in the process, the greater the potential for falling through the cracks," he adds.

Don't Fail to Close the Loop

Burton says that two issues are of paramount importance. The first is the process for assuring that Radiology abnormal reads are available at the time of the patient encounter, and/or communicated and incorporated into the index visit for the patient.

The other important process ensures that "over-reads," studies where the initial radiology impression or "wet read" is later amended, are incorporated back to the index visit. This clarifies any change in clinical care from the decisions made based on the index visit and initial radiology interpretation.

Burton says that regarding the first process, the easiest solution is a system of "real-time" reads by radiology that are immediately communicated to the ED physician. "The shorter the time from the initiation of the radiology study to the radiologist interpretation, the more likely that the ED physician can integrate the information into the patient's index visit and clinical decision-making at the bedside," he says.

The ED physician must assure the fastest time to the radiologist interpretation as possible, along with a system whereby the ED physician has immediate access to that read when it has been completed.

Also, a system should be in place with the most efficient communication possible between the ED physician and radiologist for situations where increased delays may occur due to time of day, such as overnight coverage, or type of study, such as one that requires a specialist.

"If a situation is in place where the emergency physician is commonly making decisions based upon their own interpretations of the radiology studies,, then the radiologist should have access to the emergency physician's impression of the study to assure concurrence," says Burton.

Audie Liametz, MD, JD, assistant medical director of the ED and chairman of the ED Quality Improvement Committee at Mineola, NY-based Winthrop University Hospital, says he reviews as many x-rays and CT scans himself whenever possible. " Even though ED physicians are not radiologists, the more you do, the better you get. It is not an official reading, but I know that I've picked up x-ray findings when a radiology resident didn't see anything there, and I say, 'I want to wait for the official read,' and not infrequently I've been right."

Steckl says that he believes errors mainly surround failure to "close the loop" after notification comes of an x-ray discrepancy. Lack of repeat follow-up calls made when contact is not achieved on the initial call can also be a problem as the ED physician goes off shift.

"Another error I see is failure to take responsibility for completing that loop, once 'tagged' by the radiologist," he says. "I commonly see in our institutions a tendency to assume another party will take care of this rather mundane duty of discrepancy follow-up-'Oh, the midlevel usually takes care of this.''

Steckl says that once "tagged" by the radiologist as the practitioner notified, the ED physician can now be held legally responsible for failure to carry the process through to completion.

"This is in part a systems problem where policies are not in place or routinely followed. Responsibility remains diffuse and in the end, no one takes responsibility," says Steckl.

Another problematic issue involves discrepancies between what is relayed to an ED physician verbally over the phone by a reading radiologist, and what is dictated in the final radiologist report as having been relayed to the ED physician.

"This can be critical in cases of unanticipated bad outcomes. It leads to finger pointing between the ED physician and the radiologist, as to what information was and was not discussed. This is a plaintiff attorney's dream," says Steckl.

Steckl says that in his own clinical experience, he has been surprised to find a number of important clinical findings were included in a much more detailed dictated report that followed his own brief conversation with the radiologist. "This has led to insistence at several of our institutions on faxed 'wet reads' from the radiologist at the time of their interpretation," he says.


For more information, contact:

• John Burton, MD, Residency Program Director, Department of Emergency Medicine, Albany Medical Center, Albany, NY. Phone: (518) 262-4050. Fax: (518) 262-3236. E-mail:

• A. Clinton MacKinney, MD, MS, Senior Consultant, Stroudwater Associates, St. Joseph, MN. Phone: (800) 947-5712. E-mail:

• Pete Steckl, MD, FACEP, Director of Risk Management, Emerginet, Atlanta, GA. Phone: (770) 994-9326. E-mail:

• Audie Liametz, MD, JD, Assistant Medical Director, Emergency Department, Winthrop University Hospital, Mineola, NY. E-mail: