The Referring Physicians Wish List: What Clinicians Need from Radiologists

Twitter icon
Facebook icon
LinkedIn icon
e-mail icon
Google icon
 - Businessmen
There are easy answers to the question "what do referring physicians want from radiologists?" A few brave clinicians offer glib replies that range from "rapid turn-around time" to "perfect certainty." But physicians readily admit that these aren't their real needs, which are far more complex and place much greater demands on radiologists. Many cite intangibles such as clinical value and relationships, while acknowledging the pressures that radiologists face.

"As we get better at interpreting our own images, radiologists have to bring more value and information to the process," says Laura J. Trombino, MD, a pediatric orthopedic surgeon with Essentia Health in Duluth, Minn.

It's challenging as each specialty and each physician brings different expectations. The common denominator on the wish list, however, is the patient. "The whole idea is to provide the best possible information to help the patient get better," says Thomas C. Peterson, MD, chair of family medicine at Fletcher Allen Health Care in Burlington, Vt.

The irony, notes David M. Panicek, MD, vice chair for clinical affairs in the department of radiology at Memorial Sloan-Kettering Cancer Center in New York City, is that imaging is integral to many clinical activities. Yet, radiologists' contact with patients is often limited to the diagnostic image.

"Radiologists often are not looking at the entire imaging exam history of a patient; more often viewing only the ordered study in isolation," observes Alec J. Megibow, MD, MPH, director of outpatient imaging services at New York University Langone Medical Center in New York City. Viewing the study as a patient with the context of patient history could satisfy referring physicians and help radiologists reframe their workflow, reports and relationships—for the better.

The imaging order: Dissected

The patient care process begins with the order, and as with every other step in the imaging chain, some approaches are superior to others. "A request for a radiology study is more than a test order. It's a consultation," says Peterson.

A simple, but occasionally overlooked, first step: Read the chief complaint. That way, the radiologist understands what the referring physician is looking for in the study. It guides the report and may inform the study and diagnosis.

"We may make a good request, but there may be better capability. If the radiologist can preview the order and say that for this question about intracranial hemorrhaging an MRI would better answer the question than a CT, the best test can be done for the patient," Peterson says.

In addition to ensuring that physicians have the optimum imaging data, this approach helps patients avoid unnecessary diagnostic imaging (and costs). Consider, for example, the pediatric patient who presents to the emergency department with a suspected shoulder fracture. It's not uncommon for the patient to be sent for generic x-rays of the arm. "There are better uses of imaging. A physician can miss a subtle growth plate fracture without the right views," says Trombino. After a suboptimal initial study, the patient often requires repeat imaging.

A study, published online Nov. 15, 2011 in Radiology, estimated that 6 percent of radiologists' recommendations for high-cost imaging exams were preceded by an inappropriate or suboptimal imaging order by the referring physician.

"These errors in imaging exam choice indicate a knowledge gap among referring physicians, which could be better addressed. One of the many roles of a radiologist is to serve as a consultant on the choice of imaging," wrote Lee et al in Radiology.

Lee and colleagues suggested that embedding decision support tools into computerized physician order entry could help reduce inappropriate requests. However, referrers may be less-than-receptive to another IT intervention. "Sometimes back-tracking in technology to the phone is what we need," says Peterson.

A thorough review of the order may eliminate unnecessary studies, offers Robert L. Wears, MD, professor of emergency medicine at University of Florida College of Medicine in Jacksonville, Fla. "The radiologist might turn up previous images, which is sometimes helpful. A patient may have had a chest CT two weeks prior and then present with the same chest pain." If the radiologist can review the prior study, and communicate the findings to the physician, it may be possible to eliminate a repeat study.

There is, however, a time element. "If the radiologist can notify me about what's been ordered or if there is a better study for the indication, it's a great help if he or she can let me know," says Alfred Sacchetti, MD, chief of emergency medicine at Our Lady of Lourdes Medical Center in Camden, N.J. "Getting feedback after the tests is not nearly as effective."

Reporting: Back to basics

Another simple truism: the imaging order should guide the report. The radiology report is the primary link between radiologists, their colleagues and the patient. Equally important, it is the core product that radiologists deliver to their customers. Providing a referring physician- centric report is a gold mine in terms of customer service; it can build relationships with clinicians and optimize decision-making and patient care. Or, the report can confuse and frustrate clinical colleagues, and perhaps send them searching for an imaging provider with clearer  reporting capabilities.

The ideal report may be best defined by negation. Poor reports come in an array of forms. One extreme is the prose-laden, jargon-heavy epic novel. "An internist has 12 minutes with a patient and can't take 20 minutes to read a radiology report," says Megibow. "Think about the text-to-verbiage ratio." Formal or informal structured reporting that includes a useful discussion of findings puts them in context and should satisfy most referring physicians.

Pixel catalogers—radiologists determined to prove that they have reviewed every corner of the image—present another challenge. As they endeavor to examine every pixel, the result may be a report overloaded with incidental findings, lesions "too small to characterize" and recommendations for follow-up imaging.

"What we want," explains Sacchetti, "is an interpretation of the image. I can look at an image and measure the aorta. I need the report to tell me what that measurement means. Is there a disease process? What are the differentials and next steps?"

Answer the clinical question, says Eric vanSonnenberg, MD, vice chair of radiology and intervention and chief academic officer at Kern Medical Center, a University of California, Los Angeles affiliate. "It sounds easy and straightforward, but it isn't always done routinely."

Key data include pertinent positive and pertinent negative findings and the extent of any findings. Radiologists' reports are most useful to oncologists when they define lesions by six degrees (anterior/posterior, left/right, superior/inferior) every time, characterize nodes using standard nomenclature and annotate the images and reports, says Charles R. Thomas, MD, chair of the department of radiation medicine at the Oregon Health and Science University Knight Cancer Institute in Portland, Ore.

Peterson offers a refresher on A-plus reporting: "Review the indications and techniques. Include comparison studies and dates. Present findings in a numeric list or summarized conclusion." Pointing out images or slice numbers that illustrate findings is helpful when the clinician reviews the report, adds Trombino.

The wise radiologist avoids putting clinicians into a corner with recommendations for follow-up imaging. "I try to keep my recommendations for how clinicians should further evaluate the patient to a minimum. I'd rather do that in a conversation where we can more fully appreciate the nuances of an individual case. Clinicians resent being told 'this needs to be done,'" says Megibow.

Suggestions for follow-up studies may be helpful when findings suggest a particular secondary study is preferable, says Thomas. For example, "Recommend MRI for evaluation of perineural involvement due to suspicious findings on CT."

Megibow advises radiologists to proceed thoughtfully before recommending follow-up imaging. "I go as far as I can with the image I have. There are many things we can do with 3D visualization to extract information from the current study rather than recommending another."

The hitch is that radiologists are pressured to read quickly, which may account for the much-disparaged phrase "clinical correlation suggested." Often, clinical context is available for the radiologist who looks for it. The situation is particularly relevant in oncology.

"The radiologist needs specialized knowledge to interpret oncologic studies. He or she can't be a film reader and circle everything that's not normal. The radiologist needs to understand the disease process and how tumors spread and respond to different types of therapies, including newer therapies like antiangiogenic therapies, cryoablation or radiofrequency ablation, where the tumor may not shrink, but other parameters like perfusion are affected," says Panicek.

While prior imaging studies offer a wealth of information, the dissemination of PACS has made it easier to overlook prior studies and interpret a study in isolation. A straight reading on a patient with previous images not only lacks context, but could be incorrect. For example, a radiologist reviewing a single image of a prior tibia fracture might detect and report a congenital lesion. Reviewing the prior studies would show the fracture, explains Trombino.

Timing is everything

Although the report is the cornerstone of radiology's service, other issues can make or break radiologists' relationships with referring physicians. In addition to providing a clear, thorough report that considers the clinical indication and prior studies, radiologists need to deliver results in a timely way, loosely defined as 24 or fewer hours.

Once again, context matters. "If the patient is in the emergency department, it helps if the physician has a preliminary report while the patient is there. Same-day imaging results for inpatients keep patient progress moving forward. And urgent findings should be communicated within hours," says Peterson, who recommends radiologists use the phone to report critical findings.

Audience matters as well. Primary care providers may wait for the radiologist's report to make a decision about patient treatment, but others clinicians base decision-making on their own review of the images. "As an orthopedic surgeon, I might not read the report for two or three days, so it's very helpful if the radiologist can call to point out a finding that was hard to visualize," says Trombino.  

Solid radiologists, those continually recognized by referring physicians as the go-to MD, understand the importance of being available for post-consult questions after hanging up the phone or signing the report. "The consultation doesn't end with the dictation. It's helpful if we can discuss the report in conjunction with the patient's pain or symptoms. Post-interpretation consultation can be very helpful," says Peterson, who estimates extended dialogues are needed in one out of every 20 cases.

Radiologists' oath  

A revision of the Hippocratic Oath for the 21st century radiologist may provide the optimum guide to meeting referring physicians' and patients' needs. Consider Megibow's mantra. "I am the patient's physician as it relates to imaging. I want to give him or her a safe exam with the least amount of radiation possible that limits overdiagnosis and be a steward of his or her wallet when he or she is here."

It means taking a comprehensive, patient-centered approach that:
  • Recognizes the reason for the imaging request;
  • Strives to determine and communicate the best imaging strategy for the clinical question asked;
  • Provides a thorough, organized report; and
  • Remains available for clinical consultation, pre- and post-imaging.