Should Radiologists Interact With Patients to Stay Relevant?

Saurabh Jha, MBBS, MRCS


August 10, 2017

Saurabh Jha, MBBS, MRCS

The emergence of value-based payment has encouraged a new movement in radiology: Patient-Centered Radiology (PCR).[1] While the term has instinctive appeal, what it precisely means is unclear. Murkier still is what radiologists should do differently to be patient-centered. Radiologists may reasonably wonder how interpreting a CT for Joe performed for abdominal pain is not Joe-centered radiology.

Advocates of PCR define it as care that is not only in the best interests of the patient but which encompasses their values and preferences. Such a definition is more managerial than prescriptive. Incorporating patients' values and preferences sounds reasonable, but what it actually means remains ambiguous. Should, for example, religiously inclined patients be offered a prayer room in the radiology waiting area?

However, PCR does unmask the limitations in radiologists being mere renderers of diagnoses. Arguably, patients are more likely to recall the ease in parking, the amiability of the receptionist taking their insurance details, the exam being completed without the encumbrances of incorrect precertification, or not having their deductible blown to smithereens by the MRI, than the receiver operating characteristics of the interpreting radiologist. This means that to achieve PCR, correct diagnosis on imaging might neither be necessary nor sufficient.

Talking to the Patient, Not the Referring Doc

Another element in a patient's care continuum is their understanding of their imaging. Traditionally, the explaining of the findings on imaging has been the dominion of the referring physicians; the primary care physician (PCP) tells Joe whether his scan is normal or abnormal. Many radiologists feel that as the experts in imaging, this responsibility is primarily theirs.[2]

Many radiologists feel that the primary consumer of the radiology report is the referring physician.

Radiologists discussing imaging results with patients is controversial, for several reasons. For one, it is unclear whether radiologists wish to talk to patients about their imaging studies. A recent survey conducted by the Radiological Society of North America (RSNA) highlighted the tension among radiologists in pursuing this endeavor.[3] The survey indicated that although a majority of surveyed radiologists (more so in academics) believe that they should meet patients more often, they are hindered by their job requirements. In other words, what stops a radiologist from meeting Peter to discuss the results of his lumbar spine MRI is timely reporting of Paul's MRI of the brain.

One reason cited by radiologists in the survey for being queasy about speaking to patients is pushback from the referring physician. Many radiologists feel that speaking to Peter about his lumbar spine MRI would violate the sacred doctors' contract that radiologists have with the referring physician. Many radiologists feel that the primary consumer of the radiology report is the referring physician.

The preferences of referring physicians, though important, are mostly of academic interest[4] if patients, the rightful heirs of their imaging, wish to speak to radiologists. Do patients wish to meet radiologists to discuss the findings on their exam? This can be a loaded question which cannot easily be parsed for non-judgmentalism, and the response depends on how the question is framed. Asking, "Do you wish to discuss the findings on your CT scan with the experts in imaging?" may fetch a different answer from "Do you wish to wait 2 hours to discuss the findings on your CT scan with the experts in imaging?"

Surveys asking whether patients wish to meet radiologists have found mixed results,[5,6] in part because what patients seem to want is to be informed about their imaging study as soon as reasonably possible; the timeliness of the message is more important than the nature of the messenger.

Many patients do not know[7] what a radiologist does—my mother-in-law still doesn't understand what I do, but that's a different matter—let alone that a radiologist was involved in their care. It was always so, but recently this began to grate radiologists who feel underappreciated in a high-pressure, hyperefficient working environment, driven by turnaround time, RVUs, and other metrics for speed—the efficiency trap. The radiologist with a flat plate on the alternator was once the center of the medical universe. With PACS (picture archiving and communication systems) and advanced imaging, the radiologist is now one of Jupiter's moons, innominate and abundant.

Reimbursement and Quality Metrics

Many radiologists wish to meet patients but also wish to maintain their salaries and not have to finish work later than is necessary. There are proposals to deal with the trade-off—ie, have our cake and eat it. It has been proposed that payers reimburse radiologists for patient consultation. This sounds promising except that payers will be hard-pressed to be convinced that they must flush even more cash to reimburse radiologists for services for which they already reimburse referring clinicians.

Another proposal is that meeting patients should be a quality metric for radiologists, but unless a large amount of reimbursement is tied to the metric, it is unlikely to change behavior. In healthcare, no good deed remains unmeasured, and once measured, incented, and given a dollar value, the good deed is no longer a good deed.

Of note, only 12% of radiologists [contacted] completed the RSNA survey. This is depressingly low and suggests that for many radiologists, meeting patients and patient-centeredness are not of paramount daily concern.

Radiology's Identity Crisis

Blogger and retired veteran surgeon Skeptical Scalpel astutely notes that the desire to meet patients reveals that radiologists are facing an identity crisis. This may not be a crisis of identity so much in the present; but in the future, three accelerating trends will surely change radiology. The first is that patients will increasingly own their own data, including any imaging they undergo. The second is that imaging will move towards greater and more complex quantification. No matter how complex measuring becomes, measuring is still measuring, and measuring does not require the higher cognitive ability typically demanded by a medical degree.

The third trend is that artificial intelligence (AI) will automate and complete many tasks of radiologists. The effect of AI on the radiology labor force is difficult to gauge. Predictions tend to be bimodal—either icy pessimism or naive optimism. But it would be a defiant Pollyanna who would insist that AI will have no effect on radiologists.

The future patient may just as likely present with symptoms as with data. They may be informed and confused in equal measures by the plethora of information at their fingertips. Whosoever manages their information—whether the PCP, radiologist, pathologist, or information specialist[8]—will need an extraordinary mastery of the sources of information. For radiologists, the most urgent task is not just to meet patients to tell them who they are, but to understand how these trends will affect radiology practice and pedagogy, and prepare for the future.

Since the advent of PACS, referring physicians rarely visit the reading room.

What of the present? There are areas where radiologists should engage with patients, much more than they do today. For example, the responsibility for shared decision-making for lung cancer screening, which, in fact, Medicare mandates, should be shared between radiologists and PCPs. There's no reason why radiologists, familiar with the harms of screening, including false-positive lung biopsies, can't counsel patients.

Breast imagers already talk to patients about the results of their mammograms. Discussing complex issues, such as the implications of dense breast on the sensitivity of mammography, should be the domain of mammographers, not PCPs. Shared decision-making for screening for breast cancer, which is long past its time, can be spearheaded by radiologists. This will allow PCPs to focus on who is ill today rather than on who might be ill in 20 years' time.

Explaining to patients the significance of incidentalomas, such as incidental adrenal or thyroid nodule, is another area ripe for radiologist-patient interaction. These findings do not need clinical correlation; auscultation for a fourth heart sound rarely changes the (in)significance of an incidental thyroid nodule, which is the gateway to overdiagnosis. Perhaps motivated radiologists can open "incidentaloma clinics"; they could be a huge benefit to society if they mitigated needless follow-up. It may take a friendly, smiling radiologist, armed with data and years of experience, to avoid a follow-up MRI for an adrenal nodule simply by saying, "It's nothing, Peter, dear chap. Please enjoy your cruise vacation in the Caribbean."

It must be emphasized that speaking to patients will not increase the radiologist's bottom line. However, holding this activity ransom to reimbursement is rather akin to refusing to leave the Titanic until another Titanic comes along. A more pressing need for radiologists is to meet their referring physicians, both before and after an imaging study. Since the advent of PACS, referring physicians rarely visit the reading room, making radiologists genuinely feel isolated from clinical care. If the mountain will not come to Muhammed, then Muhammed must go to the mountain—or at least attend the daily rounds in the intensive care unit.

In the meantime, I hope that the next TV medical drama portrays ta radiologist as an armchair-swiveling Bayesian who diagnoses cat-scratch fever merely by looking at the chest x-ray after secretly extracting the relevant history from the electronic health record. If you ask me nicely, I might consider playing that role.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: