Before the End of the Day

Radiology Education

Credit: arztsamui from


“Before the end of the day,” a staff radiologist placed a gentle but firm hand placed on my shoulder a few months into my first year in residency, “we should talk about your report.”  I felt a dull tugging in my stomach, worried that something had gone seriously wrong – an incorrect diagnosis, a poorly phrased finding, an embarrassing lapse in voice recognition leaving out the “no” in front of “evidence of cancer.”  Maybe I was completely off-base, having seen a finding that did not exist and perhaps called it “highly suspicious.”  Maybe the ordering physician called my attending on her personal cell phone to complain.

Maybe it was the patient who called.

“I have a few stylistic recommendations for how to draft your report.”  Nobody made a telephone complaint, no technical snafu from the dictation system, and no cancer-free organs resected by an omitted negation.   And so she and I sat and reviewed the specific wordings of each find, discussing how different word choices convey different levels of concern and judgment, even when the findings themselves remain the same.

For many radiology residents, at some point during training the inevitable question will come up, “why do we spend so much time on the body of the report if it will not change the Impression section? Isn’t the diagnosis all that matters?”

Renowned Duke University behavioral economist Dan Ariely tells a story about meeting a locksmith.  When the locksmith was an apprentice and called upon to open the door for an unfortunate homeowner, he would mull over the lock for hours, trying to identify the type of lock and figure out how to open it without destroying the mechanisms.  For his hours of work and sweat, the apprentice locksmith was paid and tipped handsomely.  Decades later, the same locksmith had become so proficient he took only seconds to open the same doors without breaking a sweat.  Ironically, clients stopped giving him a tip but began to complaint about his charges, “This much money for 30 seconds of work?”  Dan Ariely then went on to describe the phenomenon experimentally, showing that the perception of value is a function of both the quality of the work and the invested work.

While this conclusion does not mean a radiologist should produce five-page reports for each chest radiograph, it does suggest that both showing the thought process (identifying the lock and sweating over the mechanisms) and actually making the diagnosis (picking the lock) are independent contributors to value – and therefore independently important – in diagnostic radiology.

For many clinicians and most patients, the radiologist has no face.  On the other hand, the interpretation report does have a face.  A typo may suggest carelessness, an uncorrected dictation error may suggest laziness, and a disorganized report may suggest incoherent clinical thought process.  Because the radiologist is not there to defend the clinician’s judgment of her report, the report must defend itself with well-organized clinical arguments that lead to the final impression.

This is why reporting style, accuracy, and structure are among the most salient teachings of radiology residency training.  Most residents began as scribes – the first skill I acquired was to be able to buffer increasingly lengthy sentences in my head while typing out every word an attending radiologist muttered during a review.  Then residents learned from the emerging theme of words that kept resurfacing: Round mass, lobulated mass, spiculated mass, necrotic mass, heterogeneous mass, homogenously enhancing mass, ill-defined mass. 

Then radiology residents learned to combine descriptors.  Heterogeneously enhancing lobulated mass.   We learned words that can be used to keep piling features onto a finding and keep postponing the inevitable period that must accompany the end of every sentence:

There is a heterogeneously enhancing mass that is lobulated and necrotic in the center, which measures 2 by 3 centimeters along the greatest cross-section, and with some evidence of invasion of nearby structures such as the posterior 4th rib…

As our skills mature we learn to tie the clinical findings together, removing extraneous words:

Heterogeneously enhancing lobulated mass with necrotic center measuring 2 by 3 centimeters in the right posterior pleura, demonstrating invasion into the adjacent posterior 4th rib…

Ultimately, each radiology trainee develops a unique style.  Some learn to separate objective findings from subjective assessments.  Some begin their reports with an overview of the underlying findings before prescribing a line-by-line analysis.

But these are independent skill sets.   Style is not so much how a radiologist decides to pack as many features of a finding in to a sentence as knowing what not to pack into a finding.  It is also about deciding where to put the diagnosis, how long to make the differential, and how much uncertainty to associate with each diagnosis. “Finding your style” requires as much clinical knowledge as cleverly applied grammatical tricks to find the best balance of the clinically important descriptors.

If clinical knowledge is the “hard skill” one acquires through radiology training, then writing style on an interpretation report is the corresponding “soft skill.”  It is the rapport-building and the bedside manner of diagnostic radiologists.  At the end of the day, those words become our firm handshake, our confident smile, and our white coat.

Howard Chen on GithubHoward Chen on LinkedinHoward Chen on Wordpress
Howard Chen
Vice Chair for Artificial Intelligence at Cleveland Clinic Diagnostics Institute
Howard is passionate about making diagnostic tests more accurate, expedient, and affordable through disciplined implementation of advanced technology. He previously served as Chief Informatics Officer for Imaging, where he led teams deploying and unifying radiology applications and AI in a multi-state, multi-hospital environment. Blog opinions are his own and in no way reflect those of the employer.

Leave a Reply