Now we come to a final component of my ruminations about the Radiology Report. In the preceding essay, I talked about inappropriate words and phrases that create ambiguity or worse. Also I went through the various components of the report including preliminary information and the main narrative. I now wish to discuss the most important parts of the report, the impression and the resulting recommendations.

The impression should be considered like a headline, encapsulating the key information to be transmitted to the referring physician. Therefore, one should not repeat verbatim what was stated in the body of the report which will only disconcert the reader by redundancy. Another point of frustration told to me by referring physicians is the inclusion of the phrase “as above” in the impression. It is silly. There is no point in having an impression if you put that phrase in. By the way, you could put the impression on top which would make proper journalistic sense akin to what is displayed in newspapers. Could you then imagine having a headline in a newspaper that would state ‘as below’ referring to the accompanying story? If that sounds ridiculous, so does ‘as above’.

How about what you put in the recommendations? I did an informal survey of about 50 of my clinical colleagues regarding the radiology report. It revealed three things that they loathed (I mean really LOATHED) about impressions rendered in radiology reports. The first thing they do not like is an impression that includes the phrase “clinical correlation requested or suggested”. For some referring physicians, this may just be standard claptrap that they ignore. But others, with slightly heightened sensitivity about propriety will resent such a statement. After all, isn’t their business clinical correlation? That is why they request imaging studies. That is why they do a history and physical. That is why they seek lab data. For you to gratuitously suggest that they do what they are supposed to do could be resented by some and appropriately so.

The second phrase that they do not like is suggestions for further non-radiology tests specified in the impression. For example, for a patient who exhibits findings of pancreatitis a “detested” recommendation by us would be to get an amylase. Referring physicians regard as anathema even more than merely the recommendation clinical correlation suggested because now you are specifically telling them what they should do when they know what they should do. This is gratuitousness “squared”.

Third, they expressed opposition to suggestions for further imaging tests specified in the impression. Well what is the problem with that? Isn’t that a manifestation of our expertise as consultants to suggest further imaging tests? If you take a narrow view of the relevance of imaging, I guess that is true. But if you care for the whole patient, which is our responsibility, it is assuredly not true because we as radiologists know how to interpret imaging tests to be sure but we generally do not know the patient very well. If there are no old studies available, for example, and there has been no previous discussion about the patient with referring physicians, which is often the case, you know very little because you have not done a history except for the one or two sentences that have been told to you in the radiology request, you have not done a physical exam, you have not examined old records, you do not know extenuating circumstances and you do not know how the patient regards his or her disease. So how could you then make a request for imaging studies which is regarded by referring physicians as tantamount to an order when you do not have that information?