Garbage In, Garbage Out in Radiology

There’s a commonly-used phrase that got its start in the world of computer programming: Garbage In, Garbage Out. It refers to the phenomenon of getting less-than-useful results after you supplied less-than-helpful input. For instance, giving the Congressional Budget Office incomplete data about a new national healthcare system, and then getting falsely-reassuring analysis that the system will reduce costs, improve access, slice bread better, etc.

Parents make pointed use of GIGO to shape their kids’ behavior. For instance, “Ma, can I go outside?” “I don’t know, can you?” (Child rolls eyes.) “Ma, may I…”

Many of us engage in similar mechanics on a professional level, gently commenting that garbage-in from patient motion (because the tech failed to explain the importance of remaining still and/or did not rescan, or because the ER staff failed to apply appropriate restraints) is causing garbage-out in terms of our ability to diagnose anything. Or that GI from suboptimal selection of study type by our clinical colleagues (“You don’t need contrast for this study.”) is resulting in GO when we cannot exclude abnormalities in the area of their interest.

Perhaps a spin of the Wheel of Protocols would have helped.

I am frequently tempted to go beyond such quiet coaxing, especially when my hopeful pleading seems to go completely ignored by those whose behavior I am vainly hoping to reshape. Of course, indulging such temptations would swiftly get my own behavior modified, much more harshly. Still, I can dream…

… of responding to the clinical history of “flank pain” with a thorough description of renal stones - except for which side they’re on. Heck, they didn’t say which flank, why should I?

… of reporting on a patient who’s clearly just had surgery (which is inexplicably not at all mentioned in provided history) as if I’m seriously concerned about major trauma. Look at all that free air and ascites/hemoperitoneum! The vandals even removed a length of bowel! Oh, the humanity.

… of answering “R/O obstruction” with “No obstruction.” (Then, when they call about the extremely brief report and ask about the appendix, say, “Why, yes, there’s appendicitis. But you asked about obstruction. Why, was there some other clinical information? Right lower quadrant pain and such?”)

... of responding to the history of “R/O pathology” with the Impression: “Not ruled out.” Hey, you never know for sure.

… of reporting on an ultrasound for “Pregnancy” with the word “Yes.”

… of summing up a negative head CT report in response to a history of “weakness” by recommending a good strength trainer I know at my local gym.

… of responding to a one-word clinical history of “Pain” with a one-word report: “Pathology.” Does anybody else remember that old game show with a similar cadence, Password Plus?