At times, image ordering by clinicians seems downright random. Courtesy of roulette and certain game shows, I propose the Wheel of Protocols.
“The ACR Appropriateness Criteria® are evidence-based guidelines to assist referring physicians and other providers in making the most appropriate imaging or treatment decision for a specific clinical condition.”
Gosh, that sounds nice. I wonder how many clinicians out there know and/or care about this resource.
Some are downright hostile to the notion of radiologists having input into the studies they order. They’ll tell you they know their patients, and shouldn’t have to explain themselves when they order what appears to be a particularly unhelpful protocol for the purported clinical history. Some of them will try to explain how “ER throughput” trumps the importance of performing the right exam for individual patients.
They’ll further add that you, the radiologist, should have no problem making or ruling out the diagnosis of interest, no matter how their imaging choices have hamstrung you. Some will try to get academic, citing a study conducted by non-radiologists. I even had one try to tell me that the ACR’s criteria didn’t hold water because they were consensus statements, and not backed by rigorous, double-blinded claptrap.
I’ve seen enough noncontrast chest CTs for “R/O PE” and pre/postcontrast full-body scans on preteens for “pain” to start wondering whether there’s any conscious decision making going on at all. At times, it seems downright random. Basic physics (about which, as radiologists, we know a thing or two) tells us that trying to impose order on entropy is an uphill battle. Instead, I suggest introducing a different type of randomness.
Courtesy of roulette, the (fortunately) last Mad Max movie, and certain game shows, I propose the Wheel of Protocols. There can be a wheel for each body part. Sections on the Abdomen-CT wheel can be allocated for Noncontrast, PO only, IV only, PO/IV, etc. Not all sections need be equal; for instance, since even ER docs will agree that the need for a multiphasic liver scan is rare, that can be a thin sliver of the pie. Meanwhile, since noncontrast scans are really popular with clinicians, they’ll accept the wheel much more readily if that section is nice and thick, maybe a one-fifth section of the thing.
To make things interesting, there can be a couple of sections like “Radiologist’s choice” and “Spin again.” Maybe even a “Jackpot” which grows a buck or two each spin. An overworked clinician might be willing to give up some perceived autonomy if there’s a chance of having his lunch paid for. (At least, that’s what we were told when reps stopped being allowed to give us meals or even pens.) The inducement might not be all that necessary - people like games. Especially if you dress things up with flashing lights and carnival tunes.
Much like carnival attractions, of course we’d quietly give ourselves an edge. The wheel could have an adjustable bias to make it land on Radiologist’s Choice a little more frequently than it otherwise might. Not too much - mustn’t get caught. Maybe the bias could be adjusted based on which clinician was on-shift that day; if you see it’s going to be a particularly dopey NP, you might crank it up to get away with more chicanery.
Even with a conservative setting, you might be able to pull off choosing the protocols on one in 10 of the studies coming to you from the ER. If you’re currently deciding on none of them, that’s a step in the right direction.