Not Following Blindly

One of the biggest logistical wastes I have seen in the world of radiology has been in the setting of follow-up studies. I’ve poked some fun in the past at the referring clinicians who are involved (imagine, you spend as much as a decade between med school and post-grad training so you can be qualified to write orders like “F/U nodule” multiple times per day). But, today, my focus is on our end of the process.

“Process,” really, is a rather generous term for how most rad groups approach it, because things rather seem defined by a lack of process. A rad who said something should be followed almost never knows if follow-up happened, outside of some structured setting like a research project or mammo practice. The technologist whose schedule includes the follow-up imaging study doesn’t necessarily know what is being followed or whether there are particular ways the study should be done to make this follow-up effective or conclusive.

The rad receiving the study, who’s expected to render a decision as to whether the thing being followed is improved, stable, worsened, or better characterized, is the one in the “hot seat” and, thus, seems to me the variable that’s most appropriately (and effectively) adjusted to fix matters.

I’m sure this isn’t a problem in some practices—perhaps academic or otherwise heavily-sub-specialized groups. I can only say that I have yet to work in such a non-problematic environment – or know anyone who is. It seems the majority of radiology groups are simply too hectic (one might also say over-burdened, even over-extended) to make this a priority.

For more coverage based on industry expert insights and research, subscribe to the Diagnostic Imaging e-newsletter here.

One remedy I’d suggest, if at all feasible, would be to make it a policy that any follow-up study gets earmarked to be protocoled and read by the rad who recommended the follow-up. That rad knows better than anyone else what was going on in his or her mind when he or she made the recommendation, and this eliminates the chance for disagreements between original and following rads. It will also cut down on the time it takes to interpret the follow-up study, since a rad following himself or herself will recognize or remember his or her own work…as opposed to some other rad having to review the prior case.

Of course, that’s not always feasible. Even if the follow-up study is routine or non-emergent, sometimes the original rad will be unavailable, or the follow-up study will be at some facility other than the one the original rad covers. (Though, I would suggest that part of this effort could include making sure the patient got scheduled at one of the rad’s facilities, at a time when he or she would be around.)

The next-best thing would be to earmark an interpreting rad when the follow-up study is scheduled, so he or she can protocol it to whatever specifications he or she feels will be necessary to do the follow-up properly. Again, this might not be an issue in some groups where rads routinely get to protocol their own studies, but I’ve yet to see that outside of my residency or fellowship institutions.

It should go without saying that any study explicitly performed for follow-up gets done with all relevant priors ready for comparison and review by the interpreting rad. Priors done at another facility? Have them requested/delivered/fetched before the current scan gets done. The whole “Priors not available, referrer insists on a read now and will expect a detailed addendum later” is a duplication of work that should be avoidable the vast majority of the time.

So, for instance, I get an alert that there’s a pending “follow-up” case that will be mine to read, and here’s my chance to protocol it. I can now look over the prior study and see what I think is needed. Is it a nodule in the dependent portion of lung, where incomplete inspiration might obscure it? Maybe I want the CT tech to position the patient prone and urge him to make sure the patient does a maximum inspiration for the scan.

Maybe the original study was a dynamic-contrast liver MR, but I can see that the follow-up has been scheduled non-contrast. Now I have a chance to get the study modified to include IV enhancement. Or, maybe I think another imaging modality would provide more answers than just re-doing the last type of scan.

Maybe I don’t think the study should be done right now—the prior recommended follow-up in six months, and it’s been a couple of weeks. I, now, have the chance to get the referrer on the phone and see if this was an error, or if he or she has another reason to re-scan early.

Maybe I don’t think the study should be done at all—the original rad didn’t suggest follow-up, but some referrer is doing it anyway—again, a chance for me to get on the horn with the referrer and possibly spare the patient some radiation. Or, the original rad recommended follow-up contrary to guidelines of the Fleischner Society or Lung-RADS. Perhaps I don’t even see the lesion he or she is referring to, and I’d like him or her to point it out.

All of this amounts to more work on our part, in exchange for zero extra reimbursement. Heck, it might even result in a little net-loss if we canceled or delayed a few studies—although that could possibly be counterbalanced by more efficient reading of the follow-up studies when they do happen. I can also imagine a reduction of medmal-exposure with this approach.

Most importantly, I have a hard time seeing how it could fail to improve patient-care, and referrers might just come away with the impression that, when we recommended “follow-up,” we’re actively managing it as a plan, rather than some knee-jerk CYA maneuver.