Imaging Results and the Callback Queue Nobody Owns

An imaging center takes a large number of calls a week about imaging results it is not the one to release. The patient had the scan at your site, so they call your number, and your staff spends the conversation explaining a division of responsibility the patient did not know existed.

That queue is expensive and invisible. It generates no revenue, resolves nothing for the patient, and consumes the same front desk that is trying to book tomorrow’s schedule. Worse, the calls that genuinely need action look identical to the ones that do not until somebody has already spent four minutes on the phone.

Two different calls wearing the same words

Every call in this queue opens the same way, and they split into two very different jobs.

The first is a patient who wants to know what their scan showed. Your center cannot answer that, because the report goes to the ordering provider and that provider decides what the patient is told and when. The right handling is fast, courteous, and identical every time: confirm the study was completed, confirm the report was sent and when, name the practice that will be calling, and offer to send the report on to them again if it may have been missed.

The second is a patient whose study never happened, whose prep instructions were wrong, whose authorization lapsed, or who needs a repeat study the report recommended. Those are yours, and every one of them has revenue or a scheduling action attached.

When both call types land in one queue, the second gets the leftover attention of staff worn down by the first. Splitting them at the front of the call is most of the value, and it is pure routing rather than interpretation.

The follow-up gap is well documented

The pattern behind these calls is not unique to your center.

Work on imaging specifically has found that patients for whom radiologists recommended additional imaging were frequently lost to follow-up. From an imaging center’s point of view that is a recommendation sitting in a report that never became a scheduled appointment.

The broader notification record is no more reassuring. A review of more than 5,400 patient records in primary care found a 7.1% rate of failure to inform patients of clinically significant outpatient results, or to document having done so. Read that second clause the way an operator should: delivery and proof of delivery are one obligation, and your center is on the hook for the proof even when the conversation belongs to someone else.

That matters commercially and not only clinically. A recommendation for additional imaging is a study your center is positioned to perform, and it goes unbooked because closing that loop is nobody’s assigned job. The referring office assumes the patient was told. The patient assumes silence means nothing was needed.

The complication: the report changed after you sent it

Here is the failure mode that survives even a well-run notification process.

A report is finalized and delivered. Later it is amended, because a comparison study arrived, a second reader disagreed, or a correction was made. The amended version goes into the record. Whether anyone tells the referring office a second time depends entirely on local habit, and the original delivery already came back marked sent.

The result is a referring practice acting on a version of the report that has been superseded, with no signal that anything changed. Nothing in the workflow is broken in a way that shows up on a report, which is precisely why it persists.

Subscribing to changed imaging results turns that into an event rather than a discovery. When a report is amended, the automation re-delivers it to the referring practice, records the second delivery separately from the first so the audit trail shows both, and calls the referring office when the change is material enough that the practice has said it wants a call.

The boundary stays firm. The automation does not read the report or characterize what changed. It knows that this document was superseded, and its job is to make sure the people who received the first one receive the second one. Deciding what the change means belongs to the radiologist and the ordering provider.

Turning a recommendation into a booked study

The most valuable thing in the callback queue is the recommendation for follow-up imaging, and it needs an order before it can become an appointment.

That is the real bottleneck, and it is administrative all the way through. The recommendation lives in your report. The order has to come from the referring provider. The patient sits between two offices, each assuming the other is handling it.

The workable sequence starts with the referring office rather than the patient. A request for the order, sent with the report reference and the recommended study already identified, converts far better than a patient asking their doctor for something they only half understood. Once the order exists, the patient call has something concrete to offer, which is a time.

Where the order does not come back, the useful behavior is a bounded chase with an end and an escalation, not an indefinite queue. Two attempts and then a report to the practice manager showing which referring offices have outstanding recommendations is more effective than a task that quietly reopens forever.

Everything a licensed person owns stays with them. Your center recommends, the referring provider orders, and the automation moves the paperwork between the two and books the result.

What to instrument first

Three numbers turn this from an impression into an operation, and none of them exist in a standard report today.

Count the result calls your center takes and split them by whether your center could act on them. That ratio tells you how much of your phone capacity is spent on a conversation that ends in a referral to somebody else, and it is usually larger than anyone guesses.

Count amended reports and how many were re-delivered. If the second number is not close to the first, you have found a silent gap worth closing this quarter.

Count recommendations for additional imaging and how many became scheduled studies. That is the number that pays for the work, and it is the one most imaging centers have never produced.

Key Takeaways

Split result calls at the front into ones your center can act on and ones it cannot. The second group needs a fast scripted handoff, not staff improvisation.
Never let the automation characterize a report. It confirms the study happened, states when the report went out, names who will call, and routes everything else.
Subscribe to changed imaging results so an amended report triggers re-delivery. A superseded report that was marked sent is the failure nobody detects.
Chase the order from the referring office before calling the patient, because a recommendation without an order cannot become an appointment.
Give the chase a bounded end and an escalation report by referring office. Indefinite tasks reopen forever and get worked never.
Measure recommendations for additional imaging that became scheduled studies. It is the number that funds the whole workflow and almost nobody has it.

The imaging results callback queue is treated as an unavoidable cost of being the place the patient physically visited. Most of it is routing, and the part that is not routing is unbooked volume your center already earned. An AI team working imaging result documents inside athenaOne can separate the two, re-deliver the amended reports that quietly superseded the originals, and chase referring offices for the orders that turn recommendations into scheduled studies.

Sources