Thoracic Surgery Referral Intake From Pulmonology and Oncology

Thoracic surgery referral intake is not one queue. It is two, and they behave nothing alike. A pulmonology office sends a nodule follow-up with an imaging report attached and no images. An oncology office sends a surgical evaluation request with a tumor board note, a pathology report, and a date it would like the patient seen by. Your intake coordinator opens both from the same fax folder and has to know, without being told, which one is ready to book and which one is going to sit for nine days.

Most thoracic surgery practices are small. Two to six surgeons, a scheduler, an intake coordinator, and an authorization person who is also the intake coordinator on Fridays. The referral volume is not enormous, which is exactly why nobody has ever built a real process around it.

So the packets pile up in a shared folder. Someone opens them in the order they arrived rather than the order they can be acted on, and the ones missing an outside imaging disc go back into the pile to be chased later. Later usually means when the referring office calls to ask what happened.

More than three medical groups in four (76%) manage referrals in their EHR or in referral management software, so the tooling is there. What is missing is somebody with the hours to work the queue every day, in both directions, on packets that arrive in two different shapes.

Two sending specialties, two different intake paths

The referral that arrives from a pulmonology office and the one from an oncology office need different things before a slot is worth offering. One usually needs outside images pulled and loaded. The other usually arrives more complete but carries a timing expectation the referring office already communicated to the patient.

athenaOne can tell these apart before a person opens them. Referral sources are configured records, not free text, so an inbound referral can be tagged by who sent it and dropped into the right department bucket automatically. The AI reads the inbound fax or email, builds or matches the chart, files the documents under the correct document class, and creates the referral order against the sending practice.

You get a queue sorted by what it needs next instead of by what time it arrived. The pulmonology packets sit in a records-pending bucket with an outbound chase already running. The oncology packets go straight to scheduling.

The fax folder is still where referrals live

Fax has not gone away in specialty referral work and it is not going to. An MGMA Stat poll found nearly 1 medical practice in 4 (24%) does not have a digital fax solution fully integrated with their EHR and workflows, while 73% do.

That gap is where the intake problem actually sits. A digital fax that lands in an inbox is a PDF someone has to open, read, key in, and route. A digital fax that is integrated is a document that can be classified and attached to a chart without a person reading it first.

It is the same pattern as an orthopedic fax referral queue, just at lower volume and higher stakes per referral. For a thoracic practice, the difference shows up in how fast a referral becomes a booked appointment. Intake work that happens overnight means the referring office gets a status back the next morning rather than the next week.

Do not offer the slot until the packet can hold it

The failure mode in surgical referral intake is booking too early. A consult happens, the outside imaging never arrived, and the visit produces a second visit. The patient drove two hours for a conversation that had to be repeated.

This is the same timing problem that shows up as prior authorization windows in vascular surgery, where the appointment has to land inside a window somebody else controls. Minimum lead-time rules exist for exactly this reason, and most practices already run two of them: new patients cannot book inside three or four business days so paperwork and records have time to land, while established patients can book next business day. The AI applies the rule at booking time rather than hoping a scheduler remembers it, searching open slots that respect the lead time for that appointment type and provider.

When records are still outstanding, it opens a follow-up task in athenaOne against the referral, calls the sending office for the missing item, and only offers the slot once the task closes. The referral stops being a thing somebody has to remember.

Where the work hands back to a person

The AI does not decide whether a referral is appropriate, what the patient needs, or which surgeon should see them. It moves paperwork and it books time. Anything requiring a judgment about the patient goes to your surgeon or your nurse with the packet already assembled and the gaps named.

In practice the handoff is specific, and it looks a lot like pre-op packet chasing in general surgery. A referral arrives with a request for an evaluation your practice does not perform, and it goes to a named person. The tumor board note references a case discussed at a facility your practice does not have records access to, and it goes to a named person. The referring office asks for a date the surgeon’s block cannot hold, and it goes to your scheduler.

That is the honest version of what automation does to intake work. It takes away the opening of PDFs and the retyping, and hands your staff a shorter list of harder decisions.

Referral management software is not referral management

The referral tooling most groups own does a good job of holding a referral and a poor job of moving one. An MGMA Stat poll found 76% of medical groups use their EHR (66%) or referral management software (10%) to manage patient referrals, while about one in five (21%) still rely on manual tracking.

Owning the system and working the queue are different problems, and buying another point tool that automates the easy half of intake and hands the rest back is how practices end up with three logins and the same pile.

The question to put to any vendor is not what it automates. It is what happens to the referrals it cannot finish, and whether the status the referring office sees updates without somebody typing it.

Key Takeaways

Sort inbound referrals by sending specialty at intake, because a pulmonology packet and an oncology packet need different next steps.
Tag referrals to configured athenaOne referral sources and department buckets so the queue organizes itself instead of arriving as undifferentiated PDFs.
Apply your new-patient lead-time rule at booking time so a consult never happens before the outside records land.
Open a follow-up task against the referral for every missing item, and let the outbound chase run on it rather than on someone’s memory.
Judge referral automation by what it does with the packets it cannot finish, not by the ones it can.

Thoracic surgery referral volume is low enough that no one ever built a process and high enough that the gaps cost real cases. Two sending specialties, two packet shapes, one folder nobody owns. Sort the queue at intake, chase what is missing automatically, book only when the packet can hold the visit, and hand your staff the exceptions with the context already attached. That is a week of coordinator time back, and a referring office that gets an answer the next morning.

Sources