Infectious Disease Referral Intake and the Clock It Starts

Infectious disease referral intake has a specific and unforgiving property: the referring clinician has usually already told you how fast this needs to move, and that message is sitting in a fax queue where nobody has read it yet.

ID practices are small. One to five physicians is typical, and the practice is consult-driven, which means almost all volume arrives as a referral from somebody else rather than as a patient who called.

That combination produces a front office where the entire new-patient pipeline runs through one inbound queue and one or two people. When those people are busy, the queue does not slow down. It just stops being read.

The referring side experiences this as silence. They sent a patient, they marked what they needed, and then nothing happened that they can see. Their next step is to call, which consumes more of the same staff time that caused the delay, or to send the next patient somewhere else.

The practice experiences it as a paperwork backlog. Both descriptions are correct, and the fix is the same: read the queue continuously, act on what the referring clinician already wrote, and tell them what happened.

The queue is still paper in most of the specialty

Before designing anything, it is worth being honest about the channel these referrals arrive on.

A March 2026 MGMA Stat poll found that nearly one practice in four, 24%, report not having a digital fax solution fully integrated with their EHR and workflows, while 73% do and 3% were unsure. Fax has not gone away in referral-driven specialties, and a partially integrated fax is arguably worse than an unintegrated one, because it looks solved on an org chart while a person still opens each document.

The work sitting on that channel is completely mechanical. A document arrives, somebody reads it, decides which patient it belongs to, creates the chart if there is not one, files it to the right document class, and puts it in a department bucket. Then somebody calls the patient.

Every step there is describable as a rule, which is why it automates well. Ingest, classify, match or create the chart, capture the referring source, route to the bucket the practice actually uses. What is left for a person is the part that needs judgment about the practice’s own capacity, and that part is much smaller than the queue suggests.

The referring clinician already set the priority

This is the sentence that keeps the workflow inside administrative scope, and it is worth putting in writing before anyone configures anything.

A referral arrives carrying a priority the referring clinician assigned. That marking is a fact about the document. Reading it, routing on it, and moving faster because of it is document handling. Forming an independent view about how quickly a patient needs to be seen is not, and it is not something a front-office system should ever do.

So the rule is narrow and easy to audit. If the referral is marked time-sensitive by the sender, it goes to the top of the queue and the patient is contacted the same day. If the practice’s own clinicians have set standing guidance about which referral types get expedited, that guidance is configuration they own and can change. Anything ambiguous goes to a person.

The reason to be strict about this is practical as much as regulatory. Staff will not trust a system that appears to be making calls it is not qualified to make, and a system that only ever acts on what a clinician already wrote is easy to explain, easy to review, and easy to defend.

Credentialing is what actually blocks the booking

The bottleneck in a small specialty practice is usually not the calendar. It is which physician can legitimately see this patient under their plan.

Provider enrollment resolves per provider, per payer, and often per state, and the grid describing it almost always lives in a spreadsheet outside the EHR. Nothing in scheduling knows about it. So a patient can be booked, seen, and treated, and the claim dies weeks later, which in a three-physician practice is a material amount of revenue against one avoidable error.

Many practices cannot hand over a clean enrollment grid because theirs is genuinely messy. The workable artifact is the inverse: for each provider, the list of plans they do not take. That is a document a practice manager can usually produce in an afternoon, and it is enough to prevent the expensive mistake.

With that in place the intake call gets simpler. Confirm coverage, check it against the exclusion list, and offer times only with providers who can actually be billed for this patient. When the answer is that nobody in the practice is enrolled, that is a conversation with a person, not a booking, and it should reach a coordinator the same day rather than surfacing at check-in.

Close the loop back to the sender

The referring office is the customer in this specialty, and their experience of you is almost entirely about whether they hear anything.

An MGMA Stat poll found that more than three medical groups out of four, 76%, use their EHR at 66% or referral management software at 10% to manage patient referrals, while about one in five, 21%, still rely on manual tracking. Either way the referring practice is tracking what it sent, not what happened. The status only exists on your side.

Sending it back is cheap. Received, patient contacted, appointment scheduled for this date, or unable to reach after three attempts. That last one is the most valuable message you can send, because the referring office frequently has a working phone number or a family contact that you do not, and they can solve in one call a problem that would otherwise consume five of yours.

Doing this consistently is also the cheapest referral marketing available in a small specialty. Practices that are easy to refer to get more referrals, and easy almost always means responsive rather than sophisticated.

What stays with a person

The boundaries here are clean, and naming them is what lets a small practice turn this on without anxiety.

The automation reads inbound referrals, files them, creates or matches the chart, captures the referring source, checks coverage against the practice’s exclusion list, calls the patient, books the visit against the correct appointment type, and reports status back to the sender. It works the queue continuously rather than when someone has an hour.

It does not decide how sick anyone is, it does not rank patients by anything other than what the referring clinician marked and what the practice’s own clinicians configured, and it does not answer questions about care. A patient who asks one gets a person.

Authorization requirements are worth mentioning as an adjacent workstream rather than folding into intake. Federal rules on prior authorization processing and decision timelines are changing under the CMS interoperability rule, and practices should track how their payers implement it. For the front office the immediate effect is the same as always: find out early whether an authorization is needed, because discovering it at check-in costs the appointment.

Key Takeaways

Treat the inbound fax and referral queue as a continuously worked channel, not as a stack somebody clears when they have time.
Route on the priority the referring clinician assigned, and never let the front-office system form its own view of urgency.
Build the per-provider list of plans not accepted, which is far easier to produce than a full enrollment grid and prevents the expensive booking error.
Offer appointment times only with providers who can actually be billed for that patient’s plan.
Send status back to the referring office at every step, especially when the patient cannot be reached.
Check authorization requirements at intake rather than at check-in, where discovering them costs the appointment.

A small infectious disease practice does not need a bigger front office to be easy to refer to. It needs the inbound queue read continuously, the priority the sender already wrote acted on, and a status message going back the other way.

Sources