Where ENT Referrals Die Between Approval and Appointment

Every ENT referral that goes wrong gets blamed on the fax queue, and most of them do not die there. They die later, in the quiet stretch after the authorization comes back approved and before anyone books the patient. Nothing is broken in that window. There is just no task assigned to it, so it holds referrals indefinitely and nobody notices until a referring office asks.

The reason this gap survives is that it looks like success from every angle. The referral was received, so intake did its job. The authorization was obtained, so the auth team did theirs. The schedule has open slots, so scheduling has capacity. Each function reports green while the patient sits unbooked, because the handoff between the last two is a human remembering, and the person remembering is also working the next authorization.

The wait the patient experiences is not the wait you measure

Most practices measure access from the moment a patient calls to book. The patient measures it from the moment their doctor said they needed to see someone.

New-patient access is not improving on its own. In a July 1, 2025 MGMA Stat poll, 40% of groups reported no change in new-patient wait times, 31% reported longer waits, and 26% reported shorter ones. Groups working the problem described adding providers, increasing provider days, adjusting templates, and opening same-day slots.

Those are all capacity levers, and they are expensive. The approval-to-appointment gap is not a capacity problem, which is what makes it worth attacking first. The slots exist. The patient is approved to use them. The two facts have simply not been introduced.

For an ENT practice this window is often the largest single component of the patient’s real wait, and it is entirely invisible in a report that starts the clock at booking.

Approval is a milestone, not an appointment

The fix conceptually is to stop treating authorization as the finish line of one workflow and start treating it as the trigger for the next.

When an authorization comes back, three things are true at once and can be acted on immediately. There is a patient, there is an approved service, and there is a set of open slots that match the appointment type that service requires. Reading the referral authorization record against the open-slot search turns approval into a booking attempt rather than into a queue entry.

The practical implementation is an outbound call the same day, offering specific times, rather than a note asking the patient to call the practice. Referral patients are the least likely population to call, because from their side the referral was something their doctor arranged and they are waiting to hear from you.

Where the call does not connect, the useful pattern is a bounded sequence across different times of day, then a message with a way to book, then escalation to staff with everything already gathered. What should not happen is the referral quietly returning to a queue that nobody works on Fridays.

The plan can invalidate a booking you already made

This is the complication that turns a tidy workflow into a real one, and it comes straight from a group living with it.

On one large plan’s HMO product, every change of the referred-to physician requires a phone call to the payer. That includes the situation where the originally referred physician has stopped accepting patients, which is not a rare event and is entirely outside the patient’s control. So the practice cannot simply move the patient to the colleague with earlier availability, even when doing so is obviously better for everyone.

In ENT this bites hardest on the referrals that matter most. Surgical consults and hearing-related services often carry authorization tied to a specific provider, and the practice’s instinct to load-balance across the group is exactly the action that breaks the authorization.

The workable behavior is to check the constraint before offering times rather than after. Where the authorization is provider-specific, offer only that provider’s slots. Where a change is genuinely needed, generate the task for staff to make the payer call, and tell the patient honestly that the timeline depends on it. The automation is applying a rule about what the plan permits, and handing the payer conversation to a person, which is where it belongs.

Authorizations expire on a clock nobody is watching

The second failure mode in this window is slower and costs more, because it surfaces after the visit.

A specialist referral for a chronic condition may be valid for a year, and then somebody has to notice and renew it. There is rarely a role that owns noticing. The renewal gets done when a claim is denied, which means the practice has already delivered care it will now fight to get paid for.

ENT has several workflows that run past a typical authorization window. Allergy immunotherapy courses extend over long periods with repeated visits. Post-surgical follow-up series and ongoing hearing-related care behave similarly. Any of them can outlive the authorization that started them.

This is a monitoring job, which is to say it is the kind of work automation is genuinely good at and humans are genuinely bad at. Read the authorization records, find the ones approaching expiration with visits still scheduled against them, and raise the renewal task before the visit rather than after the denial. Nothing about that requires a clinician, and skipping it is a pure revenue leak.

Instrument the gap before you try to close it

Practices that attack this without measuring it first tend to fix the part that was already working.

The number to produce is the distribution of days between authorization approval and booked appointment, for referrals over the last quarter. Not the average, the distribution, because the average will look tolerable while a tail of referrals sits at sixty days and beyond. That tail is the actual problem and it is where the referring-office complaints come from.

Produce the same view for referrals that never resulted in a booked appointment at all. Most ENT practices are surprised by the size of that group, and every one of them represents an authorization somebody worked for nothing.

Once those two numbers exist, the business case writes itself and the improvement is measurable in weeks rather than quarters. It is also the number worth reporting back to your referring offices, because being able to say what your median approval-to-appointment time is puts you in a very small category of specialty practices.

Key Takeaways

Measure the patient’s wait from referral, not from booking. The approval-to-appointment window is usually the largest invisible component of it.
Treat authorization approval as the trigger to book, not the end of a workflow. Read the authorization record against open slots and place the call the same day.
Call the patient with specific times rather than asking them to call you. Referral patients are waiting to hear from you and rarely initiate.
Check provider-specific authorization constraints before offering times. On some plans, moving the patient to a colleague requires a payer call first.
Hand payer conversations to staff. The automation applies the rule about what the plan permits and generates the task.
Monitor authorizations for expiration against scheduled visits. Long ENT workflows like immunotherapy courses routinely outlive the authorization that started them.
Produce the distribution of approval-to-booking days, not the average, plus the count of approved referrals never booked. The tail is the problem.

The gap between an approved ENT referral and a booked appointment is not a capacity problem, a staffing problem, or a fax problem. It is an unassigned handoff, and it is costing the practice patients it already won and authorizations it already worked. An AI team reading referral authorizations and open slots inside athenaOne can turn approval into a booking attempt the same day, watch the expiration clock, and escalate the payer calls to the people who should be making them.

Sources