Behavioral Health No-Shows Cost More Than an Empty Slot

Behavioral health no-shows get benchmarked against numbers that were never about behavioral health. Someone quotes a group rate, your practice is above it, and the conversation turns into a reminder policy discussion. That skips the part that decides whether it matters, which is what one missed visit is actually worth in a practice where the appointment is long, the panel is fixed, and the replacement has to be somebody you already know.

The pricing is different at both ends. A behavioral health visit is typically 45 or 60 minutes rather than 15, so a single miss opens a hole four times the size of a miss in a medical clinic. And the hole is harder to fill, because the patient who can take it is usually an established patient of that same provider, not whoever happens to be on a waitlist.

Meanwhile the practice is running a queue it cannot get to. A provider may be capped at a small number of new intakes a day while a very large follow up panel sits behind them waiting for time. The schedule is carved so a 45 minute hole opening at noon cannot absorb a new patient who is ready to book today, even though that new patient has been waiting three weeks.

So the empty slot is not just lost revenue for the hour. It is capacity that existed, that somebody needed, and that nobody could reach in time.

The benchmark you are being measured against is the wrong denominator

Group level no-show benchmarks are built mostly from medical practice data. MGMA DataDive Practice Operations data show the single-specialty aggregate no-show rate, combining primary care, nonsurgical and surgical single-specialty practice data, fell to 5.55% in 2020, stayed at 5% in 2021 and 2022, and rose to 6.81% in 2023, close to the pre-pandemic benchmark of 7% recorded in 2019.

Those are useful numbers for a primary care group deciding whether it has a problem. They are the wrong denominator for you, because almost none of the practices behind them run 45 or 60 minute visits against a fixed continuing panel.

What is worth tracking instead is narrower and entirely yours. The rate on first visits against the rate on continuing visits. The rate by lead time between booking and appointment. The rate on slots that were rebooked once already.

Those three tell you where to spend effort. A single practice-wide percentage compared against a cross-specialty aggregate produces either false comfort or a panic that leads nowhere.

Published psychiatry rates sit in a different band

There is no single behavioral health benchmark you can hold your practice to, and it helps to know that before somebody hands you one.

A study of an academic psychiatry practice, published in Psychiatric Quarterly, notes that psychiatry clinics have long contended with higher no-show rates than other settings. That clinic recorded an overall no-show rate of 18.1% before the period the study examined, and 15.3% during it after care shifted toward telehealth.

Those are one practice’s numbers rather than a national benchmark, and that is the point. Rates in this specialty depend heavily on visit mix, intake volume, lead times, and how a practice defines a miss in the first place. Two practices delivering identical care can report rates several points apart because one counts late cancellations and the other does not.

So stop treating the rate as a score and start treating it as an input. Your rate tells you roughly how many hours a year are at risk. What you do about those hours is a separate decision, and it is the one that pays.

The fee is a policy, not a recovery

The first lever most practices reach for is a no-show fee, and it is more common than it used to be. A Jan. 7, 2025, MGMA Stat poll found 42% of medical group leaders reported their practices using a no-show fee, against 58% that do not, from 622 applicable responses.

A fee is a reasonable policy and it is worth having. It is also not a recovery mechanism. Charging for the missed hour does not put anybody in the room, and in a practice built on continuing relationships the collection conversation itself has a cost your front desk absorbs.

Recovery is a different activity with a different clock. It starts the moment an appointment is going to be missed and it ends when somebody else is booked into that time, which in practice means it has to happen in minutes rather than once the day is over.

That is where the automation earns its keep, because the recovery window is short, the outreach is repetitive, and it always falls to whoever at the front desk has a free moment, which on a busy day is nobody.

Recovery is a rebooking problem, not a reminding problem

Reminders reduce misses at the margin. They do nothing about the slot once a miss happens, and most practices have no motion at all for the hour that just opened.

Inside athenaOne the raw material is already there. The booked appointment list for the department tells you what was scheduled, what is still expected today, and what has gone quiet. The gap is not a mystery, it is a query, and it can be read continuously instead of once the day is over.

Once the gap is visible the work is matching. Which established patients of that provider are overdue, are already authorized where authorization applies, and can hold a visit of that length and type. Those patients get called and offered the time, in order, and the first one to say yes gets the appointment moved into it.

The patient who missed also needs a rebooking, and that is the same motion in the other direction. The visit gets rescheduled rather than cancelled, which keeps it as one appointment with a history rather than a gap in the record and a new booking that looks unrelated. None of that requires the AI to judge anything about the patient. It is looking at what was booked, what is open, and who is administratively eligible to fill it.

The complication: two lead time rules that pull against each other

Here is the constraint that breaks naive backfill in behavioral health practices. Two different minimum lead times usually coexist. New patients cannot book inside three or four business days, because intake paperwork and benefit checks need to complete first. Established patients can often book for the next business day, or same day, because none of that applies to them.

That rule is correct and it should stay. It is also exactly why the slot that opened at 9:40 this morning cannot go to the new patient who has been waiting three weeks, which is the outcome everybody wants and nobody can have.

A provider ramping up may also be running a cap that expires on a date, four new intakes a day for the first two weeks and unlimited after. The cap has to lift itself on schedule, because nobody remembers to remove it, and a stale cap silently throttles the new patient pipeline for months.

The automation has to respect both rules rather than route around them, which means today’s recovered slot is offered to the established panel and the new patient queue gets worked against slots far enough out to be legal. Where it cannot resolve something, it stops. A slot nobody eligible can take, a patient whose authorization has not returned, or a series that has now been missed repeatedly all go to a named person with the history attached. Those are the calls a scheduler should be making, and there are only a few of them a day once the routine ones are handled.

Doing the arithmetic on your own numbers

The reason to price this rather than benchmark it is that the answer decides how hard to work the recovery.

Illustrative model, figures are assumptions, not measured results. Consider a practice with 8 providers, each holding 30 patient hours a week, at an average of $150 collected per visit hour, running a 15% no-show rate. That is 8 x 30 x 52 = 12,480 patient hours a year, of which 15% is 1,872 hours, and at $150 the exposure is $280,800 a year. If a recovery motion refills a third of those hours with patients already on the books, it returns 624 hours, or $93,600. Your own visit length, collection rate, and no-show rate will move that a long way in either direction, and the point is the shape rather than the total.

Run it with numbers from your own practice management reports and the decision usually makes itself. If the recovered hours are worth less than the effort, keep the fee policy and stop there. If they are worth more than a staff line, the question is only whether anyone has minutes to do the calling, and the honest answer at most practices is no.

What matters is measuring recovery separately from prevention. Reminder performance and slot recovery are two different programs with two different numbers, and reporting them as one no-show rate hides which of them is working. That is the same discipline as treating a recurring series as the unit rather than a single visit, and it pairs with working the intake waitlist on the other side of the schedule.

Key Takeaways

Stop comparing your rate to cross-specialty aggregates and track first visits, continuing visits, and rebooked slots separately.
Treat a no-show fee as policy, not recovery, because charging for the hour does not put anyone in the room.
Read the department’s booked appointment list continuously so the gap is visible in minutes rather than at end of day.
Offer the recovered slot to established patients of that provider, since new patient lead time rules usually make same day impossible.
Reschedule the missed visit rather than cancelling it, so the appointment keeps its history instead of becoming an unrelated new booking.
Let ramp-up caps on new intakes expire on their own date, because a stale cap throttles the new patient pipeline invisibly.
Price the exposure with your own visit length and collection numbers before deciding how much recovery effort is worth funding.

The empty behavioral health slot is expensive because it is long, because the panel that could fill it is narrow, and because the rules that protect your intake process are the same rules that stop you filling it today. None of that is fixed by a better reminder. Watch the booked list, move on the gap in minutes, offer it to the patients who are actually eligible, rebook the person who missed, and hand your scheduler the handful of cases that need a human. The rate may barely move. The hours you get back will.

Sources