At a recent MGMA Summit Operations Discussion Group, the conversation turned to a familiar pain point: cancellations and the automated waitlist tools meant to recover them. Our staff facilitators asked whether anyone had found "magic" for filling canceled slots. One administrator's answer was more useful than magic — it was a set of rules.

Their practice uses an automated waitlist tool (Epic's Fast Pass) to backfill openings, but the value, the administrator explained, isn't in the automation itself. It's in the constraints layered on top of it.

For their combined neurology and neurosurgery service, the tool isn't allowed to book a patient into any slot opening less than 72 hours out. If a 3 p.m. complex visit cancels at 8 a.m. the same day, the system leaves it open rather than auto-filling it. And every fill is tied to the provider's template, so a new-patient slot never gets backfilled by a hospital follow-up just because the two happen to run a similar length.

That's the insight most discussions of waitlist technology miss. Having a backfill tool does not add value without rules governing how it fills, and those rules add the most value when they are specialty- and visit-type-specific. Otherwise, backfill tools may quietly work against you.

The stakes are real. New-patient wait times averaged roughly 31 days across 15 major metro markets in 2025 — up 19% since 2022¹ — which means every cancellation you recover is found capacity in a system with very little to spare. MGMA’s own benchmarking sharpens the point: appointment cancellations climbed sharply in 2024, and only about 27% of canceled visits were rebooked within 30 days, down from roughly 62% the year before.² Leaders told MGMA that overbooking and waitlists are among the levers they're leaning on for patient access in 2026.³ But a recovered slot only helps if the right patient lands in it.

The hidden cost of "fill every gap"

The intuitive way to configure a backfill tool is the most aggressive one: any opening, filled by anyone on the waitlist, as fast as possible. It looks efficient on a dashboard — your cancellation recovery rate climbs. The problems show up everywhere else.

A patient who can take a complex new-patient slot on three hours' notice usually isn't ready for it. They may not have completed intake, transferred records or imaging, or arranged the prep a complex visit requires. The slot gets "filled," but the visit is low-value or has to be redone. Worse, when the tool ignores visit type, it drops the wrong kind of appointment into a slot built for something else — a 15-minute follow-up into a 40-minute new-patient block, or the reverse — and the day's template starts to buckle. You've traded an empty slot for a misused one, and your schedulers spend the time they saved cleaning up the mess.

Designing backfill by specialty

The fix is to treat backfill as an operational design problem. Three rules do most of the work:

1. Set lead-time floors by visit complexity: The 72-hour floor for neurosurgery is a good model: the more prep a visit requires, the longer the minimum notice before the tool is allowed to auto-fill it. Complex, procedural and prep-heavy visits need a generous floor; routine, low-prep visits — an established-patient recheck, a rash, a medication follow-up — can be filled same-day and benefit from it. The floor isn't one number you set once; it's a per-visit-type setting.
2. Tie every fill to the template's visit type: Slot duration is a proxy, not a definition. A tool that matches on clock time alone will eventually drop a hospital follow-up into a new-patient slot because the minutes line up. Configure auto-fill to respect the visit-type definitions already built into your templates, so duration, documentation and workup expectations travel with the slot.
3. Account for mixed provider types: With 48% of groups reporting they've added advanced practice providers (APPs) relative to physicians in 2025,⁴ your backfill logic now has to know which openings an APP can appropriately take. Routing eligible overflow to an APP slot can recover access without overbooking a physician — but only where scope and visit type genuinely fit, so the rules have to encode that distinction.

Where overbooking pays off

Backfill and overbooking are different tools for different problems, and they don't belong in the same specialties. Backfill recovers a known opening; overbooking deliberately schedules beyond capacity to absorb expected no-shows. MGMA’s own access guidance lands in the same place: overbook only where it’s clinically appropriate, and recover canceled visits by removing friction rather than overbooking indiscriminately.⁵

Overbooking earns its keep in high-volume, low-complexity, predictable-no-show settings — established primary care, some behavioral health, routine recheck visits — where the no-show rate is stable enough to forecast and the cost of an occasional overlap is low because visits are short and rooming is flexible. It's actively dangerous in procedural and resource-intensive settings. Double-booking a surgery consult, an infusion chair or any visit that ties up a room, equipment or another clinician can't be absorbed when both patients show; you waste an expensive resource and damage the experience. In those specialties, a vetted waitlist with lead-time floors is the right option.

The tool is only as good as the rules behind it

It's worth naming the trap directly, because it's where practices lose the value. The waitlist engine — Fast Pass or any competitor — is just plumbing. Everything that makes it work or backfire lives in the configuration: the lead-time windows, the visit-type mapping, which templates participate and how far ahead fills are offered. A group that switches it on with blanket defaults gets the failure mode described above. A group that maps its rules specialty by specialty before configuring gets recovered capacity without schedule chaos. The work is operational rather than technical, and it's the part you can't always hand off to the vendor.

Start narrow, and use what you already have

None of this requires new software or a budget line. The first move costs an afternoon. Pull a sample of the slots your tool backfilled over the past month — the report should already exist in your system — and mark how many were the right visit type and how many got reworked, rescheduled or ran over. That ratio is your baseline and your case for adding rules.

Before you change a setting, talk to your schedulers. They already know which backfills blow up the day, and 20 minutes with them surfaces the real failure patterns faster than any report. They also have to live with whatever rules you set, so bringing them in early makes the rules both cheaper to build and more durable.

• Start with a deny-list, not a full map. You don’t have to define every visit type across every specialty before you get value. Name the handful of slots that should never auto-fill on short notice — complex new patients, procedures, infusion chairs — and exclude just those. One afternoon of configuration removes most of the downside while you work on the rest.
• For lead-time floors, three buckets cover most schedules: same-day is fine, a 24-to-48-hour minimum, and a 72-hour floor for prep-heavy visits. Sort your visit types into those three rather than assigning a custom number to each one. You can refine later; you rarely need the precision up front.
• Pilot the rules on one service line before rolling them everywhere — the specialty where backfill hurts most, or the one whose lead is most willing. Watch it for a few weeks, fix what breaks, then expand with a working example instead of a theory.

For the overbooking question, the data you need is already in your PM system. No-show rates by visit type tell you which clinics are predictable enough to overbook safely — you’re reading a report you already run, not buying anything to find out.

Put a 30-minute review on the calendar each quarter to check the floors against what’s actually happening. The rules aren’t set-and-forget, but they aren’t a daily chore either.

Measure appropriate fills, not filled slots

Cancellation recovery rate — the share of openings the tool filled — is a vanity number on its own, because it counts the wrong fills alongside the right ones. Track instead the share of backfills that matched the intended visit type and didn't overrun, get reworked or get rescheduled. Watch fill latency (how quickly gaps close) and the downstream signal that actually tells you access is improving: time to third-next-available appointment, measured by specialty rather than as a single house figure. 

It's also worth tracking the no-show rate of backfilled slots specifically, since waitlist patients pulled in on short notice can behave differently than patients who booked weeks out. A concrete companion marker is the 48-hour fill rate — the share of near-term openings closed within two days — which MGMA’s guidance puts at 85% or better.⁶

Set lead-time floors that scale with visit complexity, tie every automated fill to your template's visit types, and reserve overbooking for the high-volume, low-stakes corners of your schedule where the math works. The tool will recover real capacity — but only if the rules behind it know the difference between a filled slot and the right one.

Notes:

1. AMN Healthcare, 2025 Survey of Physician Appointment Wait Times (May 2025). https://ir.amnhealthcare.com/news-releases/news-release-details/new-survey-shows-physician-appointment-wait-times-surge-19-2022 
2. MGMA, 2025 MGMA DataDive Financials and Operations data report, September 2025. https://www.mgma.com/2025-financials-and-operations
3. Harrop C. "Patient access priorities for 2026: Tackling wait times, phones, no-shows and more.” MGMA. Dec. 10, 2025. https://www.mgma.com/mgma-stat/patient-access-priorities-for-2026 
4. Harrop C. "APP utilization and how care team redesign continues into 2026." MGMA. Nov. 12, 2025. https://www.mgma.com/mgma-stat/app-utilization-and-care-team-redesign-in-2026 
5. 2025 MGMA DataDive Financials and Operations data report.
6. Ibid.