Sizing your recall backlog: how many patients are overdue, and what they're worth

What counts as the recall backlog

A patient belongs in the backlog when three things are true: they are an active patient of the practice, they were expected to return by a certain date, and that date has passed with no future appointment scheduled. Leave any one of those out and the number gets distorted.

"Active" is the part practices skip. A patient last seen eight years ago who moved across the country is not a recall opportunity; they are a closed chart. Pick a lookback window that fits your specialty, for example patients seen at least once in the last 36 months, and use it consistently.

"Expected to return" also varies. A dental hygiene patient is usually expected back on a fixed interval. A primary care patient with a chronic condition may have a follow-up interval set by the provider. A physical therapy patient may have remaining visits on a plan of care. Use the interval your providers already set; this is an operational count, not a clinical judgment.

Step by step: building the count

You can do this in a spreadsheet from a patient list and an appointment export. No integration is required.

1. Export two lists. One: every patient with a visit in your lookback window, with their last completed visit date. Two: every appointment scheduled from today forward.
2. Set an expected return date. For each patient, add their recall interval to the last visit date. In a spreadsheet, if the last visit is in column B and the interval in months is in column C, the formula is `=EDATE(B2, C2)`.
3. Remove anyone already booked. Match the patient IDs against the future appointment list. In a spreadsheet, `=COUNTIF(Future!A:A, A2)>0` flags anyone with a booking. Drop those rows.
4. Keep only the overdue. Filter to rows where the expected return date is before today.
5. Calculate months overdue. `=DATEDIF(D2, TODAY(), "m")` gives whole months past due.
6. Band them. Group into 0-3, 3-6, 6-12 and 12-24 months overdue. Anyone beyond 24 months is usually better treated as a separate reactivation list.

Why age bands matter more than the total

A patient who is three weeks overdue often just hasn't gotten around to calling. A patient who is eighteen months overdue may have changed insurance, moved, or found another office. Both show up as "overdue" in a single total, but they are very different opportunities.

That is why you assign a different expected return rate to each band. Your own history is the best source: look at patients who were overdue a year ago and see what share eventually came back, by how late they were. If you don't have that history yet, use conservative placeholders and replace them once you've measured.

Here is an illustrative example of what bands might look like for a single-location practice. These are placeholders for the method, not benchmarks.

• 0-3 months overdue: 210 patients, assumed return rate 40%, about 84 visits
• 3-6 months overdue: 165 patients, assumed return rate 25%, about 41 visits
• 6-12 months overdue: 240 patients, assumed return rate 12%, about 29 visits
• 12-24 months overdue: 310 patients, assumed return rate 5%, about 16 visits

Putting a dollar value on it

In the illustrative example above, the raw backlog is 925 patients. Multiplying 925 by an average visit value would produce a huge, misleading number. Instead, add up the expected visits by band: 84 + 41 + 29 + 16 = 170 visits.

Now apply your average collected revenue per recall visit. Use collected, not billed. Say, for illustration, a recall visit collects $135 on average. Then 170 x $135 = $22,950 of realistically recoverable revenue sitting in the backlog today.

That figure is a one-time stock, not a monthly flow. The backlog refills every month as new patients pass their return dates. To get a monthly number, count how many patients newly become overdue in a typical month and apply the same banded math. If, illustratively, 70 patients cross their due date each month and 40% of the freshest band can be brought back with prompt outreach, that is 28 visits, or about $3,780 a month at $135.

Common mistakes that inflate or hide the backlog

• Counting inactive charts. Patients who transferred records, moved, or are deceased should be excluded. Keep your lookback window tight.
• Ignoring future bookings. If you don't remove patients who already have an appointment, you double count people the front desk already recovered.
• Using billed charges. Billed amounts overstate value. Use what the practice actually collects per visit type.
• One rate for everyone. A single return rate across all ages of overdue makes old, cold patients look as valuable as warm ones.
• Treating the stock as monthly revenue. The backlog is a pool you draw down once; the monthly opportunity is the new overdue flow plus whatever share of the pool you work through.

Turning the number into a plan

Once you have bands, the plan writes itself. The freshest band gets the fastest, lightest touch, because these patients are most likely to say yes. Older bands get fewer, spaced-out invitations. The oldest band may only be worth one well-timed message.

Most important, the work has to happen every week, not in a once-a-year cleanup. A backlog that is sized once and then left alone grows right back. The practices that keep it small are the ones where outreach runs continuously and every booked recall is tied back to the patient and the dollar value, so you can see whether the plan is actually working.