Estimated insurance, and why it moves your A/R without anyone touching it

What estimated insurance is

Every practice management system carries some version of this figure, and the label varies. The concept does not. It is the software's answer to one question: of this charge, how much do we expect the plan to cover.

The calculation is mechanical. The system starts with the charge for the procedure. It applies whatever fee schedule is attached to the patient's plan, because the amount the plan considers allowable is often not the amount you charge. It applies the coverage share the plan setup records for that category of work, since preventive, basic and major treatment are usually covered at different levels. Then it adjusts for what it knows about the deductible and the remaining annual maximum. What comes out is the estimated insurance portion, and whatever is left of the charge is treated as the patient's responsibility.

The important part of that sentence is what it knows. The fee schedule and the coverage shares are inputs your office maintains, so they are as current as the last time someone updated them. The remaining annual maximum, the deductible already met elsewhere, and the frequency history are usually not things the software can see, because a patient may have been treated by another office in the same benefit year. That gives the estimate a shape worth remembering: it is at its best on routine, predictable work, and at its least reliable on exactly the large cases where being wrong costs the most and produces the worst conversation.

Why it moves without anyone touching it

A prediction updates when its inputs update. That is the whole mechanism, and it explains almost every case of an A/R total that changed while the practice was closed. Five causes account for nearly all of it, and each one leaves a different trail.

1. 1A fee schedule was updated. This is the big one. Changing a fee schedule re estimates every open claim attached to every plan that uses it, all at once, so a single afternoon of maintenance can move a practice wide total by an amount that makes people assume the software broke. Where to look: ask who last touched fee schedules, and check the date the plan's schedule was last edited before you check anything else.
2. 2A payer paid an amount other than the estimate. When the payment is posted, the estimate on that claim is replaced by what actually happened, and any difference lands in the patient column or in an adjustment. Nothing is wrong here. It is the prediction being retired by a fact. Where to look: the payments posted since the last time the total was read.
3. 3The plan on the patient changed, or a coverage table was edited. Attaching a different plan, or changing what a plan covers for a category of work, re estimates every open claim that touches it. Where to look: the patient's insurance setup, and the edit history on the plan itself if your system keeps one.
4. 4The benefit year rolled over. Maximums and deductibles reset, so estimates that were being suppressed by an exhausted maximum can come back to life on the same night for a whole group of patients at once. Where to look: the plan's benefit year start date, and whether the change lands on the first of a month or of January.
5. 5A claim was closed. Closing a claim removes its estimate entirely, and the balance it was covering moves to the party the software now considers responsible. This is the one that quietly turns an account nobody was billing into an account that is now due. Where to look: the claims closed on the day the total moved.

Whose balance is it

A balance does not belong to a treatment. It belongs to a responsible party, which is the person the practice will actually bill. In a family that is usually a guarantor rather than the person who sat in the chair, and that single fact is behind a whole class of totals nobody can reconcile.

Three consequences follow, and all three show up in real offices.

• The same family produces two different totals. A per patient view splits the balance across the people treated. A per account view holds it against the guarantor. Both are correct. They are answers to different questions.
• Summing a per patient export across a family double counts. If a balance is recorded against the guarantor and also visible on each family member's record, adding the rows together produces a total the practice cannot possibly collect. When a spreadsheet built from an export comes out larger than the software’s own total, this is the first thing to check.
• One member's open claim can hold the whole account. If the account is treated as awaiting insurance because a claim is open somewhere on it, the balances of every other person on that account can sit unbilled behind it.

This is the mechanism behind a conversation every office has had, where a staff member is certain an account owes nothing and a report insists it does. Both statements can be true, of the same family, on the same day, because they are counting from different sides.

Why it decides who gets a statement

This is where the figure stops being a reporting curiosity and becomes something a patient receives in the post. A statement routine generally holds an account out of a run when there is estimated insurance still outstanding on it, or when a claim on the account is still open. The reasoning is sound: billing a patient for money the payer has not yet declined to pay is both wrong and a phone call, and the phone call is the expensive part.

The trouble is that the rule can be built on either half of that sentence, and each half fails on its own in a different direction.

Keyed only on the estimate

An account with a claim genuinely open, but with no estimate recorded against it, looks clear and gets a statement it should never have received. The patient is billed for money the payer may well be about to send. This is the failure the patient notices.

Keyed only on the open claim

An account whose claim has been closed, but whose estimate was never cleared, is held back from every run forever. Nobody is billed, nothing chases it, and the balance ages quietly until someone happens to open the account. This is the failure the practice notices, much later.

The honest rule combines both. An account is held when there is estimated insurance outstanding or a claim still open, and it is billed when neither is true. Written that way, neither failure mode is reachable.

How to find out which rule you are actually running

Do not read the manual for this. Run your statement process the way you normally would, stop before anything is sent, and look at three accounts you already understand: one with a claim open and no estimate recorded, one with an estimate but no open claim, and one with neither. Whether each account appears in the run tells you exactly which half of the rule your process is keyed on, in about ten minutes, using your own data rather than a description of how it is supposed to behave.

Reading your A/R with it in mind

Once you know what the estimate is and how it moves, an A/R total becomes readable. Before you act on one, work through the following, in this order.

1. 1Whose balance is this? Per patient, or per responsible party. Everything downstream depends on the answer, and it is the one most reports do not state on the page.
2. 2Is estimated insurance netted out of the patient column? A total that leaves the expected insurance portion sitting in the patient column is describing money you are not going to ask a patient for.
3. 3Is the estimate current? An estimate calculated against a fee schedule or a coverage table that has since been replaced is a number about a plan that no longer exists.

A total is only comparable with last month's total when all three answers are identical in both months. This is worth being strict about, because the comparison fails in a way that looks like nothing is wrong. If a fee schedule was updated between the two readings, both numbers can be perfectly correct and the difference between them still means nothing at all. Nobody collected more or less. The prediction was rebuilt.

Which leads to the recommendation this whole subject has been building towards. Keep two figures, and keep them apart on purpose. One is the worklist figure: what has been billed to payers and is still outstanding, aged from the date of service so a resubmission does not reset the clock. That is the number the team works from every week. The other is the financial reporting figure: balances by responsible party, net of current estimated insurance. That is the number the owner and the accountant look at.

Both are real, they will never be equal, and they are not supposed to be. The failure mode is not having two figures. It is letting someone quote one of them as though it were the other.