Case Study 1 — The Package That Outlived the Coverage: A Composite

Constructed. The practice, the patient, and the figures are not real. The mechanism — a global package billed months after the care began, against coverage that changed in between — is ordinary, and it is specific to obstetrics because obstetrics is the only place in this book where a single code covers nine months.

Background

Section 18.6 said the global obstetric package covers antepartum care, delivery, and postpartum care in one code, and that it is billed at the end of the episode.

Then it listed three failures that follow from that one scheduling fact. This is all three of them, on one account.

The composite

Constructed.

An obstetric practice. A patient establishes care at nine weeks and is seen on the standard schedule — monthly, then biweekly, then weekly. Thirteen antepartum visits over about seven months. An uncomplicated pregnancy, an uncomplicated vaginal delivery, and a postpartum visit six weeks later.

Textbook global package. One code. Billed the week after the postpartum visit.

It denied.

What had happened

At about six months, the patient changed jobs.

She told the practice at her next visit. The front desk noted it, took a photograph of the new card, and scanned it into the record. Nobody did anything else, because there was nothing in the workflow that said to — the practice was not billing anything, so there was no claim to correct and no balance to re-verify.

Seven months of antepartum care had been provided under one plan. The delivery and postpartum care were provided under another. And the claim, submitted in one piece at the end, went to the plan on file at the time of billing.

That plan denied the antepartum portion, correctly: it had not covered the patient during most of it.

Why the fix was hard

The package had to be unbundled, which §18.6 describes as straightforward and which is straightforward only if you have the information.

The practice needed a visit count. How many antepartum visits fell under the first plan, and how many under the second? The antepartum care codes are divided by number of visits, so the count is the code.

Nobody had counted. Nobody had needed to. The count was reconstructible from the chart — it took an afternoon — but it was reconstructible only because the visits were documented, and the dates had to be matched against the exact effective and termination dates of two plans, which the practice did not have and had to request.

And then the timely filing question. The first plan's filing limit runs from the date of service. The earliest antepartum visits were eight months old, and the claim for them had never been submitted, because the practice had been holding everything for the global package.

   ANTEPARTUM VISITS, BY PLAN

     Visits 1–8    Plan A    earliest is now ~8 months old
     Visits 9–13   Plan B
     Delivery      Plan B
     Postpartum    Plan B

   Plan B's portion: recoverable.
   Plan A's portion: depends entirely on Plan A's filing limit,
                     which nobody had checked, because nobody
                     had ever needed to file with Plan A.

(Constructed.) Some of it filed in time. Some of it did not.

Where the patient came in

She was told about it at the postpartum visit, obliquely, by someone at the front desk who said there was "an insurance issue."

She then received, over the following two months, correspondence from two insurers and statements from the practice that did not agree with each other, for care she had received while pregnant and had assumed was settled.

She had done the one thing patients are asked to do: she told the practice her insurance had changed. She told them in person, at a visit, and she watched them scan the card.

What actually failed

Not the front desk. They recorded the change correctly.

What failed is that nothing downstream consumed the information. In a practice that bills at the time of service, an insurance change triggers an immediate cascade — new eligibility check, new claim routing, sometimes a re-verification of benefits. In a practice holding a nine-month package, there is no claim in flight for the change to affect, so the change lands in the chart and stops.

The information was captured and never used. That is a different failure from not capturing it, and it is invisible to any control that measures whether insurance information is being collected.

What it shows

First, a global package is a bet that nothing will change for the duration of the package. Over nine months, in a working-age population, something changes reasonably often — employment, marriage, eligibility for a public program, a plan year. The obstetric package is the only place in this book where a coder is exposed to nine months of a patient's life at once.

Second, holding charges suspends every control that runs on claims. Eligibility checks, scrubber edits, denial work queues, and remittance review all trigger on something being submitted. A practice that has submitted nothing for eight months has run none of them for eight months — and will run all of them at once, on stale information, in a single claim.

Third, the visit count is a data requirement nobody knows they have until the package breaks. It costs nothing to track and it is the difference between an afternoon of reconstruction and a field in the system.

Fourth, timely filing is measured from the date of service, and a package does not pause it. This is worth stating plainly because it feels unfair and is nonetheless how the limits work. A practice holding a nine-month package is accumulating filing risk on every visit it holds.

And fifth — the patient did everything right. This is the second time in three chapters that the person harmed is the one who behaved correctly. Chapter 16's Case Study 1 was the first. It is worth noticing that the revenue cycle's failures land disproportionately on people who have no way to see them coming and no leverage when they arrive.

The lesson

A global package suspends billing, not risk. Everything that can change, changes, and nothing checks.

Four carry-forwards:

Re-verify eligibility during the package, on a schedule. Not at intake and not at billing. In between — at least at the start of the third trimester, and at any visit where the patient mentions a change. It is one query per patient per quarter.

Track the antepartum visit count as you go. A field, incremented at each visit. When the package holds, you never look at it. When it breaks, it is the difference between a code and an afternoon.

Know both plans' filing limits before you need them. The practice in this composite had never filed with Plan A and therefore had never read Plan A's limit.

And build a trigger on the insurance change itself, not on the claim. The change is the event. In a practice that bills at the time of service, the claim carries the signal. In a practice holding a package, nothing does — unless someone builds it.

Discussion questions

1. The front desk recorded the insurance change correctly and nothing downstream used it. Design the trigger. What fires, who receives it, and what do they do?

2. Is it reasonable that timely filing runs from each date of service when a single code covers nine months? Argue both sides, then say what a practice should do regardless of the answer.

3. The visit count costs nothing to track and is needed only when the package breaks. How often does that have to happen to justify tracking it always? Estimate honestly.

4. The patient told the practice, in person, at a visit. What would you have said to her at the postpartum visit? Write it.

5. Compare this with Chapter 16's Case Study 1, where a patient also did everything right and was harmed anyway. Is there a common structural feature? If so, name it.