PBM Reconciliation: How Pharmacies Find Underpayments

All right, here is the plain answer first. PBM reconciliation is the claim by claim comparison of what a payer promised to pay at adjudication against the cash that actually landed in your bank account, and the investigation of every difference. Net-Rx describes it as matching "the promise-to-pay amount, from the third-party payer to the amount actually paid."¹

Here is the part that trips up almost every owner I talk to. The adjudication response your system prints at the counter is not a payment. It is a quote. It is the payer telling you what it intends to pay if nothing changes, and a great many things change between that moment and the deposit: a performance fee, an effective rate true up, an audit takeback, a claim that simply never gets paid at all and never gets flagged, because nobody was watching the gap. Reconciliation is how you find out which of those happened to you, on which claims, for how much.

So let me walk you through it the way I would in your back office. What reconciliation is and is not, because the word covers three different jobs. Where the money leaks, in the order I would go looking. The workflow itself. A real failure, so the stakes are not abstract. What the 2024 rule changed and what it did not. When to stop chasing and escalate. And what a reconciled pharmacy looks like on the financial statements, which is where I live.

Reconciliation is three different jobs wearing one word

When a pharmacy owner tells me the books are reconciled, I have learned to ask which kind. Bank reconciliation ties the deposits on the statement to the deposits in the ledger, which tells you nothing about whether a deposit was the right amount. Cash application matches a check or an ACH to the batch of claims it paid, which is better but still lives at the check level. Claim level reconciliation is the only one that finds underpayments, and Net-Rx is direct about the standard: "It is best practice to reconcile at the claim level to ensure you are reimbursed correctly for each claim."¹

The reason claim level is the only level that works is arithmetic. A payer sends one payment covering hundreds of claims, net of adjustments that are themselves aggregated, so a plausible looking total proves nothing. Two claims short paid by eleven dollars each, inside a batch of four hundred, will never announce themselves in a total. They announce themselves in a variance report, and only if something is generating one.

The mechanical problem is that the explanation arrives in a format built for machines. The electronic remittance advice is the ASC X12N 835 transaction, and a Medicare contractor puts the limitation plainly: "This file is not easily readable and must be converted to a readable format."²⁴ Even converted, the volume is the point. A single remittance can "potentially contain hundreds or thousands of lines of details," and every payer codes its adjustments a little differently.² That is why manual reconciliation quietly dies in most pharmacies around month two. It is a data volume problem, and it needs software plus a defined process, in that order.

Where the money actually goes missing

There are five leaks, and they behave differently, which matters because a fix for one does nothing for another. The first is the short pay: the claim pays, but for less than the adjudicated amount, with a remittance adjustment code that may be "vague or misleading," in Net-Rx's own words.² The second is the non payment: the claim never pays and no remittance ever arrives, which is the leak that hides best because there is no document to review. The two most common causes are boring and fixable, a stale physical address and stale bank account information on file with a payer.³

The third leak is the performance fee, formerly the retroactive DIR fee. The scale of that line is not a matter of opinion. CMS itself put the growth of pharmacy price concessions at 107,400 percent between 2010 and 2020,⁸ and NCPA, citing CMS, notes that "in 2020, pharmacy price concessions were 4.8 percent of Part D gross drug costs," against 0.01 percent a decade earlier.¹⁰ One vendor's measurement of the trend inside actual pharmacy sales tracks the same arc, from "less than 0.5% of total prescription sales in 2015 to 3.7% of total prescription sales in 2023."¹⁹ Earlier still, in 2021, NCPA's CEO B. Douglas Hoey called the trajectory what it was: "This is absolutely unsustainable."¹¹

The fourth leak is the effective rate true up, the generic effective rate and its brand cousin, and this one has a structural feature that catches good bookkeepers out. Effective rate contracts are not measured on your claims. As Elevate Provider Network puts it, "BERs/GERs are measured at the aggregate pharmacy services administrative organization (PSAO) level, not the individual pharmacy or claim level."²³ A law firm that litigates these describes the consequence bluntly, that the mechanism "has only brought financial harm and burden upon the pharmacies," applied both retroactively and at the point of sale.²¹ You cannot reconcile your way out of a settlement calculated on somebody else's claims. You can, however, know it is coming and accrue for it, which is a different and entirely winnable fight.

The fifth leak is the reimbursement floor itself, the maximum allowable cost list. The Supreme Court explained the mechanic better than any trade publication has: "PBMs' contracts with pharmacies typically set reimbursement rates according to a list specifying the maximum allowable cost (MAC) for each drug," and the lists are the PBM's own.¹⁵ When a MAC price sits below what you paid the wholesaler, the claim is not underpaid. It is paid exactly as contracted, at a loss. That is an appeal and a contract problem, and telling it apart from a reconciliation problem saves a great deal of wasted effort.

The workflow, in the order I would run it

None of this needs to be elaborate. It needs to be the same every week, which is a different and harder thing. Here is the sequence I would put on one page and tape inside a cabinet door.

1. Capture the promise. Pull the adjudicated amount for every claim from the dispensing system, daily, and treat that figure as the receivable. That is the number everything else gets measured against.
2. Load every remittance. Electronic where the payer offers it, and manually keyed where it does not, so coverage is complete rather than convenient. A vendor neutral service can obtain electronic files from payers that offer them and convert paper for the rest.⁴
3. Match payments to remittances to claims, in that order. Bank deposit to remittance advice first, then remittance to claim. Set aside time each week to compare the payments received to the remittance advices, and chase any missing check or deposit with the payer directly.³
4. Produce a variance list, not a total. Every claim where cash received does not equal the adjudicated amount, with the adjustment code and the dollar difference, sorted by size.
5. Classify each variance. Short pay, non payment, performance fee, effective rate adjustment, audit takeback, or your own billing error. That single classification step is what turns a report into a decision, because each category has a different owner and a different remedy.
6. Age what is left, by payer. Days sales outstanding by payer is a standard report for exactly this purpose, identifying "the average number of days between claim transaction dates and their payment dates."⁴ Payers behave consistently, so their aging patterns are predictive.
7. Work the tail on a clock. Unpaid and short paid claims expire. Reconciliation exists to catch them "before they can no longer be recovered," and the recovery window is contractual, not moral.¹

How big is the tail? The one measured figure I found from a party that publishes its own study is this: "after 60 days over 4% of a pharmacy's receivables remain unpaid."³ Hold it lightly, because the source sells reconciliation software and the study is its own, but hold it, because the arithmetic is easy to run on your own book. The same study set found 86.1 percent of claims paid within 15 to 30 days of the promise to pay, across 2,187 pharmacy identifiers over six months.² The interesting money is not in the 86 percent. It is in the remainder, and nobody stumbles into it.

The failure I would rather you hear from me

In 2018 the Ohio Auditor of State went looking for the spread on Medicaid pharmacy claims, at the request of state legislators. Across 39,378,594 prescriptions in a single year, auditors found $224.8 million in spread retained by PBMs, of which $208.4 million came from generics, a 31.4 percent share of the $662.7 million the managed care plans paid on generic drugs.¹⁶ Read that ratio twice. Nearly a third of what the state paid for generics never reached a pharmacy.

The part that should worry an owner is not the number. It is who could not see it. This was a state government with subpoena power and a statutory audit function, and the Auditor still said the work was limited by what the PBMs would hand over, calling it a "black box" in the delivery of public services. His summary of the whole review ran nine words: "The more we learn, the more troubling this becomes."¹⁶ If Ohio needed a formal audit to see its own spread, an owner working from the total on a remittance will not see anything at all. That is the case for a process rather than a hunch.

And the consequences of not seeing it are measurable in storefronts. A peer reviewed study in Health Affairs linked the national pharmacy registry to census and health statistics data and found that of the 88,930 retail pharmacies operating between 2010 and 2020, 29.4 percent had closed by 2021, with the authors reporting that "Independent pharmacies were at greater risk for closure than chain pharmacies across all neighborhood and market characteristics."¹⁷ Its first author warned that "closures may widen health disparities in access to prescription and other essential pharmacy services," with closure rates running 37.5 percent in predominantly Black neighbourhoods against 27.7 percent in predominantly white ones.¹⁸ Reconciliation will not fix a national reimbursement structure. It will tell you the truth about your own numbers in time to act on it.

What the 2024 rule changed, and what it did not

Effective January 1, 2024, CMS required Part D plans to apply all pharmacy price concessions to the negotiated price at the point of sale, and it said what it was doing in one sentence: "CMS is redefining the negotiated price as the baseline, or lowest possible, payment to a pharmacy."⁷ The reasonably determined exception that had allowed retroactive reporting was eliminated, and the change applies across every phase of the benefit.⁸ On paper, that ends the retroactive clawback. In practice it moved the money, not the problem.

Two things happened at once in early 2024. The 2023 fees came due while the 2024 fees were already being taken at the counter, which the trade nicknamed the DIR hangover and which one vendor measured at 1.91 percent of sales in January alone.¹⁹ Practitioners put it more bluntly. "Back in 2023, when we knew this was coming, it strangled pharmacy cash flow," one multi store owner and consultant told Drug Topics, adding that she knew owners who went six figures into debt over the change.²⁰ Owners who saw it coming behaved like operators, and one vendor's sales lead reported that "Most pharmacies I spoke with were building a reserve to cover the first two quarters of 2024."²⁶ CMS acknowledged the cash flow concern in the rule and said it lacked authority to mandate payment plans to smooth it.⁹

The second thing is subtler, and it is the reconciliation problem. The fee now sits inside the contracted rate formula rather than on a separate line, so the amount is knowable in aggregate and invisible per claim. Inmar's summary of the trade off is fair: the change is "creating additional complexity that is reducing the transparency needed by pharmacies to operate with financial confidence."¹⁹ So the target has shifted. You are no longer chasing a fee that shows up later. You are testing whether the rate you were paid matches the rate you contracted for. And the other side of this argument is well funded: Milliman's ten year estimate of the point of sale change says on its own report page that the work "was commissioned by the Pharmaceutical Care Management Association," the PBM trade association.²⁵ Both things can be true. The rule improved what patients pay at the counter and made your own margin arithmetic harder to see.

When to stop chasing and escalate

Not every variance is a collections problem. Some are legal rights, and knowing which is which is the difference between a productive Tuesday and a wasted quarter. I am not a lawyer and this is not legal advice, so treat the following as a map of where to point your counsel rather than a plan of action.

Start with the clock, because it is federal and specific. Under Part D, a plan sponsor must pay clean claims from network pharmacies within 14 days for an electronic claim and 30 days for any other, with a clean claim defined as "a claim that has no defect or impropriety," and interest owed on clean claims paid late.¹⁴ A sponsor also has only ten days to notify you of a deficiency in an electronic claim, after which the claim is deemed clean.¹⁴ If a payer is habitually outside that window, you are not looking at an aging problem. You are looking at a compliance one.

Then look at your state, because the ground has moved. Six categories of state law now bear on this: any willing provider, fair pharmacy audit, MAC appeal and minimum reimbursement, prompt pay, anti patient steering, and 340B parity.²² The MAC statutes are the ones that touch reconciliation most directly, because they typically require a defined appeal process with turnaround timelines and, critically, "Adjustment of similarly situated claims if an appeal succeeds."²² That last clause is where the real power sits. One won appeal can reprice a whole category. And the reason those statutes reach self funded plans at all is Rutledge, where a unanimous Court held Arkansas Act 900 was "merely a form of cost regulation" and not preempted by ERISA.¹⁵

Finally, keep the scale of the counterparty in perspective, because it explains why documentation beats indignation. The FTC found that the six largest PBMs manage nearly 95 percent of prescriptions filled in the United States, and that FTC staff wrote of contract clarity that "The rates in PBM contracts with independent pharmacies often do not clearly reflect the ultimate total payment amounts."⁵ In its second report, the FTC's policy planning director said staff "have found that the Big 3 PBMs are charging enormous markups on dozens of lifesaving drugs," alongside more than $7.3 billion in dispensing revenue above estimated acquisition cost on specialty generics from 2017 to 2022.⁶ None of that is news to a pharmacy owner. What is useful is that it is now on the record, from a federal agency, in language you can attach to a letter.

What a reconciled pharmacy looks like on the financials

This is the part I care about most, because it is where reconciliation stops being an administrative chore and becomes a financial statement. Third party receivables should be recorded at the adjudicated amount and reduced by an estimate for expected concessions and takebacks, so gross margin does not lie to you for six weeks and then correct violently. In accrual terms the concession is a reduction of revenue, not a cost of goods line, and putting it in the wrong place is how a pharmacy ends up with a gross margin percentage nobody can explain.

I ran a three year lookback once, restating a specialty pharmacy's books to GAAP after years of them being wrong, and what I remember is not the accounting. It is the owner's face when the real margin appeared. That company went on to grow from about $50 million to about $500 million in revenue, and none of that would have been financeable on books nobody trusted. Later, at a fertility pharmacy built from concept to about a $50 million operation across roughly 40 state permits, the reconciliation file was the artifact that made diligence survivable. Buyers do not pay for optimism about receivables. They pay for a receivable you can prove.

The macro picture makes the discipline urgent rather than optional. NCPA's 2025 Digest found independent pharmacy to be a $103 billion marketplace in 2024, in a year that combined a ten year high in cost of goods with a ten year low in gross profits, and with 52 percent of prescriptions covered by Part D and Medicaid combined.¹² Hoey's remedy is a policy argument, that "it is critical that policymakers change the pharmacy payment model to compensate for value."¹² Fair enough, and slow. Reconciliation is the part you control this month. GAO documented the underlying asymmetry years ago, finding that "rebates and other price concessions increased 66 percent, to $29 billion" while gross Part D expenditures rose 20 percent.¹³ When concessions grow three times faster than the spend they modify, claim level precision is not fussiness. It is survival.

Three numbers I would want on your dashboard every month, and none are hard once the process exists. Percentage of claims fully reconciled to cash. Dollars in the unreconciled tail aged past 60 days, by payer. And net revenue per prescription after concessions, trended, because that is the only figure that tells you whether the last contract amendment helped or hurt. Get those three, and deciding whether to sign a network agreement stops being a guess.