Hospital Drug Acquisition Costs: The Purchase Records Behind Medicare’s Survey

Hospital drug acquisition cost reporting connects what a hospital paid for outpatient drugs, after discounts and rebates, to the quantities and net costs it reports to Medicare. Leaving an applicable rebate out can make the reported cost higher than the hospital actually paid. For health care purchasing and finance staff, the invoice is only the beginning: wholesaler credits and buying-group rebates also belong in the picture.

The 2026 Outpatient Prospective Payment System drug acquisition cost survey covered purchases from July 1, 2024, through June 30, 2025. The Centers for Medicare & Medicaid Services surveyed hospitals paid under that system to inform proposed changes to Medicare payment for outpatient drugs. The question now is how the hospital’s purchasing records became the numbers CMS received.

What CMS Asked Hospitals to Report

The Centers for Medicare & Medicaid Services asked for total packages purchased and total net acquisition costs for each listed outer National Drug Code, in an 11-digit format. Hospitals reported those quantities and costs separately for purchases through the 340B drug discount program and purchases outside it. The split follows the purchase channel, so the discounts stay with the drugs they apply to.

The survey FAQ calls for the aggregate cost of the acquired units, rather than a per-unit price or a weighted average across purchasing accounts. It also covers outpatient acquisitions beyond drugs billed to Medicare. Purchases intended solely for inpatient use and returned purchases were excluded.

That distinction reaches the hospital’s own accounting. When inpatient and outpatient purchases share an account, the FAQ acknowledges that pharmacy purchasing, finance, billing and vendors may have to help estimate site-of-care use. A single account total does not answer how much belonged in the outpatient submission.

From the Wholesaler Invoice to Net Cost

The reported cost follows the purchase after the applicable discounts, rebates and credits. The agency’s technical report identifies group purchasing organization savings, buying-group rebates, wholesaler or distributor credits, and the discounts attached to 340B purchases. A purchasing report may identify the product and quantity while a separate credit or rebate record explains the final cost.

Product-specific concessions belonged in that product’s net acquisition cost. CMS also included discounts that a hospital could reasonably allocate to a product, such as an invoice-level discount. Additional buying-group concessions that could not reasonably be allocated to an individual drug belonged in the separate Price Concessions field. The rebate agreement and the allocation behind the reported number matter together.

For an insider, this makes the handoff between purchasing and finance consequential. We look at which concessions appeared in the purchase report, which arrived through another record, and how each reached the submission. An invoice price can look complete while leaving the later credit out of view.

The wider gap between list and net drug prices helps explain why the labels need care. In The Association Between Drug Rebates and List Prices, the USC Schaeffer Center reported in 2020 that increases in estimated rebates were associated with increases in list prices. Its rebate measure included payments and discounts to several intermediaries, including wholesalers and pharmacies. Hospital reporting asks a narrower question: which concessions changed this hospital’s acquisition cost for these purchases?

One Purchasing System, Separate Hospital Submissions

The survey required a separate response for each CMS Certification Number. The same person could submit for multiple certification numbers, but the preparation fact sheet allowed only one submitter per certification number. The hospital’s point of contact selected that person, who collected, attested to and submitted the data.

Multiple purchasing systems or wholesaler reports had to feed one complete report for the particular certification number. The FAQ expressly rejected combining different certification numbers into a single submission. Whoever assembled a systemwide purchasing file therefore had to preserve which purchases belonged to each reporting entity.

The technical report shows what happened when identical data reached more than one entity’s submission. CMS found 27 subunits reporting exactly the same drugs, quantities and net costs as their parent units. It removed the subunit records while retaining the equivalent parent records. For someone inside that workflow, the useful question is where the shared file acquired its entity labels.

A Package Quantity Is Different From a Billing Unit

The survey’s unit was the complete package identified by the outer drug code. CMS’s example uses ten cases containing twelve vials each: the reported quantity is ten package units. The package definition, purchase quantity and submitted quantity have to describe the same thing.

CMS found that some respondents used international units for hemophilia clotting factors and milligrams for selected respiratory therapies, matching the units those products are billed in. The resulting unit counts made acquisition costs appear substantially lower relative to payment rates based on average sales price. To avoid selection bias, CMS excluded all survey records for the affected drug codes from its acquisition-cost analysis.

The conversion is where purchasing and billing staff can explain something a spreadsheet total hides. We look at the package definition in the purchasing record, the unit used in the reporting file and any conversion between them. The cost can be copied correctly while the quantity changes its meaning. Same purchase, cheaper on paper.

What CMS Removed From Its Analysis

The Centers for Medicare & Medicaid Services excluded reported data with these problems:

• A purchasing-channel mismatch. Hospitals outside the discount program reported acquisitions in the 340B fields.
• A missing ceiling price. Some drugs reported as 340B purchases had no applicable ceiling price during the purchase period.
• Costs above the ceiling. Some reported per-unit costs exceeded the maximum applicable 340B ceiling price.
• Duplicated entity data. Subunit submissions repeated their parent units’ purchasing figures.
• The wrong unit. Billing quantities replaced the package quantities the survey required.

Those exclusions removed 14,313 records from the 160,430 received. CMS then separately screened extreme per-unit costs for each drug and purchasing channel. An employee who knows the original account, credit or package definition can explain which part of a reported figure changed, and why.

How the Purchase Records Reach Medicare Payment Policy

CMS used the survey to compare hospital acquisition costs with payment benchmarks based on average sales price. Average sales price is a manufacturer-reported, volume-weighted price used principally for Medicare Part B. The hospital’s acquisition cost comes from its own purchases and applicable concessions.

The July 2026 technical report found that hospitals reported acquisition costs for 340B drugs that were, in aggregate, 33.4% below mean average sales price. For drugs acquired outside that program, reported costs were approximately 2.7% above that benchmark.

The Centers for Medicare & Medicaid Services connected those findings to proposed payment policy for calendar year 2027. We care about the price behind the report because taxpayers fund Medicare payments shaped by these numbers. Leaving a rebate out makes the hospital’s purchase look more expensive in the very analysis meant to align payments with costs.

What Evidence Is Needed to Prove Fraud?

We build a suspected false reporting case with you by connecting the purchases, the submitted figures and what people inside the hospital knew to government money. The survey FAQ identifies a point of contact with authority over purchasing or finances, such as a financial executive or pharmacy manager, and a submitter responsible for attesting to the data. Report versions and communications about a known discrepancy can show how that discrepancy reached the people responsible for the submission.

A separate manufacturer pricing case shows why a concession’s route matters. In the whistleblower complaint against Regeneron, the whistleblowers alleged that Eylea discounts passed through distributor Besse Medical in a way that concealed them from the manufacturer’s Medicare price reporting. The alleged arrangement made Regeneron’s discount appear to be the distributor’s, avoiding the reduction in reported average sales price that the manufacturer’s discount would trigger.

That allegation concerns manufacturer average sales price reporting. Hospital acquisition-cost reporting follows the purchase from the buyer’s side, but the records question is familiar: who received the discount, and where did it appear in the number reported?

The Centers for Medicare & Medicaid Services said it would keep individual hospital submissions confidential while allowing aggregate data to appear in rulemaking. An insider can explain the connection between purchases and reported costs that the public analysis cannot show. A whistleblower case under the False Claims Act can pay 15% to 30% of what the government recovers.

When You Know What the Hospital Actually Paid

If you know that a hospital’s reported costs left out rebates or carried figures someone knew were wrong, speak with our whistleblower experts about the purchases behind the report. We build your strongest case from the cost records and the Medicare connection, bring in healthcare fraud whistleblower lawyers who understand drug purchasing and payment, and stay beside you through the case.