CVS Caremark: A Prescriber Field That Identified Nobody

The CVS Caremark prescriber-identifier case ended in a judgment for Caremark over dummy identifiers in early Medicare Part D records. In its November 16, 2017 opinion, the Third Circuit held that the inaccurate identifiers did not matter to the government’s decision to pay for prescriptions patients actually received. For someone inside Part D claims processing, the case turns on a concrete question: what did the altered field change about the payment?

What a Prescriber Identifier Did in Medicare Part D

A prescriber identifier connected a prescription record to the health care professional authorized to order the medicine. In the early Part D system described by the court, that identifier traveled through a payment process involving pharmacies, pharmacy benefit managers, private plan sponsors and the Centers for Medicare & Medicaid Services.

The pharmacy sent an electronic claim to the sponsor or its processing agent. The PBM returned an approval or a rejection, generally while the patient waited at the counter. Following approval and dispensing, the sponsor reimbursed the pharmacy, less the patient’s copayment.

The sponsor also sent Prescription Drug Event records to CMS, often through its PBM. During the period at issue, those PDE records supported the year-end reconciliation between the government’s advance payments to sponsors and their actual costs. The pharmacy transaction and the sponsor’s report served different jobs in the same payment chain.

CMS’s guide to Part D data describes a PDE as a summary extract prepared by the sponsor. That distinction matters to an insider who sees an identifier change between the pharmacy’s original claim and the data ultimately sent to the government.

The early PDE format accepted a National Provider Identifier, a Universal Provider Identification Number, a state license number or a DEA number. The Prescriber ID Qualifier told the system which kind of number it was reading. Blank identifier or qualifier fields caused the CMS system to reject the record. Few prescribers used the National Provider Identifier then because no universal prescriber identifier had been issued to all U.S. prescribers.

The Default Value That Identified Nobody

Caremark’s workaround replaced missing or incorrectly formatted prescriber identifiers with a default that passed the system’s checks. According to the court’s account of the programming, employees identified approximately 4,500 PDE records in March 2006 that Caremark had authorized for payment but had not submitted to CMS because they lacked a compatible prescriber identifier.

Caremark entered AA0000000 in those records. It then programmed that value into its system so that later claims with missing or incorrectly formatted prescriber identifiers would receive the same substitute automatically. The resulting PDEs could pass the error checks that would otherwise have stopped them.

The practice expanded beyond that initial default. The opinion describes 56 different dummy identifiers in Caremark’s PDE records during 2006 and 2007. None identified the actual prescriber or corresponded to anyone with prescribing authority.

A field built to identify a prescriber was carrying a value that identified nobody. Automatically. We see the absurdity in a system that rejected an empty field but accepted an invented number.

The Audit Behind the Case

A former pharmacist who co-founded a pharmacy-auditing company identified the dummy identifiers during a 2007 audit of one of Caremark’s insurance-company clients. The whistleblower later brought a qui tam case under the False Claims Act. The government declined to join it.

The whistleblower alleged that Caremark populated the prescriber field with dummy values and then falsely certified the accuracy of the PDE records supporting reimbursement. That connected the data problem to the sponsor’s certification of payment information. The underlying prescriptions had been filled. The dispute concerned the information used to support payment for them.

Why the Judgment Turned on Payment

The Third Circuit decided the identifier issue on whether the inaccurate information mattered to the government’s payment decision. Its reasoning rested on what CMS knew about the early Part D difficulties and what the government received for its money.

CMS knew that pharmacies and PBMs were struggling to obtain usable prescriber identifiers. It wanted patients to receive their medicines, recognized the dummy numbers and continued paying the affected claims. The court emphasized that the government had paid for prescriptions actually supplied to Medicare recipients.

The government saw the fake numbers and paid anyway. The court explained why: stopping the payments would have interfered with dispensing the medicines patients needed.

We read the case through that payment chain. The computer accepted an invented identifier, but the government’s payment decision rested on the medicine being supplied. The data field and the dispensing record each tell part of the story.

The Rules Changed After These Claims

The opinion placed the workaround in the first years of Part D and described later CMS efforts to prohibit dummy identifiers. It discussed policy clarification in 2010 and revisions in 2011 and 2012.

When Inaccurate Data Drives Extra Dispensing

In a separate case against CVS Pharmacy, the government alleged that inaccurate days-of-supply data helped generate premature insulin refills. Where the prescriber-identifier case ended in a judgment for Caremark, the December 2, 2025 insulin settlement required CVS Pharmacy to pay $37.76 million. The allegations concerned excess medicine and early refills, giving the reported data a direct connection to what patients received.

The government’s complaint alleged that CVS understated how long the insulin would last, often preventing PBMs from detecting premature refills. It then described how the same inaccurate information fed the pharmacy’s refill scheduling:

According to the complaint, pharmacy staff repeatedly dispensed substantially more insulin than patients needed, substantially sooner than their prescriptions called for. The complaint also alleged that PBM audits identified premature refills, invalid days-of-supply data and quantities exceeding the prescription. The government alleged that management knew about repeated chargebacks and failed to address the problem.

That allegation reaches into the next dispensing decision. The inaccurate field allegedly helped determine when another refill occurred, creating the excess supply the government challenged.

What the Claims and Dispensing Records Show

For an insider with a Part D data concern, we connect the prescription and dispensing records with the original claim, the system’s response, the sponsor’s PDE report and the payment. An override or rebill shows what the processor did after the original claim, while the pharmacy remittance shows the resulting reimbursement.

CMS’s data guide distinguishes original, adjusted and deleted PDE records so payments can reflect revised data. Your knowledge of the processing decision can give meaning to a discrepancy that appears only as a number in an extract. We work through those connections with you to understand who changed the field, what the system did next and how the payment followed. A whistleblower case under the False Claims Act can pay 15% to 30% of what the government recovers.

You Know Why the Field Changed

If your work connects prescriber data, claim edits or PDE reporting to payments that should have stopped or changed, speak to one of our Whistleblower Experts about how those claims were paid. We work beside you to build your strongest Part D whistleblower case from your knowledge of the system and bring in healthcare fraud whistleblower lawyers who understand Part D payment records. From the first discussion through the pursuit of a recovery, we are your partners in making that knowledge count.