Lessons learned on bona fide service fees and fair market value-is there a 'right answer'?

“Bona Fide Service Fee (BFSF),” and “Fair Market Value (FMV),” are two of the most important and visible terms today for pharmaceutical manufacturers participating in federal healthcare programs, such as Medicaid, Medicare, Public Health Service (PHS) and the Federal Supply Schedule (FSS). For many manufacturers, up to half of their US customers receive some benefit through one of these publicly funded programs, with the government purchasing or reimbursing for drugs at significant discounts. The price the government pays is based upon data provided to government agencies by the manufacturer.

As government programs grow and the government spends more on pharmaceutical products, enforcement authorities have increased their scrutiny of the accuracy of reported pricing to ensure against overpayment. The “statutory pricing” that determines the prices paid by the government is performed by manufacturers who collect, aggregate and filter thousands or millions of lines of commercial data, including direct sales, indirect sales, and payments to third parties, such as wholesalers, group purchasing organizations (GPOs), managed care organizations (MCOs) and pharmacy benefit managers (PBMs). Currently, the math involved can be seen as a “gross to net” exercise, where the manufacturer starts with its gross direct sales and finds a net price to a certain class of customer, typically retail customers in the case of Medicaid Average Manufacturer Price (AMP).

If the manufacturer does not perform and submit accurate calculations, this could cause government overpayment, and the government could pursue a False Claims Act (FCA) action against the manufacturer. Substantial fines and penalties could result.

Therefore, manufacturers have to get the calculations correct. They must know with as much certainty as possible what is considered a discount and taken into a calculation (lowering AMP, for example), and what is considered a BFSF and excluded from the calculation (raising a price point like AMP).

The problem is this: the Centers for Medicare & Medicaid Services (CMS), the federal agency that runs the Medicaid and Medicare programs and issues the rules for how to calculate these statutory pricing statistics, has issued limited guidance for manufacturers to use when determining which payments constitute BFSFs. This is particularly true with respect to FMV, which CMS has repeatedly stated it will not define. CMS did publish criteria for BFSF determinations in the 2007 AMP Final Rule, which established what we call now the “Four Part Test.”

A BFSF is defined as:

“a fee paid by a manufacturer to an entity, which represents Fair Market Value (FMV) for a bona fide, itemized service actually performed on behalf of the manufacturer, that a manufacturer would otherwise perform (or contract for) in the absence of a service agreement, and that is not passed in whole or in part to a client or customer of an entity, whether or not the entity takes title of the drug.” (42 C.F.R.§ 414.802 (2011))

The first and most difficult element of this test to determine is the FMV. FMV is difficult because 1) it is not defined, 2) the payments are often related to a variety of products and services and are percentage based, raising the question of how to determine FMV when the payment is variable, and 3) there are variations across manufacturers and service providers in the fees paid for services.

Having performed a significant number of BFSF and FMV evaluations now, often done through a partnership between a consulting and a law firm, it seems good faith effort and due diligence are almost more important than having the exact right answer because there may not be a “right” answer. FMV is inherently subjective and can be a range. It is the documentation of the assumptions, methodology, and rationale used in establishing that range that must give an outside review organization, such as the OIG, comfort in the appropriateness of the established range.

So, why is there so much scrutiny on this area now?

The government program price points are now more visible and under a great deal of public and government scrutiny. Just as importantly, the financial impact of certain price points is inconsistent across government payers and providers.

A common approach for manufacturers used to be that if they were uncertain whether a payment should be considered a price incentive, they would be “conservative,” consider it as such, and include the payment in the AMP and Medicaid Best Price (BP) analysis. These statutory reporting requirements are used to determine the Unit Rebate Amount (URA), which is a rebate that the manufacturer pays to the state for Medicaid utilization. Then, in 2007 CMS published its Final Rule implementing the Deficit Reduction Act (DRA), which required AMP to be used, not just to define the URA, but also to determine the Federal Upper Limit (FUL) for multiple-source drugs (most, but not all of which are generics).

FULs, considered in the aggregate, set a cap on the federal share of the costs incurred by a state Medicaid program for generic prescriptions. To ensure consistent compliance with that cap, many states also use FULs on a drug-specific basis to set the maximum reimbursement available to pharmacies for multiple source drugs dispensed to a Medicaid patient. The dual role assigned to AMP effectively eliminated that potential for taking a conservative course in the face of uncertainty about requirements for calculating AMP—requirements like those associated with the definition of a BFSF—since AMPs that are too low would reduce Medicaid rebate revenues and AMPs that are too high would cost the state Medicaid programs money when they pay pharmacies for generic drugs dispensed to Medicaid patients.

AMP’s dual role caused the retail pharmacy industry to hone in on the potentially adverse financial implications of AMP-based FULs under the DRA. The industry sought and got an injunction against the use of AMPs calculated in accordance with the DRA Final Rule for reimbursement purposes, but the concept of using AMP for rebates and reimbursement was reintroduced in 2010 with the enactment of the Patient Protection and Affordable Care Act (PPACA), which also redefined AMP, further increasing the public and government scrutiny of AMP. So what we are left with now is that improper designation of a payment to a customer as a BFSF or a discount can have a dramatic impact on various price points across the government programs, including AMP, BP, the Medicare Average Sales Price (ASP), the PHS price and the VA FSS pricing. If you were to treat a specific payment as a BFSF and exclude it, the result would most often be raising the affected price points.

The effect on one price point could be favorable to the government, where the effect on another price point of the same decision could be unfavorable. For example, higher AMPs generally mean higher Medicaid rebate payments but also higher costs to Medicaid in pharmacy reimbursement. Further, although higher AMPs usually generate more rebate money for Medicaid, they typically hurt PHS entities by raising their costs for drugs. The exclusion of a BFSF (which is defined somewhat differently by the VA) from the calculations that support FSS pricing also leads to higher costs for federal agencies buying the product for their own use.

This puts the manufacturer between a rock and a hard place.

There has always been a serious lack of clear guidance surrounding the inclusion or exclusion of fees in the various calculations. In the recent court ruling, US ex rel. Streck v. Allergan (aka the Streck case), the court clearly stated that prior to the 2007 CMS Final Rule, the manufacturer had to make “good faith efforts” to evaluate the fees paid. This court finding, as well as recent reports by the OIG and the GAO, demonstrate that the government is looking closely at this problem. There now seems to be a recognition that there has been a history of a lack of guidance about how service fees should be identified and treated under government pricing programs, and that perhaps what is most important is manufacturers making a good faith effort to perform adequate due diligence around the treatment of fees, including developing and consistently applying a rationale and methodology for distinguishing between a BFSF and a discount, and then documenting the outcome of that analysis to each and every fee payment to a customer or managed care payer.

These reference points, as well as language in the 2012 AMP Proposed Rule also brought into question the treatment of GPO administration (admin) fees. The traditional approach had been to exclude any admin fees at 3% or less based on the provisions in the GPO Safe Harbor regulations. In other words, such payments became the generally recognized industry standard for FMV for certain administrative functions that GPOs were performing on behalf of manufacturers. The reports by the OIG and GAO called in to question the “pass through assumption,” part of the four-part test, suggesting that some of the payments to GPOs may be passed on to members, thus negating, at least in part, their treatment as BFSF. The AMP Proposed Rule specifically suggested that GPO admin fees should only be excluded to the extent that they meet the four-part test, which seems to indicate that manufacturers need to include such fees in their BFSF analyses even though GPOs are not usually customers that take title to drugs.

Manufacturer scrutiny related to Medicaid Program integrity will continue, and the topic of BFSFs will continue to be one of the most visible areas of scrutiny under audit. Manufacturers must conduct a high level of due diligence when making a BFSF determination. The trading partners, especially wholesalers, have a role and an interest, but the compliance accountability and risk is not on them. CMS has a role in issuing guidance, but that will take a while, and the agency will probably never define FMV. In the mean time, the message is clear: The manufacturers are expected to “get it right.”

So, what approach works?

With the combination of PPACA, the Streck case and generally increased scrutiny through government audits, and even Congress’ recent inquiries into the workings of the PHS program, many manufacturers have begun conducting thorough BFSF and FMV evaluations, as well as VA service-fee evaluations. For some, it has been the first time they have conducted such a broad and deep exercise, but the awareness of the risk is there.

CIS and Arent Fox have been on the cutting edge of many such evaluations, and have gained valuable and cumulative experiences over the past year or so. It is important that manufacturers go to an appropriate and informed external source, including law firms and consultants, to show independence and objectivity, as well as critical experience across the industry.

An important beginning is guidance and methodology. The manufacturer should be informed of the specific authoritative guidance and reference points to be used in the evaluation criteria. The next important component is methodology. We have employed a complex industry benchmarking approach, to conduct an “apples to apples comparison” across the industry, considering types and sizes of manufacturers, types of agreements, specified services, and ranges of payments. As stated earlier, this is very difficult, as the agreements, services and ranges vary greatly. It requires the appropriate collection and analysis of a wide range of industry information and data, and the parsing out of the specifics in each and every contract in order to compare like services within a range of payments.

Certainly other approaches can and should be considered, such as a “Cost-Based Approach,” particularly where specialty services unique to one or a handful of products are at issue. Multiple approaches should be evaluated and used as appropriate.

What is most important is that “good faith effort.” To us that was the term that resonated most with the Streck case, where the court discusses how a lack of regulatory guidance meant that manufacturers had to make good faith interpretations of AMP historically and how that same lack of guidance makes it difficult to suggest that a manufacturer knowingly falsified AMP if it made a good faith attempt to correctly understand the required elements of the AMP calculation. We recognize that the court’s comment applies specifically only to the historical, pre-2007 AMP, but as there is still a lack of guidance, especially with no definition of FMV, we think it is a valid consideration going forward as well.

ABOUT THE AUTHORS

Chris Cobourn is senior vice president, commercial compliance, at Compliance Implementation Services (CIS), where he works with manufacturers in the management of government programs, including audit, policy and procedures, methodology development and documentation, systems implementation, as well as class of trade and commercial systems. Chris also supports manufacturers as they work with the federal government in self-reports, restatements, and investigative activity.

Rick Moore is manager, US commercial compliance, at CIS, and oversees projects including many aspects of commercial compliance policy and practice, and government pricing programs. Rick has led many BFSF evaluation projects to determine the fair market value of service fees for purposes of AMP, BP, and ASP. Rick has handled litigation support projects for clients with alleged False Claims Act violations related to government pricing.

Larri Short, is a partner in the Health Law Group, where she focuses principally on regulatory, pricing, reimbursement, and compliance matters facing pharmaceutical manufacturers, distributors, and their customers. Her practice also extends to HIPAA privacy program developments and counseling. Larri is chair of Arent Fox’s Prescription Drug Committee.