Photo of Michelle Yeary

You’ll find any number of places on this blog where we argue emphatically that FDA warning letters are far from final and therefore should not be able to be used in products liability actions.  See here for our latest post on the issue (which also collects our other posts). These posts cite case law, FDA internal manuals, FDA’s own position in formal briefing, and learned treatises.  All of which say that an FDA warning letter has no binding legal effect.  Warning letters are informal and tentative and rarely lead to formal agency action.

But, a recent decision involving a letter issued by the Centers for Medicare and Medicaid Services (“CMS”) has us wondering if there is a chance warning letters could be viewed as final agency action after all.  The case is Ipsen Biopharmaceuticals, Inc. v. Azar, — F.3d —, 2019 WL 6482392 (D.C. Cir. Dec. 3, 2019).   At the heart of the case is the self-reporting scheme in the Medicaid drug-rebate program.  Without going into the details that aren’t important to us, the bottom line is CMS disagreed with defendant’s self-reporting and issued a series of letters advising defendant to change its data.  The letters stated that they were not “a final agency action or even an initial determination on a reimbursement claim.”  Id. at *1.  Defendant filed suit and CMS moved for summary judgement on the ground that the letters were not final agency action and therefore not appropriate for judicial review.  Id. at *2.  The D.C. Circuit disagreed.

An agency action is final if (1) “the action marks the consummation of the agency’s decision-making process” and (2) “it is an action by which rights or obligations have been determined, or from which legal consequences will flow.”  Id. (citing 5. U.S.C. § 704).  And, the provision open to debate is “from which legal consequences will flow” – which the Supreme Court has called a “pragmatic” inquiry.  Id.  That means the court has to look at the consequences of the agency’s action in the context of the specific regulations at issue.

Here, the defendant argued that the Medicaid statute provides penalties for “knowingly provid[ing] false information.”  Id. at *3.  Therefore, the CMS letters increased the probability that defendant could be found to have knowingly provided false information.  The court agreed that because the letters carried an “increased risk of prosecution and penalties,” it was a “legal consequence” and thus met the definition of a final agency action.

CMS also argued that the risk that defendant may incur penalties in some future action was too attenuated and uncertain to render it a legal consequence.  Apparently, the court found that since the letter could some evidence against defendant — that was enough.  The court also found that because defendant had a self-reporting obligation, each time defendant submitted the same data to CMS after receiving the letters, defendant was exposed to civil penalties.  For this the letter itself was the trigger.  Id. at *4.

The court also was persuaded by the fact that there is no further action defendant can take at the administrative level.  In this way, the letters constitute the agency’s final decision which includes informing defendant that it must change its rebate calculations.  Id. Finally, the court noted that agencies often communicate decisions and interpretations via letter without formalizing or finalizing their action.  Id. at *5.  Here the court found that because the letter “expressly applied CMS’s interpretation of governing law to the specific facts of [defendant’s] case . . . [it] closely resembles an individual adjudication, which is a well-recognized form of final agency action.”  Id.

Put that all together and you have a decision by the D.C. Circuit that informal agency letters asserting specific regulatory violations sufficiently increase the risk of enforcement action so as to be final and appealable.  Unfortunately, that sounds like an FDA warning letter.

Think about an FDA warning letter that concludes that a specific sales aid is false and misleading because it omits risk information or includes unsubstantiated claims.  We’ve all seen them.  The letter will state that the specific sales aid misbrands the product under the FDCA and makes its distribution violative of FDA regulations.  The letter will usually contain at least some examples of what the FDA finds misleading or false.  It will conclude with a request that the manufacturer cease using the sales aid (or cease misbranding the product) and take corrective action.  The letter requires a response by the manufacturer in a given timeframe.

It seems to tick a lot of the same boxes the court found convincing in Ipsen.  Warning letters do more than merely interpret a regulation, they apply those interpretations to the specific facts and reach conclusions regarding whether those facts violate the FDCA.  FDCA violations carry the risk of both civil and criminal penalties.  If a manufacturer continues to use sales materials that violate the FDCA as found in the warning letter, manufacturer is open to additional penalties.  FDA uses warning letters as a routine method of conveying violations to manufacturers.

We surely aren’t persuaded that the court reached the correct conclusion in Ipsen and if confronted with Ipsen in a pharmaceutical warning letter context, we’d be relying on all those sources we mentioned at the outset of this post and diving deeper to find differences between the FDA scheme and the CMS scheme.  For now, it’s enough to know it’s out there and some more work may need to be done.