In U.S. ex rel. J. William Bookwalter, III, M.D. et al. v. UPMC et al., the US Court of Appeals for the Third Circuit endorsed two controversial interpretations of the Stark Law’s “volume or value” standard, known as the correlation theory and the practice “loss” theory. Specifically, the court held that the relators had made out a plausible allegation of an indirect compensation arrangement between surgeons and University of Pittsburgh Medical Center (UPMC)-affiliated hospitals. The court held that the relators were entitled to proceed to discovery because of the correlation between the amount of the productivity-based compensation paid to the surgeons and the volume of the surgeons’ referrals for inpatient hospital services (e.g., operating room and hospital room and board). Repeatedly invoking the concept of “where there is smoke, there might be fire,” the court also stated that the fact that at least three of the surgeons allegedly received compensation in excess of the hospital’s collections for their professional services supported the plausibility of the relators’ allegation that the compensation “takes into account” the volume or value of the physicians’ referrals to the hospitals.

If this holding sounds familiar, that is because it is based on the same logic advanced by the Fourth Circuit in U.S. ex rel. Drakeford v. Tuomey, the infamous Stark Law/False Claims Act (FCA) case that first put the hospital industry on notice that common productivity-based compensation to hospital-employed surgeons could implicate the Stark Law. While distinguishable from Tuomey, UPMC has important implications for hospitals and health systems that employ surgeons.

Summary of Allegations and Procedural History

In UPMC, the plaintiffs alleged that the UPMC hospitals where the neurosurgeons performed cases each had an indirect compensation arrangement with the surgeons and thus triggered the Stark Law’s prohibitions against referrals and the associated Medicare claims for reimbursement. Based on this alleged Stark Law violation, the plaintiffs claimed that the hospitals violated the FCA by submitting false claims for hospital services referred by the surgeons. The surgeons were paid a base salary and a productivity bonus of $45 per work RVU above a specified target. If a surgeon did not hit the target, her base compensation would be reduced the following year. The government had intervened in and settled another aspect of the case, but declined to intervene on these allegations.

The compensation arrangement between the surgeons and the UPMC hospitals was evaluated as a potential indirect compensation arrangement because the surgeons were employed by UPMC-affiliated medical practices, not directly by the UPMC hospitals. For Stark Law purposes, an indirect compensation arrangement requires, among other things, that the compensation paid to the physician “varies with” or “takes into account” the volume or value of the physician’s referrals to the hospital. In this case, the plaintiffs alleged that the compensation greatly exceeded fair market value and that at least three surgeons were paid more than the hospital collected for their services. The plaintiffs also asserted that “[e]very time . . . [the surgeons] performed a surgery or other procedure at the UPMC Hospitals, the Physicians made a referral for the associated hospital claims pursuant to the Stark Statute.”

UPMC filed a motion to dismiss plaintiffs’ second amended complaint, and the US District Court for the Western District of Pennsylvania granted the motion with prejudice, agreeing with UPMC that use of a common productivity-based compensation methodology to pay surgeons does not support a plausible allegation that the compensation “varies with” or “takes into account” the volume or value of the surgeons’ referrals to UPMC hospitals. The district court was also unpersuaded that the compensation to the neurosurgeons “takes into account” the volume or value of the surgeons’ referrals to UPMC hospitals because the compensation is allegedly in excess of fair market value. In support of their allegation that the compensation exceed fair market value, the plaintiffs alleged:

• The physicians were paid compensation exceeding collections from their work.
• The physicians fraudulently inflated their work RVUs to extremely high levels (resulting in compensation for work not performed).
• The physicians were paid compensation above the 90th percentile.
• The physicians were paid compensation per work RVU in excess of Medicare rates.
• Plaintiffs appealed.

On appeal, the Third Circuit reversed the district court, holding that the plaintiffs’ complaint satisfied the basic pleading requirement of “plausibility” and the heightened pleading requirement for Rule 9(b) particularity applicable to all FCA actions. Accordingly, the plaintiffs can now proceed with discovery.

Key Implications for Hospitals and Health Systems

Productivity-Based Compensation Treated with Suspicion

A key implication of the Third Circuit’s opinion in UPMC for hospitals and health systems is that the mere allegation that a surgeon is paid productivity-based compensation is enough to survive a motion to dismiss a Stark-based FCA complaint in the Third Circuit, because the amount of productivity-based compensation paid to a surgeon employed by an affiliated medical practice arguably correlates with the volume of the surgeon’s referrals for inpatient hospital services. Hospitals and health systems employing surgeons in the states under the Fourth Circuit’s jurisdiction already lived with this legal reality. Now that the Third Circuit has in significant part adopted the Fourth Circuit’s construction of the “volume or value” prong of the Stark Law’s “indirect compensation arrangement” definition, there is some concern that this decision is a harbinger for other circuits. Ultimately, no hospital or health system employing surgeons and paying them productivity-based compensation can be certain that they are safe from a frivolous suit, despite the fact that the Centers for Medicare and Medicaid Services (CMS) has expressly stated, on multiple occasions, that productivity-based compensation is permitted under the Stark Law.

Circuit Judge Ambro disagreed with the majority’s construction of the “volume or value” prong of the “indirect compensation arrangement” definition in a concurring opinion, arguing that the correlation between productivity-based compensation to surgeons and referrals to inpatient hospitals is insufficient to make out an allegation that the compensation “varies with” the surgeon’s referrals for inpatient hospital services. Judge Ambro argued that “varies with” requires a causal relationship between the surgeon’s compensation and referrals for inpatient hospital services, a relationship that is missing when productivity is measured solely on the basis of the physician’s personally performed work. The majority disagreed, concluding that only the “takes into account” language of the Stark Law (i.e., “takes into account the volume or value of referrals”) requires a showing of a causal relationship.

Judge Ambro also noted that the majority’s construction “suggests . . . that any hospital that pays affiliated physicians according to some metric of the work they personally perform at the hospital falls under suspicion of violating the Stark Act, and it can only restore its good name by pleading one of the . . . exceptions—presumably at the summary judgment stage at the earliest, i.e., after discovery has taken place.” The majority opined that this “opening the floodgates of litigation” concern is overstated, however, because the government can dismiss frivolous qui tam actions. This is true, but affords the industry cold comfort in light of the fact that the government has exercised this authority in relatively few cases. Given the hundreds of new cases filed every year, the US Department of Justice (DOJ) cannot reasonably be expected to affirmatively move to dismiss every case the government believes is not meritorious. Moreover, the purpose of Rule 9(b) is for the courts to separate the wheat from the chaff and not rely on the government to take the first pass.

Practice “Loss” Theory

While not central to the court’s holding that the plaintiffs met their pleading burden, the Third Circuit also expressly endorsed the practice “loss” theory. This theory postulates that where a physician costs a hospital more than the hospital can collect for his services (thus operating at a “loss” on the physician’s services), the physician’s compensation might be above fair market value and take into account the volume or value of the physician’s referrals to the hospital. The underlying premise is that, absent special circumstances, the hospital would not operate at a loss on the physician’s services were it not “taking into account” the volume or value of the physician’s referrals for hospital or other Stark designated health services.

This theory has been advanced by relators and the DOJ in other Stark Law cases. Nonetheless, it has been the subject of significant criticism because it fails to recognize the financial realities of hospital operations, such as minimum staffing requirements for trauma centers and reimbursement rates (particularly where hospitals serve a more challenging payor mix). It also disregards the fact that many hospitals are nonprofit, charitable enterprises that serve a mission. Those hospitals may need to maintain a service line that is not profitable to ensure a community has access to necessary services. Even in the absence of such special circumstances, the practice “loss” theory is based on the fallacy that physicians in private practice are not paid more than collections from their personal productivity will support, when, in fact, the local market rates of compensation to physicians reflects income from ancillary services, such as in-office lab and other diagnostic testing. In fact, physician salary surveys, which the government agrees can be used to set fair market value physician compensation, are based on private practice physicians whose salary is not limited to their collections from personal productivity. Moreover, the practice “loss” theory renders the “volume or value” standard a subjective intent test, i.e., why is the health system paying this level of compensation, rather than an objective “bright line” standard that CMS claims to have intended. Finally, the theory relies on the fallacy that the hospitals and health systems employing the physicians are in the business of physician services; in today’s health care market where bundled or global payments, shared savings and other payment methodologies have upended the historical distinction between the physician and non-physician/facility components of care and payment, it is arguably misplaced to judge the reasonableness of a physician’s level of compensation based solely on collections from the physician component of the care.

Given the problems with the practice “loss” theory of “volume or value,” arguably it should not meet the basic pleading requirement of “plausibility.” However, in this case, the Third Circuit found that the theory, along with other allegations raising “suspicions” regarding the fair market value of the compensation, made out a plausible allegation that the compensation may “take into account” the volume or value of the physicians’ referrals and thus create an indirect compensation arrangement between the surgeons and the UPMC hospitals.

Low Pleading Standard

A third implication of the UPMC majority opinion is that, at least in the Third Circuit, plaintiffs need not plead facts showing that the defendant knew or should have known that the physician compensation arrangement failed to satisfy a Stark Law exception to survive a motion to dismiss. Following its own precedent, the Third Circuit held that compliance with a Stark Law exception must be plead by the defendant as an affirmative defense.

The hospital argued that a defendant in an FCA case can only have the requisite knowledge (i.e., actual knowledge, reckless disregard or a deliberate ignorance) that the claim is “false” if the defendant knew that the compensation arrangement did not satisfy a Stark Law exception. In other words, knowledge of falsity for purposes of the FCA must entail knowledge that the compensation arrangement did not satisfy a Stark Law exception, because only then can the defendant have the requisite knowledge that the claims were prohibited by the Stark Law and, therefore, false. The court acknowledged that the hospital’s argument “has force,” but nonetheless rejected it, largely on the basis of Third Circuit precedent. Accordingly, the court held that “[i]t is thus the defendants’ burden to plead a Stark Act exception, not the relators’ burden to plead that none exists.”

This aspect of the court’s holding is open to criticism, although its characterization of the exceptions as affirmative defenses is one that has been embraced in other circuits. The Stark Law’s fundamental prohibition on certain referrals and claims only applies if an exception does not apply. The first five words of the Stark Law are “(e)xcept as provided in subsection (b),” with subsection (b) setting forth the Stark DHS exceptions. Further, while the court recognizes that the plaintiff must plead a plausible allegation that a “financial relationship” exists between the physicians and the hospitals, and the court recognizes that there is nothing inherently wrong with a physician and hospital having a financial relationship, it does not require the plaintiffs to plead that none of the compensation exceptions apply and the defendants knew or should have known that no compensation exception applied. Pleading that a hospital has a compensation arrangement with a referring physician (without regard to compensation exceptions), the physician made referrals to the hospital, and claims were submitted for those referrals, arguably does not plead a plausible Stark Law violation. If like homicide or other inherently wrongful conduct it was inherently wrongful for a physician who has a compensation arrangement with a hospital (before exceptions are considered) to make referrals to the hospital, it is logical to treat the exceptions as affirmative defenses that must be plead by the defendant. In this case, however, the court even acknowledged that there is nothing inherently wrong with the referrals and financial relationship, only that there is the potential for abuse.

Thus, in an FCA claim predicated on an alleged Stark law violation, there is a good argument that plaintiffs should bear the burden of pleading on the issue of whether the defendant knew or should have known that the compensation arrangement or referrals did not qualify for a Stark Law exception. Indeed, the crucial state of mind question under the FCA is whether a defendant knew or recklessly disregarded whether it was submitting claims in violation of law, thereby committing fraud. It is an FCA plaintiff’s burden to plead this element of her claim. Accordingly, in a Stark-based FCA case, it would make eminent sense to require a plaintiff to plead the absence of an applicable exception.

What Happens Next?

The defendants have moved for an extension to file a petition for rehearing en banc. Their petition is due on October 15, 2019.

The court here considered the correlation theory and the practice “loss” theory in the context of a motion to dismiss, not summary judgment. Notably, the Third Circuit’s correlation theory is based on the “varies with” language of the indirect compensation definition. The court still required a causal relationship between a physician’s referrals and compensation for purposes of the Stark Law’s “takes into account” prong of most applicable Stark compensation exceptions, including the indirect compensation exception and the special rule on unit-based compensation that will no doubt be raised as affirmative defenses. Thus, if the plaintiffs proceed to discovery, it remains to be seen how they will fare in the post-discovery context where the “smoke” of “suspicion” regarding compliance with the “fair market value” and “takes into account” standards will be scrutinized for “fire.”

In response to the Third Circuit’s decision, hospitals and health systems should thoroughly document why their physician compensation arrangements comply with Stark Law exceptions, including documentation of internal and external valuation and Stark Law analyses, especially for compensation arrangements with physicians with high referral volume or value. There is speculation that proposed amendments to the Stark Law regulations, which are currently in review with the Office of Management and Budget, will address the Tuomey correlation theory of the “volume or value” prong, and possibly the practice “loss” theory. Consequently, hospitals and health systems may want to wait for the proposed amendments before making changes to surgeon and proceduralist compensation structures to avoid the correlation theory risk (now increased by the UPMC decision).