On December 20, 2019, the Federal appeals court panel that heard U.S. ex rel. Bookwalter v. UPMC, No. 18-1693 (3d Cir.), amended its September 2019 opinion by removing a controversial interpretation of the “volume or value” standard under the Stark Law. The September opinion had adopted a “correlation theory,” holding that a physician’s compensation “varies with” the volume or value of referrals if the physician is paid based on his personally performed services, such as on a work relative value unit (wRVU) basis, and there is a “correlation” between the physician’s referrals and those personally performed services. The court relied on this correlation theory to support its finding that the physicians had an indirect compensation arrangement with the hospitals to which they referred, thereby allowing the case to proceed and shifting the burden to the defendants to prove the availability of a Stark Law exception. Although the amended December opinion removed the correlation theory rationale, the court maintained its September holding to allow the case to proceed based on alternative reasoning that there were adequate allegations that the physicians’ compensation “took into account” their referrals.
The Stark Law prohibits a physician’s Medicare referrals for “designated health services,” including hospital services, to an entity with which the physician has a direct or indirect financial relationship, unless the requirements of an applicable exception are satisfied. One element of the Stark Law’s test to determine whether a physician has an indirect compensation arrangement with an entity is whether the physician’s aggregate compensation “varies with, or takes into account, the volume or value of referrals” to the entity. For these reasons, a critical component in a Stark Law analysis is frequently whether a referring physician is compensated in a manner that “varies with” or “takes into account” the volume or value of his referrals.
The relators in Bookwalter allege in their False Claims Act qui tam case that employed neurosurgeons had an indirect compensation arrangement with University of Pittsburgh Medical Center (UPMC) hospitals that resulted in a prohibition of their referrals under the Stark Law. The neurosurgeons were employed by UPMC subsidiaries and referred to hospitals owned by other UPMC subsidiaries. Each neurosurgeon’s compensation involved a base salary and a productivity bonus. Each neurosurgeon’s base salary was subject to prospective reduction if the neurosurgeon did not achieve a certain wRVU target for personally performed services, and the productivity bonus was calculated from the number of the neurosurgeon’s personally performed wRVUs in excess of the target. The relators claim that the compensation for some neurosurgeons exceeded their employer’s collections for their services, some neurosurgeons received compensation in excess of the 90th percentile, many generated very high wRVUs, and the per wRVU productivity bonus rate exceeded the employers’ reimbursement rates.
The District Court dismissed the case for failure to state a claim, prompting the relators’ appeal to the Third Circuit. On appeal, a central issue was whether the employed neurosurgeons had a financial relationship with the hospitals under the Stark Law’s definition of indirect compensation arrangement. If not, the Stark Law’s referral prohibition would not be triggered and the case would be dismissed – whether or not a Stark Law exception was satisfied. As noted above, an indirect compensation arrangement exists if, among other things, a physician’s aggregate compensation “varies with, or takes into account, the volume or value of referrals.”
The September and December Opinions
In its September opinion, the court separately discussed the “varies with” and “takes into account” elements of the indirect compensation arrangement definition, ruling that an indirect compensation arrangement can exist if either of these two elements exist. It concluded that a physician’s compensation would “vary with” referrals if there was a correlation between compensation and referrals – that is, “[i]f compensation tends to rise and fall as the volume or value of referrals rises and falls.” According to the court, since the neurosurgeons referred to a UPMC hospital every time they personally performed a procedure that generated wRVUs, structurally, the physicians’ compensation varied with their referrals. Addressing the other element of the definition, the court concluded that a physician’s compensation “takes into account” referrals if there is a “causal relationship” between the two. Here, the court pointed to the “suspiciously high compensation” paid to the neurosurgeons in support of its holding that the neurosurgeons’ compensation plausibly took into account their referrals.
In its December amended opinion, the court eliminated its interpretation that “varies with” means “correlation.” Instead, the court simply said that it need not resolve the meaning of “varies with” because it has concluded that the neurosurgeons’ compensation “took into account” their referrals. On that basis, it reversed the dismissal of the relators’ complaint.
Implications of Bookwalter
Physician compensation structures that include incentives to reward physicians for personally performed services are ubiquitous in the health care industry. And it is not unusual for physicians to be employed or contracted to provide these services by affiliates of hospitals or other entities to which those physicians refer. Oftentimes, physicians personally perform professional services in connection with many of their referrals, such as where a surgeon performs a surgical procedure at a hospital. Knowing how to analyze such compensation structures to maintain compliance with the Stark Law – the topic at issue in Bookwalter – is of critical importance to the industry.
The court’s December amendment of its September opinion is some good news for those providers that employ or contract with physicians through affiliated entities. It annuls the portion of the Third Circuit’s September opinion, which was precedential, that had adopted the “correlation theory.” However, it is important to recognize that the court did not reject that theory either, thereby preserving the ability of relators in other cases to argue, and courts to conclude, that the Stark Law may be violated when a physician is paid based on personally performed services but there is some “correlation” with his referrals.
Given the Bookwalter court’s December turnabout and the uncertainty that remains concerning the volume or value standard, clear guidance from the Centers for Medicare & Medicaid Services (CMS) is needed now more than ever. As noted, a critical factor in the indirect compensation arrangements definition is whether a physician’s aggregate compensation varies with, or takes into account, the volume of value of his referrals. Historically, for purposes the indirect compensation arrangements definition, CMS has focused on the word “aggregate.” In contrast, many Stark Law exceptions require that a physician’s compensation cannot take into account the volume or value of his referrals (omitting the word “aggregate”). Unlike CMS, the Bookwalter court did not seem to give any import to the word “aggregate” in interpreting the indirect compensation arrangements definition, indicating that the Third Circuit’s analysis is not in line with CMS’s approach to the volume or value test.
Moreover, in its October 2019 proposed Stark Law regulatory changes, without any discussion, CMS proposed to change the indirect compensation arrangements definition to eliminate the “varies with” element. As proposed, a physician would have an indirect compensation arrangement with another entity only if, among other things, the physician’s compensation “takes into account” the volume or value of referrals. While it is hard to be certain due to the lack of explanation, this change further suggests that CMS may not interpret the volume or value test in the indirect compensation arrangements exception in the same manner did as the Bookwalter court.
Also in its October proposal, CMS offered a new test, focused mainly on whether referrals are an explicit variable in the physician’s compensation formula, to determine whether compensation “takes into account” referrals. In noteworthy discussion, CMS reaffirmed that a physician’s productivity bonus will not take into account the volume or value of the physician’s referrals solely because the referral recipient bills for a service that corresponds to such personally performed services. While this guidance certainly applies to the volume or value standard in many Stark Law exceptions, it is less clear whether that same analysis applies to the volume or value standard in the indirect compensation arrangements definition. Here again, the health care industry would benefit greatly from a full and clear explanation from CMS when it finalizes the Stark Law changes.
Finally, the practical implications of the indirect compensation arrangements definition are worth noting. As Bookwalter shows, if the relators had not been able to make a plausible case that the neurosurgeons had an indirect compensation arrangement with the hospitals to which they referred, the case would have been dismissed. Because, according to the court, they were able to make this case, the burden shifts to the defendants to prove the applicability of a Stark Law exception. As a result, the defendants will be subject to significant litigation costs and risks as the case moves forward, giving the relators substantial leverage, even if the defendants believe that they could ultimately demonstrate the applicability of a Stark Law exception. Given the costs and risks associated with defending a qui tam case based on meeting a Stark Law exception, the need for clear guidance that enables providers to structure their physician compensation arrangements to avoid having to defend such allegations is tremendous.