By Richard Zhao*
The Stark law is a federal anti-fraud statute that prohibits physicians from referring Medicare and Medicaid patients for designated health services to an entity in which the physician has a financial interest.[i] Named after Congressman Pete Stark who sponsored the original bill, the law was enacted at a time when healthcare services were reimbursed separately on a fee-for-service basis.[ii] This model incentivizes physicians to either perform more procedures themselves or refer patients to other providers who offer financial compensation in return for such referrals. This leads to increasing cost for the patient and an overutilization of healthcare resources.[iii] The Stark law is in place to curb this medically unnecessary referrals and resource overuse.
Since the enactment of the law, healthcare in the United States has been shifting from a fee-for-service, volume-based scheme to what is now known as a value-based model.[iv] In a value-based model, payment is not calculated by the amount of services offered but instead by the quality of such services. Improvement of patient health outcome thus takes the center stage.[v] A major avenue healthcare providers have used to realize outcome improvement is through the formation of value-based networks of providers to achieve better care coordination and integration.[vi]
The current version of the Stark law and the corresponding regulations introduced by the Centers for Medicare and Medicaid Services (CMS) do not adequately accommodate this landscape shift. The core of many of these value-based networks is to establish meaningful partnerships that necessarily entail some type of financial incentive arrangement among participating providers. The current Stark law regulations effectively foreclose this possibility in many instances.[vii] While exceptions applicable to certain narrow categories of value-based networks exist, they are nonetheless incompatible with the extent of value-based payment innovation.[viii] Recognizing the need for reform, in October 2019, CMS published a proposed rule to modernize its interpretations of the Stark law in an increasingly value-based healthcare ecosystem.[ix] Specifically, the proposed rule defined six value-based concepts and carved out three value-based exceptions.[x]
Under the Stark law, the Secretary of the Department of Health and Human Services has wide latitude to design exceptions that do not pose a substantial risk of program and patient abuse.[xi] The question left open is whether these proposed changes comport with the objectives of the Stark law—namely, to deter program and patient abuse. The current version of the proposal expressly encompasses two layers of safeguards to ensure that Stark law’s anti-fraud objectives remain intact. The first layer is the requirements embedded in the six proposed value-based definitions: value-based activity, value-based arrangement, value-based enterprise (VBE), value-based purpose, VBE participants, and target patient population.[xii] The idea is that these definitions will function as a preliminary filter to screen out schemes that are still based on the traditional fee-for-service rationale. Arrangements that pass this threshold then get tested for compatibility with the three proposed exceptions: arrangements that assume full financial risk, arrangements with meaningful downside financial risk to the physician, and value-based arrangements in general.[xiii] The implication seems to be that models that fall within one of the proposed exceptions are less likely to be susceptible to program and patient abuse.
The underlying mechanism at work here is the risk-sharing nature of a mature value-based network. In a fee-for-service world, the payor, public or private, assumes full financial risk. The more services are rendered by providers, the greater financial risk payors must assume.[xiv] This lopsided risk-taking arrangement necessitates the enactment of anti-fraud statutes like the Stark law, which stipulates that physicians “may not make a referral” to an entity with which the physician “has a financial relationship”.[xv] In contrast, in a value-based system, financial risk is spread across the various stakeholders. Providers are incentivized to cut back on unnecessary services because they will have to assume at least part of the cost spillover, and where savings are rendered, they are entitled to a share of that surplus.[xvi] The drafters of the proposed rule view this mechanism embedded in value-based arrangements as a powerful check on the behavior of providers, which, in tandem with the requirements incorporated through the definitions and exceptions, serves the objectives of the Stark law.
Are the existing built-in deterrents sufficiently adequate to ward off program and patient abuse? For the first two exceptions, one could argue that the risk-sharing mechanisms required to meet the standard of the exceptions serve as a strong enough disincentive for providers to engage in fraudulent or abusive behaviors, but with the third exception, this inquiry is particularly pertinent. In its current form, the third exception will apply to all value-based arrangements as defined by the proposal, regardless of the level of financial risk that the participating providers undertake.[xvii] Recognizing the potential fallibility of this exception, CMS proposed a requirement unique to this exception that remuneration to physicians in this type of arrangements “not conditioned on the volume or value of referrals.”[xviii] This approach is reminiscent of language in the existing exceptions that remuneration shall not be determined in any manner that “takes into account volume or value of referrals.”[xix]
Historically, this latter standard has been interpreted in various ways. Under one approach, remuneration would take into account the volume or value of referrals only if “it varied proportionally with increases or decreases in such referrals, either during the term of the compensation arrangement or from one compensation arrangement to the next.”[xx] Under an alternative approach, the mere existence of a physician remuneration scheme implicates violation of this standard.[xxi] A third view is that the better interpretation of the standard should focus on “the formula the parties use to determine compensation over the term of the arrangement, not on whether referrals or other business generated were considered in structuring the arrangement or the intent of the parties.”[xxii] CMS appeared to be in agreement with this last approach. The proposal provided that this standard is breached “only when the mathematical formula used to calculate the amount of the compensation includes as a variable referrals, and the amount of the compensation correlates with the number or value of the physician’s referrals.”[xxiii]
None of the foregoing interpretations make total sense in the context of value-based healthcare, though they are consistent with the anti-abuse objectives of the statute. As one commentator suggested, in certain value-based arrangements, remuneration is set to be inversely correlated with the volume or value of patient referral.[xxiv] Such programs would be barred by the overly restrictive requirement that compensation schemes not in any way take into account referral volume or value. In contrast, the modified requirement does accommodate schemes with an inverse correlation between physician remuneration and referral volume and value because compensation in these structures will be conditioned not on the occurrence of referral but rather the absence of it. The modified language also more precisely addresses the concern that physicians base their referral decisions not on medical necessity but on knowledge that such referrals will generate financial gains. Therefore, this proposed safeguard to the third value-based exception strikes the right balance between accommodation of innovation and prohibition of abusive referral behavior. In conclusion, while the proposed exceptions seem to significantly loosen the Stark restrictions on financial alignment between providers, they are accompanied by appropriate safeguards that disincentivize value-based participants from engaging in fraudulent and abusive conduct. With proper auditing measures and a robust program exit mechanism, these proposed changes will likely go a long way towards providing a regulatory boost to the development of value-based care.
*Richard Zhao is a Junior Editor on MJEAL. They can be reached via email at email@example.com.
[i] 42 U.S.C. § 1395nn(a)(1) (2018) (“Except as provided in subsection (b), if a physician (or an immediate family member of such physician) has a financial relationship with an entity specific in paragraph (2), then the physician may not make a referral to the entity for the furnishing of designated health services for which payment otherwise may be made under this subchapter”).
[ii] Billy Wynne et al., Proposed Stark Law, Anti-Kickback Reforms Aim to Facilitate Value-Based Care, Health Affairs (Nov. 27, 2019, 9:52 AM), https://www.healthaffairs.org/do/10.1377/hblog20191014.13154/full.
[iii] Carmel Shachar, Are Fraud and Abuse Laws Stifling Value-Based Care?, New England J. Med Catalyst (Nov. 27, 2019, 10:01 AM), https://catalyst.nejm.org/fraud-and-abuse-laws-stifling-value-based-care.
[iv] Modernizing and Clarifying the Physician Self-Referral Regulations, 84 Fed. Reg. 55,766, 55,768 (Oct. 17, 2019) (to be codified at 42 C.F.R. pt. 411).
[v] What Is Value-Based Healthcare?, New England J. Med. Catalyst (Nov. 27, 2019, 10:29 AM), https://catalyst.nejm.org/what-is-value-based-healthcare (The “value” in value-based healthcare is derived from measuring health outcomes against the cost of delivering the outcomes).
[vi] What Is Care Coordination? New England J. Med. Catalyst (Nov. 28, 2019, 9:44 AM), https://catalyst.nejm.org/what-is-care-coordination (“The goals of coordinated care are to improve health outcomes by ensuring that care from disparate providers is not delivered in silos, and to help reduce health care costs by eliminating redundant tests and procedures”).
[vii] See supra note 3 (arguing that the current fraud and abuse laws “have a chilling effect on innovative health care financing and delivery models” because the current exceptions are “narrow and hard to comply with”).
[ix] Press Release, Department of Health and Human Services, HHS Proposes Stark Law and Anti-Kickback Statute Reforms to Support Value-Based and Coordinated Care (Oct. 9, 2019) (on file with the Department of Health and Human Services).
[x] See supra note 4, at 55,773-83.
[xi] Supra note 1, § 1395nn(b)(4).
[xii] Supra note 4, at 55,773.
[xiii] Supra note 4, at 55,779-83.
[xiv] Marsha Gold, Financial Incentives: Current Realities and Challenges for Physicians, 14 J. Gen. Internal. Med. S6 (1999).
[xv] Supra note 1.
[xvi] Andrew Meola, How And Why Value Based Purchasing Is Trending In the Healthcare Industry, Business Insider (Nov. 28, 2019, 11:19 AM), https://www.businessinsider.com/value-based-care-pay-for-performance-healthcare-model.
[xvii] Supra note 4, at 55,783 (“We are proposing an exception at §411.357(aa)(3) for compensation arrangements that qualify as value-based arrangements, regardless of the level of risk undertaken by the value-based enterprise or any of its VBE participants”).
[xviii] Supra note 4, at 55,783 (“Because the exception proposed at §411.357(aa)(3) would be applicable even to value-based arrangements where neither party, but especially not the physician, has undertaken any downside financial risk, we believe that safeguards beyond those included in the promised meaningful downside financial risk exception are necessary to protect against program or patient abuse. Specifically, we are proposing. . . .a requirement that remuneration is not conditioned on the volume or value of referrals of any patients to the entity. . . .”).
[xix] Supra note 1, § 1395nn(h)(C)(iii)(III)
[xx] U.S. ex rel. Villafane v. Solinger, 543 F.Supp.2d 678, 692 (W.D. Ky 2008)
[xxii] David M. Deaton, What is “Safe” About the Government’s Recent Interpretation of the Anti–Kickback Statute Safe Harbors? … and Since When was Stark an Intent–Based Statute?, 36 J. Health L. 549, 552 (2003)
[xxiii] See supra note 4, at 55,793.