Caught between Scylla and Charibdis: Regulatory Parameters for Designing P4P and Gainsharing Programs Bruce J. Toppin, Esq. Vice President and General Counsel North Mississippi Health Services Daniel F. Murphy, Esq. Partner Bradley Arant Boult Cummings LLP I. Quality Incentive / Pay for Performance (P4P) Programs A. Background The federal health care regulatory environment for quality-based incentives is in flux. CMS and the OIG recognize the potential of such programs to improve the overall quality of health care delivery, but remain concerned that improperly designed programs could become a vehicle for inappropriate financial rewards for patient referrals or other adverse effects on beneficiaries. Both agencies have published guidance concerning pay for performance (P4P) and gainsharing arrangements. In July 2008, CMS published a proposed a Stark Law exception relating to hospital-sponsored quality incentive programs, but in response to extensive industry objections to the structure and scope of the proposed exception, issued a further call for public comment on a number of issues in November 2008. Since making its call for additional comments in November 2008, CMS has taken no additional action to finalize the regulatory exception. The guidelines below reflect a synthesis of available agency commentary and guidance and, except for compliance with one of the Stark exceptions, are intended to reflect best practices rather than mandatory elements. Although a number of OIG advisory opinions contain similar guidance on P4P and gainsharing arrangements, we primarily cite to the following three CMS and OIG policy guidance sources with respect to the guidelines offered below.
FY 2009 Proposed Physician Fee Schedule Revisions 73 Fed. Reg. 130 (July 7, 2008) pages 38548 et seq. (hereinafter Proposed Exception ). OIG Advisory Opinion 08-16 (October 7, 2008) (hereinafter OIG Advisory Opinion ). FY 2009 Final Physician Fee Schedule Revisions 73 Fed. Reg. 224 (November 19, 2008) pages 69793 et seq. (hereinafter CMS Commentary ). B. Regulatory Guidance Recent CMS and OIG policy statements on physician quality incentives have been relatively consistent. Both recognize similar potential advantages and risks associated with such programs. Incentive compensation arrangements like the Proposed Arrangement are designed to align incentives by offering physicians a portion of a hospital s compensation related to meeting quality targets, in exchange for implementing strategies to help the hospital meet those targets. Properly structured, such arrangements can serve legitimate business and medical purposes by improving efficiency and quality of care. However, like any payment arrangement between a hospital and physicians who refer business to the hospital, payments purportedly intended to encourage quality improvements might be misused by unscrupulous parties to induce limitations or reductions in care or to disguise kickbacks for Federal health care program referrals. Therefore, such arrangements must be evaluated in light of applicable Federal statutes and the potential for abuse. 1 Although properly structured incentive payment programs can enhance health care quality and efficiency, improperly structured programs pose significant risks of program or patient abuse, including adversely affecting patient care. Moreover, such programs could be vehicles to disguise payments for referrals, including incentives to steer healthier patients to the hospital offering the incentive payment program. Programs that cannot be adequately and accurately measured for quality would also pose a high risk of program or patient abuse. 2 1 OIG Advisory Opinion p. 5. 2 CMS Commentary p. 69794.
In particular, CMS has stated its intention to promulgate [a Stark] exception that is as broad as possible but that will pose no risk of program or patient abuse. 3 It is not clear, however, when this might occur. C. P4P Best Practices 1. The program s potential to improve the quality of patient care should be supported by objective evidence. Quality measures may include published standards established by CMS. 4 If quality measures are borrowed from another quality standards organization or are internally developed, there should be documentation of independent medical review and validation. Although CMS has not proposed to permit internally developed quality measures in its proposed Stark exception, it has since requested public comment on whether and/or how providers might be permitted to do so. 5 Similarly, the quality measures proposed to be used by the incentive program addressed by OIG Advisory Opinion 08-16 were limited to CMS approved standards, but the opinion did not necessarily foreclose the use of other quality measures under proper circumstances. The agency remains focused on the use of government-approved performance and quality indicators with specific reference to clinical evidence for quality measures found in the CMS Quality Measures Manual. Quality measures should use an objective methodology, be verifiable, be supported by credible medical evidence, and be able to be independently 3 Id. 4 Proposed Exception, p. 38605; OIG Advisory Opinion pp. 9-10. 5 CMS Commentary, p. 69795-6.
tracked. 6 In connection with advisory opinion requests, the OIG may require requestors to engage independent medical experts, selected by the OIG, to review each quality measure for which a proposed arrangement would provide compensation. Quality standards should be reasonably related to the medical needs of the affected patient population. 7 The program should not adversely affect overall health care by encouraging cherry-picking (treating only healthy patients), steering (discouraging the admission of sicker patients), or other inappropriate activities. 8 Quality standards should be reviewed periodically (e.g., annually) and be subject to change in accordance with available medical evidence. 9 2. Physicians participating in the program should be able to affect the achievement of quality goals through specific treatment interventions or other actions recognized as improving patient care. 10 3. The necessity for the active involvement and support of the physicians participating in the program to achieve the desired quality goals should be documented. 11 6 Proposed Exception, p. 38553. 7 OIG Advisory Opinion p. 7; Proposed Exception, p. 38605. 8 CMS Commentary p. 69795. 9 CMS Commentary p. 69795; OIG Advisory Opinion p. 7. 10 OIG Advisory Opinion p. 6; Proposed Exception, p. 38605. 11 OIG Advisory Opinion p. 11.
4. There should be no incentive for a physician to prescribe a medically inappropriate course of treatment to a particular patient 12 or restriction on physicians ability to make medically appropriate decisions for their patients. 13 5. The quality standards, measurement periods, and methodology used to determine eligibility for incentive payments, including specific thresholds above or below which no incentive payments will accrue to physicians, should be clearly identified in advance. 14 6. The effect of the incentive program on the patient population should be carefully monitored and appropriate steps taken to modify or discontinue the program if adverse effects on patient care are detected. 15 The proposed Stark exception would require independent medical review of the effect of an incentive program on the quality of patient care prior to commencement and annually thereafter. 16 CMS later asked for additional public comment on the nature and scope of such independent review. 17 7. Quality-related incentive payments should not be not used as a sham to reward physician referrals. 18 The program should not be used to attract new referring physicians or to increase the volume or value of referrals from existing sources (e.g., by 12 Id. at 7. 13 Proposed Exception, p. 38605. 14 OIG Advisory Opinion pp. 7, 10. Proposed Exception, p. 38605. 15 OIG Advisory Opinion p. 7. 16 Proposed Exception, p. 38605. 17 CMS Commentary p. 69796. 18 Id.
excluding from the incentive formula any increase in the volume or value of federal health program patients referred to a provider in comparison to a baseline period. 19 The incentive program addressed in OIG Advisory Opinion 08-16: o required physicians to be active members of the hospital s medical staff for at least a year in order to be eligible to participate; o capped physician incentive payments using inflation-adjusted baseline compensation received prior to the implementation of the program; and o provided for the termination of a participating physician whose referral patterns significantly changed in a manner beneficial to the hospital during the term of the program. Incentive payments made to a medical group should be allocated on a per capita basis or in some other manner that is not related to the volume or value of referrals from particular group members. 20 CMS has proposed to require physicians to participate in pools of five or more and for payments to be made on a per capita basis. In response to industry objections, CMS has since requested additional public comment on these points. 21 The value of the incentives should be reasonably related to the economic value realized by the hospital for achieving patient quality-related goals. 22 If 19 OIG Advisory Opinion p. 9-10; Proposed Exception, p. 38555. 20 OIG Advisory Opinion p. 10; Proposed Exception, p. 38606. 21 CMS Commentary p. 69797. 22 Id.; see also OIG Advisory Opinion p. 10.
incentive levels are initially based on assumptions or estimates of resulting economic value to the hospital, results should be validated (and incentives modified as necessary) based on actual program experience. The incentive program addressed in OIG Advisory Opinion 08-16 involved a proposal by a hospital to share quality incentive payments received from a commercial insurer with members of the hospital s medical staff. This set an objective value of the program to the hospital. Similarly, the proposed Stark exception would tie physician payments under a shared savings program to the difference between acquisition costs for certain items and supplies during a baseline period and during the operation of the incentive program. There is, however, nothing in the Advisory Opinion or in the CMS commentary indicating that alternative good faith measures of fair market value could not be considered. 8. Eligibility for incentive payments should require achievement of measurably superior performance in comparison to average or expected quality standards (i.e., incentive payments should not be made to physicians for achieving generally accepted standards of quality.) 23 The CMS proposed exception would prohibit incentive payments for quality improvements achieved during a prior period. 24 However, in response to industry objections, CMS has since requested additional public comment on 23 CMS Commentary p. 69796. OIG Advisory Opinion p. 10 fn. 24 Proposed Exception, p. 38606.
whether physicians may be rewarded for the maintenance of patient care quality (as opposed to improvements in patient care quality). 25 9. The program must satisfy a Stark exception (to the extent that DHS services are furnished to patients) presumably the exception for personal services agreements. The CMS commentary explicitly recognizes that properly structured arrangements involving physician participation in an incentive payment or shared savings program may meet the requirements of one or more of the existing physician self-referral exceptions for compensation arrangements. 26 Such agreements must: be in writing; be signed by the parties; specify the services to be performed by the physician (e.g. include a nonexclusive list of actions expected to be taken by participating physicians to improve patient quality); cover all of the services to be furnished by the physician (or if embodied in a separate agreement, be cross-referenced to a master list of all agreements with the physician); not include services in excess of what is reasonable and necessary for achieving a legitimate business objective (e.g., achieving specific quality standards); be for a term of at least one year; 25 CMS Commentary p. 69797. 26 Id. at 69798.
set compensation in advance (e.g., by tying fixed incentive payments to the achievement of specific quality standards or goals), not exceed fair market value, and not take into account the value or volume of referrals or other business generated between the parties. 10. Documentation of quality outcomes and determination of eligibility for incentive payments should be transparent, be administered objectively and consistently for all participating physicians (i.e., no subjective exceptions), and be retained for potential review by federal or state regulatory authorities. 27 11. Affected patients should be made aware of the program. 28 The OIG identifies patient notification as an element of the incentive program that it was asked to review in Advisory Opinion 08-16. The transparency of the Proposed Arrangement allows for public scrutiny and individual physician accountability for adverse effects of the Proposed Arrangement, should any occur. Similarly, the CMS July 2008 proposed Stark Act exception would have required written notice to affected patients (i) identifying the physicians participating in the program, (ii) disclosing the availability of incentive payments, and (iii) disclosing the performance measures use by the program in layman s terms. D. Summary As noted, government policies relating to quality-related incentive programs are in flux. The OIG and CMS recognize the potential value of such programs, but are wary that poorly or improperly designed programs could become vehicles for fraud or abuse. The above guidelines 27 Proposed Exception, p. 38606. 28 OIG Advisory Opinion p. 7; Proposed Exception, p. 38605.
have been derived from published OIG and CMS statements on this subject. Meetings with CMS or OIG staff or formal requests for advisory opinions relating to a specific proposed program structure may also be considered as a way to reducing the degree of regulatory uncertainty associated with physician incentive programs. II. Gainsharing Programs A. Background As understood by most health care providers and regulators, the term gainsharing refers to any arrangement through which a health care provider agrees to pay a physician a share of cost savings that it is able to realize as a result of actions taken by the physician. Such arrangements are designed to align provider-physician financial interests in lowering the costs of care. The OIG has recognized that properly structured cost sharing programs can serve legitimate business and medical purposes by increasing efficiency and reducing waste. 29 CMS has similarly noted that shared savings programs have been recognized by stakeholders as an effective means of controlling costs, improving efficiency, and promoting quality in the delivery of health care services provided that such programs are properly structured to ensure compliance with Federal health care program requirements. 30 However, both agencies have also expressed concern about the potential for poorly designed or sham gainsharing programs to (i) interfere with the independent exercise of a physician s medical judgment in furnishing or prescribing health care items or services to patients and/or (ii) inappropriately influence the volume or value of physician referrals to a provider. This includes so-called black box gainsharing programs that offer payment incentives to physicians for overall cost savings in a department or service line without 29 OIG Advisory Opinion No. 01-01, January 11, 2001, page 6. 30 Proposed Exception, page 38550.
identifying the specific actions that physicians are expected to take in order to achieve such savings. 31 In the government s view, such arrangements often lack objective clinical performance measures to ensure that the quality of patient care is not adversely affected. CMS has solicited public comments on whether adequate safeguards could be designed to prevent patient or program abuse arising from gainsharing programs tied to global savings from a department of service line. 32 B. Regulatory Guidance The OIG has issued a Special Advisory Bulletin (July 1999) as well as 15 advisory opinions directly addressing gainsharing programs for acute care hospitals (Advisory Opinions 01-1, 05-01 through 06, 06-22, 07-21 and 22, 08-09, 08-15, 08-21, 09-06, and 12-22). These documents outline OIG concerns that improperly designed gainsharing programs could encourage: the delivery of substandard patient care; reduced access to health care services for sicker, more costly patients; payments to physicians that reward increases in the volume or value of referrals to a provider; and unfair competition by providers using sham gainsharing programs to attract referrals. The several advisory opinions identify and provide guidance on a number of factors for evaluating gainsharing programs. The OIG s legal analysis focuses primarily on 31 Testimony of Lew Morris to the House Committee on Ways and Means, October 7, 2005, page 3 and Proposed Exception, page 38550. 32 Proposed Exception, page 38557.
The Civil Monetary Penalties ( CMP ) Law 33 prohibition of payments knowingly made by a hospital or a critical access hospital ( CAH ) to a physician to reduce or limit items or services furnished to a Medicare or Medicaid beneficiary under the direct care of the physician; and The Anti-Kickback Statute 34 prohibition of knowing and willful payment or receipt of remuneration to induce the referral of federal health care program patients or services. The CMP Law applies, by its terms, only to payments made by a hospital or CAH to a physician and not to payments made between other types of health care providers. Although the CMP prohibition does not apply directly to payments made by non-hospitals to non-physicians, it may be prudent for other providers to use the OIG analysis as a guidepost for the design of nonhospital gainsharing programs, particularly in light of the recently published rule on fraud and abuse waivers for accountable care organizations ( ACOs ). 35 That rule, issued jointly by CMS and the OIG, referred to 42 USC 1320a-7a(b) as the Gainsharing CMP 36 and noted that it would evaluate not only direct payments between hospitals and physicians, but also indirect payments. The ACO waiver of the Gainsharing CMP requirements will apply to payments of shared savings distributions made directly or indirectly from a hospital to a physician are not made knowingly to induce the physician to reduce or limit medically necessary items or services to patients under the direct care of a physician [emphasis in original]. 37 The government also emphasized in the ACO fraud and abuse waiver rule that its concern under the Gainsharing CMP is the reduction of medically necessary services. The emphasis on medically necessary services 33 42 U.S.C. 1320a-7a(b). 34 42 U.S.C. 1320a-7b(b). 35 76 Fed. Reg. 67,992 (Nov. 2, 2011). 36 Id. at 67,997. 37 76 Fed. Reg. at 68,001.
suggests that the government does not want to chill providers efforts to cut out wasteful or unnecessary costs, items and services. Factors similar to those identified in the OIG s gainsharing advisory opinions have also been addressed by CMS in federal register notices for the Proposed Exception. 38 However, neither agency has published a formal Anti-Kickback Statue safe harbor or Stark Law exception for gainsharing programs nor explicitly addressed the application of regulatory principles to gainsharing programs for non-hospital settings. C. Gainsharing Best Practices 1. Specific cost-saving initiatives expected to be undertaken by physicians and resulting savings should be clearly and separately identified. In each of the favorable advisory opinions issued by the OIG, the agency made a point of noting that specific cost-savings recommendations were identified in advance and were subject to independent medical review. 39 2. There should be credible medical evidence that implementation of cost savings initiatives will not adversely affect patient care. Each of the gainsharing arrangements that received a favorable advisory opinion from the OIG also included a provision for ongoing review of specific quality measures designed to detect and mitigate any such effect. 40 CMS has proposed a similar quality review standard, but places additional emphasis on the use of an independent entity to perform the review. 41 38 Proposed Exception and CMS Commentary, pages 69793 et seq. 38557. 39 See, e.g., Advisory Opinion 01-01, pages 8 and 13 and Proposed Exception, pages 38550 and 40 See, e.g., Advisory Opinion 01-01, pages 8-9. 41 Proposed Exception, pages 38550 and 38553.
3. Gainsharing incentives should apply to all patients, regardless of payor type. 42 4. Payments to physicians should exclude cost savings associated with an increase in the volume or value of federal health program patients referred to a provider in comparison to a baseline period. 43 5. Cost savings should be calculated using actual out-of-pocket supply costs rather than estimates or artificial accounting conventions. 44 6. Objective clinical measures should be used to establish baseline utilization thresholds below which no savings will accrue to participating physicians. Such measures should be reasonably related to the relevant provider s practices and patient population. For example, Advisory Opinion 01-01 cited a hospital analysis of utilization patterns for certain less costly sutures used in its operating rooms (and by those of other hospitals with similar practices and patient populations). The study concluded that it would be medically reasonable for the hospital to use such sutures in a certain percentage of surgeries (the specific percentage is redacted in the advisory opinion) without adversely affecting the quality of care. The OIG approved the sharing of cost savings attributable to the use of the sutures in hospital surgeries up to, but not above, the percentage identified in the utilization analysis. 45 Advisory Opinion 07-21 approved a gainsharing arrangement where no minimum thresholds were able to be set because the proposed savings initiatives appeared to have no clinical significance 42 See, e.g., Advisory Opinion 01-01, page 9 and Proposed Exception, pages 38556 and 38558. 43 See, e.g., Advisory Opinion 01-01, page 13 and Proposed Exception, page 38555. 44 See, e.g., Advisory Opinion 01-01, page 9 and Proposed Exception, page 38555. 45 Advisory Opinion 07-21, page 5.
or were in accordance with national best practice data and other quality indicators. In this instance, the OIG approved the use of more general quality indicators established by the Society of Thoracic Surgeons (STS) to correlate patient outcomes to specific operating room practices. No cost savings were to be shared for procedures which did not meet STS quality standards. 7. Physicians should have access to the same selection of devices, supplies (including pharmaceuticals), and services (including lab testing) as they did prior to the implementation of a gainsharing program as well as other items deemed necessary for the care of an individual patient. 46 8. Affected patients should receive written notice of a gainsharing program, including an opportunity to review specific cost savings initiatives. 47 9. Financial incentives offered to physicians should be reasonably limited in amount (e.g., by a fixed percentage of actual cost savings achieved through the program) and duration (e.g., one to three years). The arrangements receiving favorable OIG advisory opinions limited shared savings to 50 percent of actual cost reductions. CMS has sought public comment on whether a specific ceiling (higher or lower than 50 percent) would be appropriate. 48 10. Payments to a particular physician (directly as an individual or indirectly as part of a group) should be made on a per capita basis. The OIG believes that this will reduce the incentive for an individual physician to generate disproportionate cost 46 Advisory Opinion 05-01, page 9 and Proposed Exception, page 38554. 47 Advisory Opinion 01-01, page 9 and Proposed Exception, page 38557. 48 Advisory Opinion 01-01, pages 9 and 13 and Proposed Exception, page 38555.
savings. In addition, payments to physicians groups should not be shared with members who do not actively participate in the designated cost reduction initiatives. 49 CMS has proposed similar payment rules, but has also sought public comment on whether payments should be made to physician pools of not less than 5 members and whether gainsharing agreements with group practices should be required to include explicit pass through provisions. 50 11. Participation in a gainsharing program should be limited to physicians already on the medical staff of, or credentialed by, the provider in order to reduce the likelihood that a gainsharing program will be used to attract new referring physicians. The gainsharing arrangement described in Advisory Opinion 01-01 limited participation during the first year to a group of cardiac surgeons on the medical staff of a hospital, but left the door open for extending eligibility on substantially comparable terms to other cardiac surgeons on the medical staff in subsequent years. 51 CMS has similarly proposed that program participation be limited to current medical staff members, but has sought public comment on whether to extend eligibility to new physicians who join a medical staff as the result of routine turnover or workload demand. 52 12. A mechanism should be developed for monitoring changes in patient referral patterns related to age, severity of illness or condition, or payor type. For example, the gainsharing program addressed by Advisory Opinion 01-01 included 49 Advisory Opinion 01-01, page 13. 50 Proposed Exception, page 38556. 51 Advisory Opinion 01-01, pages 12-13 52 Proposed Exception, page 38554
a provision for monitoring patient referrals using generally-accepted standards (presumably, available demographic and clinical statistics). Physicians who significantly modified referral patterns from prior periods were subject to termination from the program. 53 13. Care should be taken in allowing participating physicians to receive a share of out-year cost savings. The majority of the gainsharing arrangements that have received favorable OIG advisory opinions were limited to one year (i.e., participating physicians received a share of savings attributable to changes in clinical or administrative practices only for the year during which they were implemented). 54 Two more recent opinions addressed gainsharing programs extending two to three years in which cost benchmarks were re-based annually using prior year performance data. 55 In neither case, was the OIG willing to allow participating physicians to receive a share of ongoing savings measured from a cost benchmark more than one year old. Multiple-year gainsharing arrangements raise a particular concern, in that they can inappropriately carry over earlier-accomplished savings across years, effectively accounting for them more than once. The resulting unearned duplicate payments can amount to unlawful kickbacks from hospitals to physicians, if accompanied by illicit intent. The annual rebasing method adopted by the Requestors removes earlier accomplished savings from the 53 Advisory Opinion 01-01, page 5 and Proposed Exception, page 38557. 54 See, e.g., Advisory Opinion 01-01, pages 9 and 13. 55 See Advisory Opinion 08-15, page 6 and Advisory Opinion 08-21, page 7.
accounting and thereby avoids improper duplication of physician payments, reducing the accompanying risk of kickbacks. 56 CMS has expressed similar concerns about sharing of out-year savings from cost reduction initiatives. But it has also sought public comment on an alternative scaling mechanism which would allow multi-year sharing of cost savings from a fixed initial benchmark, though at a declining percentage. CMS offers the example of an arrangement in which savings attributable to cost reduction initiatives would be measured against a fixed baseline, but the percentage of such savings eligible to be shared with participating physicians would decline from 50 percent in the first year, to 35 percent in the second year, and to 20 percent in the third year. 57 Some physicians could be inclined to challenge the OIG s characterization of outyear payments for physician-initiated clinical or administrative cost savings as unearned or duplicative. In practice, the scaling model described by CMS more appropriately recognizes physicians interests in receiving a fair share of ongoing savings realized by a provider as a result of physician initiatives. It also offers greater incentives for long term institutionalization of cost reduction practices. However, given publicly stated OIG and CMS concerns about misusing gainsharing programs to reward physicians for expected standards of performance, it would seem prudent to place reasonable limits on the length of time that cost reductions will be shared with physicians. The annual productivity improvement assumptions used by MedPAC and CMS in making annual 56 Advisory Opinion 08-15, page 13. 57 Proposed Exception, page 38556.
adjustments to provider payment rates may provide a basis for calculating such a limit (i.e., once savings have been subsumed into base payment rates, the rationale for continued sharing loses effect). 14. The program must satisfy a Stark exception (to the extent that DHS services are furnished to patients). 15. Documentation of quality outcomes and determination of eligibility for incentive payments should be transparent, be administered objectively and consistently for all participating physicians (i.e., no subjective exceptions), and be retained for potential review by federal or state regulatory authorities. 58 D. Summary As noted, both the OIG and CMS recognize the potential value of gainsharing programs, but are wary that poorly or improperly designed programs could become vehicles for fraud or abuse. The OIG has listed a series of gainsharing program features that it believes heighten the risk of program or patient abuse. These include: no direct connection between physician actions and reductions in the hospital s outof-pocket costs; failure to identify the individual actions that are expected to give rise to the savings; insufficient safeguards against the risk that the other, unidentified actions, such as reduction of medically necessary services, might actually account for any savings; quality of care indicators are of questionable validity or statistical significance; and 58 See, e.g., Advisory Opinion 01-01, page 10 and Proposed Exception, page 38606.
there is no independent verification of cost savings, quality of care indicators, or other essential aspects of the arrangement. 59 Because the above guidelines have been derived from published OIG and CMS statements on gainsharing, it may be prudent to address each of the elements in some manner even if to document that one or more should not apply to a program based on its particular facts and structure. Meetings with CMS or OIG staff or formal requests for advisory opinions relating to a specific proposed program structure may also be considered as a way to reducing the degree of regulatory uncertainty associated with physician incentive programs. 59 See, e.g., Advisory Opinion 01-01, page 10.