Medicare and Medicaid Programs; Organ Procurement Organizations Conditions for Coverage: Revisions to the Outcome Measure Requirements for Organ Procurement Organizations
Federal RegisterDec 2, 2020
Ask Donna
What actually matters in this document.
Text
DEPARTMENT OF HEALTH AND HUMAN SERVICES
Centers for Medicare & Medicaid Services
42 CFR Part 486
[CMS-3380-F]
RIN 0938-AU02
Medicare and Medicaid Programs; Organ Procurement Organizations Conditions for Coverage: Revisions to the Outcome Measure Requirements for Organ Procurement Organizations
AGENCY:
Centers for Medicare & Medicaid Services (CMS), HHS.
ACTION:
Final rule.
SUMMARY:
This final rule revises the Organ Procurement Organizations (OPOs) Conditions for Coverage (CfCs) to increase donation rates and organ transplantation rates by replacing the current outcome measures with new transparent, reliable, and objective outcome measures and increasing competition for open donation service areas (DSAs).
EFFECTIVE DATE:
These regulations are effective on February 1, 2021, except for amendment number 3 (further amending § 486.302), which is effective July 31, 2022.
Implementation date:
The regulations will be implemented on August 1, 2022.
FOR FURTHER INFORMATION CONTACT:
Diane Corning, (410) 786-8486; Jesse Roach, (410) 786-1000; Kristin Shifflett, (410) 786-4133; CAPT James Cowher, (410) 786-1948; or Alpha-Banu Wilson, (410) 786-8687.
SUPPLEMENTARY INFORMATION:
Table of Contents
To assist readers in referencing sections contained in this preamble, we are providing a Table of Contents.
I. Background
A. The Importance of Organ Procurement Organizations and the Need To Reform the Organ Procurement System
B. Statutory and Regulatory Provisions
C. HHS Initiatives Related to OPO Services and Executive Order (E.O.) 13879
II. Summary of the Proposed Provisions and Responses to Public Comments
A. General Comments
B. Proposed Changes to Definitions (§ 486.302) and Proposed Changes to Outcome Requirements (§ 486.318)
1. General Comments About the Outcome Measures
2. Donation Rate § 486.318(d)(1)
3. Donor Definition § 486.302 and the “Zero Organ Donors”
4. Organ Transplantation Rate § 486.302 and § 486.318(d)(2)
5. Organ Definition § 486.302
6. Donor Potential (§ 486.302 and § 486.318(d)(4))
a. Death That Is Consistent With Organ Donation § 486.302
i. Death Certificate Data
ii. International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM)
b. Age 75 and Younger
c. Inpatient Deaths
d. Waiver Hospitals
7. Risk-Adjustments § 486.302 and § 486.318(d)(2)
a. Chronic Diseases
b. Race
c. Gender and Age
d. Ventilator Status
8. OPO Performance on Outcome Measures § 486.318(e) and § 486.302
9. Non-Contiguous States, Commonwealths, Territories, or Possessions § 486.318(e)(7)
10. Assessment and Data for the Outcome Measures § 486.318(f)
11. Implementation Timeline
12. Definitions § 486.302
C. Re-Certification and Competition Processes (§ 486.316)
1. Re-Certification of OPOs § 486.316(a)
2. De-Certification and Competition § 486.316(b)
3. Criteria to Compete § 486.316(c)
4. Criteria for Selection § 486.316(d)
5. Extension of the Agreement Cycle for Extraordinary Circumstances § 486.316(f)
D. Reporting of Data § 486.328
E. Proposed Change to the Quality Assessment and Performance Improvement Requirement (§ 486.348)
1. Death Record Review in QAPI
F. Response to Solicitation of Comments
1. Out of Scope
III. Provisions of the Final Rule
A. Proposed Changes to Definitions (§ 486.302) and Proposed Changes to Outcome Requirements (§ 486.318).
B. Re-Certification and Competition Processes (§ 486.316)
C. Proposed Change to the Quality Assessment and Performance Improvement Requirement (§ 486.348)
D. Solicitation of Comments (Including Changes to Re-Certification Cycle)
IV. Collection of Information Requirements
A. ICRs Regarding Re-Certification and Competition Processes (§ 486.316)
B. ICRs Regarding Condition: Reporting of Data (§ 486.328)
C. ICRs Regarding Quality Assessment and Performance Improvement (§ 486.348)
V. Regulatory Impact Analysis
A. Statement of Need
B. Scope of Review
C. Effects on OPO Performance
D. Anticipated Costs and Benefits
1. Effects on Medical Costs
2. Effects on Patients
3. Implementation and Continuing Costs
E. Effects on Medicare, Medicaid, and Private Payers
F. Effects on Small Entities, Effects on Small Rural Hospitals, Unfunded Mandates, and Federalism
1. Regulatory Flexibility Act
2. Small Rural Hospitals
3. Unfunded Mandates Reform Act
4. Federalism
G. Alternatives Considered
1. Changes to the Denominator
a. CALC as the Denominator
b. All Deaths, Age <= 75 as the Denominator
c. Total Population, Age <75
2. Changing the Confidence Interval
3. Changing the Threshold Rates
H. Accounting Statement and Table
I. Reducing Regulation and Controlling Regulatory Costs
J. Conclusion
Regulations Text
I. Background
A. The Importance of Organ Procurement Organizations and the Need To Reform the Organ Procurement System
Organ procurement organizations (OPOs) are vital partners in the procurement, distribution, and transplantation of human organs in a safe and equitable manner for all potential transplant recipients. The role of OPOs is critical to ensuring that the maximum possible number of transplantable human organs is available to individuals with organ failure who are on a waiting list for an organ transplant. There are currently 58 OPOs that are responsible for identifying eligible donors and recovering organs from deceased donors in the United States (U.S.), with no current statutory authority to add new OPOs. Therefore, the Centers for Medicare & Medicaid Services (CMS) views OPO performance as a critical element of the organ transplantation system in the United States (U.S.)
As of November 2020, a total of 108,725 people were on the waiting lists for a lifesaving organ transplant.
1
Many people face tremendous quality of life burdens or death while on the waiting list. An OPO that is efficient in procuring organs and delivering them to recipients will help more people on the waiting list receive lifesaving organ transplants, which could ultimately save more lives.
1
Organ Procurement and Transplantation Network (OPTN) Data.
https://optn.transplant.hrsa.gov/data/
.
Based on public feedback and our own internal analysis of organ donation and transplantation rates, it is the agency's belief that the current OPO outcome measures are not sufficiently objective and transparent to ensure appropriate accountability in assessing OPO performance, nor do they properly incentivize the adoption of best practices and optimization of donation and organ placement rates.
Given OPOs' important role in the organ donation system in the U.S., some stakeholders have stated that underperformers have faced few consequences for poor performance, by noting “Performance varies across the OPO network, with many persistent underperformers failing to improve over the last decade.”
2
They further note that there are serious negative impacts to both organ transplantation and donation when OPOs are underperforming, in that “[w]hen OPOs are inefficient or ineffective, donor hospitals are reluctant to refer potential donors, and transplant programs have fewer organ offers for patients on the waiting list. The end result is a bottleneck within the system that leads to avoidable deaths and increased national health care spending.”
3
2
The Bridgespan Group. Reforming Organ Donation in America.
https://www.bridgespan.org/bridgespan/Images/articles/reforming-organ-donation-in-america/reforming-organ-donation-in-america-12-2018.pdf.
3
ORGANIZE. Organ Donation Reform Report, 2019.
Some stakeholders, including members of the OPO industry, have stated that the current OPO outcome measures should be reformed to incentivize improvements in OPO performance. Some of these stakeholders note that “[e]xisting regulations need dramatic improvement to remove perverse incentives to organ procurement (for example, OPOs are evaluated on the number of organs procured per donor, which leads to older single-organ donors being overlooked) and increase continuous performance accountability.”
2
Reforming the current outcome measures can be achieved, they indicated, through metrics that improve accountability and “by replacing current ineffective metrics for OPO performance with a simplified transparent metric that enables independent performance measurement.”
2
B. Statutory and Regulatory Provisions
To be an OPO, an entity must meet the applicable requirements of both the Social Security Act (the Act) and the Public Health Service Act (the PHS Act). Section 1138(b) of the Act provides the statutory qualifications and requirements that an OPO must meet in order for organ procurement costs to be paid under the Medicare program or the Medicaid program. Section 1138(b)(1)(A) of the Act specifies that payment may be made for organ procurement costs only if the agency is a qualified OPO operating under a grant made under section 371(a) of the PHS Act or has been certified or re-certified by the Secretary of the Department of Health and Human Services (the Secretary) as meeting the standards to be a qualified OPO within a certain time period. Section 1138(b)(1)(C) of the Act provides that payment may be made for organ procurement costs “only if” the OPO meets the performance-related standards prescribed by the Secretary. Section 1138(b)(1)(F) of the Act requires that to receive payment under the Medicare program or the Medicaid program for organ procurement costs, the entity must be designated by the Secretary. The requirements for such designation are set forth in 42 CFR 486.304 and include being certified as a qualified OPO by CMS.
Pursuant to section 371(b)(1)(D)(ii)(II) of the PHS Act, the Secretary is required to establish outcome and process performance measures for OPOs to meet based on empirical evidence, obtained through reasonable efforts, of organ donor potential and other related factors in each service area of the qualified OPO. Section 1138(b)(1)(D) of the Act requires an OPO to be a member of, and abide by the rules and requirements of, the Organ Procurement and Transplantation Network (OPTN). OPOs must also comply with the regulations governing the operation of the OPTN (42 CFR part 121). The Department of Health and Human Services (HHS) has explained that only those policies approved by the Secretary will be considered “rules and requirements” of the OPTN for purposes of section 1138 of the Act. The OPTN is a membership organization that links all professionals in the U.S. organ donation and transplantation system. Currently, the United Network for Organ Sharing (UNOS) serves as the contractor for the operation of the OPTN under contract with HHS. OPOs are required under the OPTN final rule (42 CFR 121.11(b)(2)) and 42 CFR 486.328 of the OPO CfCs to report information specified by the Secretary to the OPTN, including the data used to calculate the outcome measures for OPOs.
In addition, OPOs are required to comply with title VI of the Civil Rights Act of 1964, 42 U.S.C. 2000d (title VI), section 504 of the Rehabilitation Act of 1973, 29 U.S.C. 794 and section 1557 of the Patient Protection and Affordable Care Act, 42 U.S.C. 18116 (section 1557). Title VI and section 1557, protect individuals on the basis of race, color and national origin. Under these laws, OPOs are required to take reasonable steps to ensure meaningful access to their programs by individuals with limited English proficiency. Reasonable steps may include providing language assistance services at no cost, such as providing interpreters or translated material. Also, section 504 and section 1557 protect qualified individuals with a disability, including prospective organ recipients with a disability and prospective organ donors with a disability, from discrimination in the administration of organ transplant programs. Under these laws, OPOs must ensure that qualified individuals with a disability are afforded opportunities to participate in or benefit from the organ transplant programs that are equal to opportunities afforded others. Decisions to approve or deny organ transplants must be made based on objective facts related to the individual in question. “Individuals with disabilities are also entitled to reasonable accommodations needed to participate in and benefit from a program, and auxiliary aids and services needed for effective communication. These rights extend in some circumstances to family members of a prospective organ donor or recipient. For example, health care providers and organ donation programs are required to provide auxiliary aids and services (including sign language interpreters) when necessary for effective communication between a relative involved in a prospective donor or recipient's care and a health care provider or donation program.” Additionally, if eligibility criteria for being a transplant recipient require an individual to be able to comply with post-transplant regimens, it would be a reasonable accommodation to allow an individual with a developmental or intellectual, or other disability to meet that requirement with the assistance of a relative, attendant, or other individual.
We established CfCs for OPOs at 42 CFR part 486, subpart G, and OPOs must meet these requirements in order to be able to receive payments from the Medicare and Medicaid programs. These regulations set forth the certification and re-certification processes, outcome requirements, and process performance measures for OPOs and became effective on July 31, 2006 (71 FR 30982), which we refer to as the “2006 OPO final rule”. The current outcome measures, found under § 486.318, are used to assess OPO performance for re-certification and competition purposes (see § 486.316(a) and (d)).
Section 486.322 requires that an OPO must have a written agreement with 95 percent of the Medicare- and Medicaid-participating hospitals and critical access hospitals in its service area that have both a ventilator and an operating room, and have not been granted a waiver by CMS to work with another OPO. Meanwhile, 42 CFR 482.45
requires a hospital have written protocols that incorporate an agreement with an OPO under which it must notify, in a timely manner, the OPO or a third party designated by the OPO, of individuals whose death is imminent or who have died in the hospital. Potential organ donors may encounter Medicare- and Medicaid-certified providers prior to an emergency department visit or hospital admission to a critical care unit. Therefore, we expect that each OPO's responsibilities and work begins long before a hospital notifies the OPO of an impending death—through, but not limited to, extensive training and education of all Medicare and Medicaid-certified providers along the continuum of care and by fostering a collaborative relationship among them.
C. HHS Initiatives Related to OPO Services and Executive Order (E.O.) 13879
In 2000, the Secretary's Advisory Committee on Organ Transplantation (ACOT) was established under the general authority of section 222 of the PHS Act, as amended. See 42 CFR 121.12. ACOT is charged to: (1) Advise the Secretary, acting through the Administrator, Health Resources and Services Administration (HRSA) on all aspects of organ donation, procurement, allocation, and transplantation, and on such other matters that the Secretary determines; (2) advise the Secretary on federal efforts to maximize the number of deceased donor organs made available for transplantation and to support the safety of living organ donation; (3) at the request of the Secretary, review significant proposed OPTN policies submitted for the Secretary's approval to recommend whether they should be made enforceable; and (4) provide expert input to the Secretary on the latest advances in the science of transplantation, the OPTN's system of collecting, disseminating and ensuring the validity, accuracy, timeliness and usefulness of data, and additional medical, public health, patient safety, ethical, legal, financial coverage, social science, and socioeconomic issues that are relevant to transplantation.
4
4
https://www.organdonor.gov/about-dot/acot/charter.html
.
A 2012 recommendation by ACOT stated: “The ACOT recognizes that the current CMS and HRSA/OPTN structure creates unnecessary burdens and inconsistent requirements on transplant centers (TCs) and OPOs and that the current system lacks responsiveness to advances in TCs and OPO performance metrics. The ACOT recommends that the Secretary direct CMS and HRSA to confer with the OPTN, Scientific Registry of Transplant Recipients (SRTR), the OPO community, and TCs representatives to conduct a comprehensive review of regulatory and other requirements, and to promulgate regulatory and policy changes to requirements for OPOs and TCs that unify mutual goals of increasing organ donation, improving recipient outcomes, and reducing organ wastage and administrative burden on TCs and OPOs. These revisions should include, but not be limited to, improved risk adjustment methodologies for TCs and a statistically sound method for yield measures for OPOs . . . .”
5
5
Available at:
https://www.organdonor.gov/about-dot/acot/acotrecs55.html
.
On July 10, 2019, President Trump issued Executive Order (E.O.) 13879 titled “Advancing American Kidney Health.” The E.O. 13879 states that it is the policy of the U.S. to “prevent kidney failure whenever possible through better diagnosis, treatment, and incentives for preventive care; increase patient choice through affordable alternative treatments for end-stage renal disease (ESRD) by encouraging higher value care, educating patients on treatment alternatives, and encouraging the development of artificial kidneys; and increase access to kidney transplants by modernizing the organ recovery and transplantation systems and updating outmoded and counterproductive regulations.”
Further, the E.O. 13879 aims to increase the utilization of available organs by ordering that, within 90 days of the date of the order, the Secretary propose a regulation to enhance the procurement and utilization of organs available through deceased donation by revising OPO rules and evaluation metrics to establish more transparent, reliable, and enforceable objective outcome measures for evaluating an OPO's performance. In conjunction with the E.O. 13879, HHS set a goal to deliver more organs for transplantation and aims to double the number of kidneys available for transplant by 2030.
6
6
https://aspe.hhs.gov/system/files/pdf/262046/AdvancingAmericanKidneyHealth.pdf.
In accordance with the E.O. 13879, we published a proposed rule in the
Federal Register
on December 23, 2019 entitled, “Medicare and Medicaid Programs; Organ Procurement Organizations Conditions for Coverage: Revisions to the Outcome Measure Requirements for Organ Procurement Organizations” (84 FR 70628 through 70710), (referred to as the “December 2019 OPO proposed rule”), which proposed to revise the current OPO outcome and process measures to be more transparent, reliable, and provide enforceable objective outcome measures of OPO performance. The December 2019 OPO proposed rule would improve upon the current measures by using objective and reliable data, incentivize OPOs to ensure all viable organs are transplanted, hold OPOs to greater oversight while driving higher performance, and as a result, save more lives.
II. Summary of the Proposed Provisions and Responses to Public Comments
In response to the December 2019 OPO proposed rule (84 FR 70628 through 70710), we received approximately 834 total comments. Commenters included individual OPOs, transplant hospitals, national associations and coalitions, academic researchers, advocacy organizations, health care professionals and corporations, donor families, and numerous individuals from of the general public. Most commenters supported the proposed rule's goals to improve organ donation and transplantation in the U.S. and to update the current OPO outcome measures.
In this final rule, we provide a summary of each proposed provision, a summary of the public comments received and our responses to them, and an explanation for changes in the policies that we are finalizing. We note that this final rule is written in order by topic, discussing our primary reason for revising the regulations by revising the outcome measures first, and then discussing some necessary changes and cross-cutting requirements.
A. General Comments
The majority of the comments received on the December 2019 OPO proposed rule were received from the general public and organ donor families.
Comment:
The majority of the commenters asked for OPOs to be held accountable for poor performance and for additional oversight of OPOs. Some of the commenters expressed concern that the OPOs are operating as monopolies that are engaged in fraud, waste, and abuse. Many commenters asked CMS to increase the accessibility of organs for transplant and ensure that donated organs reach the many individuals on the organ transplant waitlist.
The comments received from donor families expressed support for the OPO in their service area and expressed concern that the proposed changes
would lead to the decertification of their assigned OPO.
Response:
We appreciate the commenters' feedback. The tremendous amount of comments that we received asking for OPOs to be held accountable strongly supports our commitment to reform the organ donation and transplantation system in the U.S. We believe that the changes we are making in assessing OPO performance will ensure positive outcomes and increases in the organ supply. There are other initiatives that HHS and CMS are currently undertaking that will also lead to improvements in organ donation and transplantation, such as the ESRD Treatment Choice (ETC) Kidney Transplant Learning Collaborative.
We also appreciate the time taken by numerous donor families to develop and submit thorough and thoughtful comments on the proposed rule. We understand that the decision to donate a family member's organs is difficult, and we praise these families for their generosity. We acknowledge that the decision to donate their loved one's organs likely saved or improved the recipient's life. The changes that we discuss in this final rule are intended to ensure that donated organs are not wasted and reach those waiting for a lifesaving organ transplant. It is our goal to ensure that OPOs are held to a high level of performance expectations and that all OPOs are pushed to perform better. We acknowledge that through changes to the procurement and transplantation process (such as enacting best practices) we can effect visible changes that can lead to an increase in the number of organs available for transplant and decreases in organ discards. We acknowledge commenters' concerns regarding decertification of OPOs and note that that is a likely potential outcome due to these new measures. However, CMS is committed to ensuring patient access to high quality health care, including access to high performing OPOs. Additionally, we believe the measures will incentivize OPOs to improve result in upward performance across most, if not all, OPOs.
Comment:
Several commenters criticized our reference to the Bridgespan study
2
and objected to our characterization of the failures of OPOs. These commenters also expressed concern about negative stories in the media suggesting that OPOs are poorly performing and do not care about the families they serve. The commenters stated that when media stories share inaccurate or outright false information about the OPO community, these stories have the strong potential to hurt public perceptions about donation.
Response:
We understand that there have been several news articles about the poor performance of some OPOs and some of these articles raise reasons about why the organ donation system needs to be reformed. Our independent assessment of OPO performance on outcome measures is the basis for our belief that more oversight is needed. As of November 2020, a total of 108,591 people were on the waiting lists for lifesaving organ transplants. An OPO that is efficient in procuring organs and delivering them to recipients will help more people on the waiting list receive an organ transplant, which could ultimately save more lives. The current OPO outcome measures are not sufficiently objective and transparent to ensure public trust in assessing OPO performance, nor do they properly incentivize the adoption of best practices and optimization of donation and organ replacement rates. Given these concerns, as well as those regarding the data quality of self-reported measures, we are finalizing new outcome measures at § 486.318 to hold OPOs accountable as a crucial step in reforming the organ donation system.
B. Proposed Changes to Definitions (§ 486.302) and Proposed Changes to Outcome Requirements (§ 486.318)
In the December 2019 OPO proposed rule, we proposed to revise the outcome measures for re-certification and the corresponding changes in definitions at §§ 486.302 and 486.318 to replace the current outcome measures and definitions. We proposed at § 486.302 the definition of “donation rate” as the number of donors as a percentage of the donor potential. We also proposed to add “donor potential,” as the number of inpatient deaths within the donation service area (DSA) among patients 75 years of age and younger with any cause of death that would not be an absolute contraindication to organ donation. We also proposed to define the term “organ transplantation rate,” which is discussed in more detail in section II.B.4 of this final rule and changes related to our use of “death that is not an absolute contraindication to organ donation” at § 486.302 of this final rule. We refer readers to section II.B of this final rule for the other definitions we proposed at § 486.302. We proposed to revise the outcome measures for re-certification at § 486.318 to replace the current existing outcome measures with the proposed two new outcome measures that would be used to assess an OPO's performance: “donation rate” and “organ transplantation rate” effective for calendar year (CY) 2022. We have organized the comments by subject matter.
The comments and our responses are below.
1. General Comments About the Outcome Measures
Comment:
Several commenters, supported our proposed new outcome measures while other commenters questioned the need for revising them. Some commenters in support of the proposed new outcome measures recognized that these measures are derived from objective data and would not present an increased burden on OPOs. One commenter was concerned that the proposed new outcome measures would result in the OPOs putting more pressure on families and next-of-kin to authorize organ donations. Other commenters expressed concern that increased pressure from the proposed outcome measures and the threat of de-certification (discussed in section II.C of this final rule) would damage the relationships between the OPOs so that they will no longer cooperate with one another.
Response:
We appreciate the comments received on the proposed outcome measures. Under our current regulations, the outcome measures at § 486.318 are used to assess OPO performance for re-certification and to determine the selection of an OPO to take over a DSA as part of the competition for an open DSA. We think the increased transparency and objectivity of the proposed outcome measures will drive improvements in organ donation and transplantation while reducing reporting burdens for OPOs. As discussed in section II.B.1 of this final rule, there are aspects of our current outcome measures that we no longer find adequate; therefore, we believe that revising the current outcome measures would be consistent with the goal of E.O. 13879, which directs CMS to establish more transparent, reliable, and enforceable objective measures for evaluating an OPO's performance. In addition, we believe revising the current outcome measures is a critical step towards achieving the Secretary's goal of doubling kidneys available for transplantation by 2030.
Our proposed outcome measures are based on verifiable and objective data and are designed to increase organ transplantations by comparing an OPO's performance to the top performing OPOs. The top performing OPOs have demonstrated success in their role and responsibilities using practices that do not place inappropriate pressures on
families to consent to an organ donation. We note that studies have shown that giving families sufficient time to make their informed decisions and not putting too much pressure on families results in successful consent.
7
We also note that by objectively identifying top performing OPOs, poorer performing OPOs can appropriately change and adopt their effective practices that improve their performance in donation and make more organs available for transplantation.
7
Siminoff, LA, Agyemang, AA, et al., “Consent to organ donation: a review,” Prog Transplant. 2013 Mar; 23(1): 99-104.
It is clear that our historical approach to measuring OPO performance has resulted in a wide range of performances. This variability is unacceptable to patients and CMS. Thus, CMS intends to hold these entities to revised and higher standards. These revised metrics are necessary in light that over the past 14 years, the sharing of best practices, if it has occurred, has not resulted in consistent improvements throughout the industry. CMS is committed to increasing the availability of organs for transplantation across all DSAs, and continues to believe that this higher standard is necessary to achieve this goal.
Comment:
We received multiple comments raising concerns that removing organs for research, other than the pancreata, as part of the outcome measures would hurt research by discouraging OPOs from using organs that are not transplanted for research. A commenter recommended CMS to include organs that are used in organ transplantation research in the outcome measures because the process for obtaining consent and managing these donors is the same as with organ transplantation. Other commenters suggested that we include organs for research as a “third performance metric” or incorporate it in some other way into our conditions. One commenter discussed our history of inclusion of organs for research and stated that OPOs would not pursue marginal organs because they would no longer get credit if the organ was not transplanted, whereas the old outcome measures allowed them to be counted to the organs for research measure.
Response:
We appreciate the comments raising concerns about the removal of other organs used for research as part of the outcome measures.
The transplant and research communities commonly described the transplantation of organs into humans using research protocols (for example, deceased donor intervention research) as both transplants and research. Generally, such research involves the transplantation of organs into transplant candidates that is generally considered clinical care while simultaneously qualifying as human subject research. For the purpose of establishing the performance measures, we contend that organs used for research is meant to apply only to organs procured and used only for research purposes. Organs transplanted into human subjects are counted as part of clinical care. Although organs procured for research may sometimes involves the same procedures and practices of donor management as organs for transplantation, we cannot easily verify the procurement of organs for research unless they are transplanted into a patient on the OPTN waiting list. Furthermore, our concern is that having an outcome measure for organs procured for research may inadvertently incentivize OPOs to direct some organs for research rather than for transplantation. Except for pancreata when procured for research, as noted in the December 2019 OPO proposed rule, we are not adopting the commenters' suggestion to include organs donated for research in the outcomes measures.
Pancreata procured for islet cell research are included in the outcome measures of this final rule. We carefully considered other options to address pancreata procured for research, such as creating a process measure for these organs, creating a unique outcome measure, and counting these organs in the outcome measures of this final rule as less than the full value of a transplanted organ. However, these alternative policy approaches did not meet the PHS Act, which states that “Pancreata procured by an organ procurement organization (OPO) and used for islet cell transplantation or research shall be counted for purposes of certification or recertification . . . .” To meet this statutory requirement, we have chosen to include pancreata for research in the outcome measures in the same way that organs procured for transplantation are included. We think that the impact of pancreata for research on the overall rankings of OPOs will continue to be minimal. From 2014 to 2018 (the most recent year of complete data), the number of pancreata procured for research has been 727, 716, 575, and 579. There is a clear downward trend, and we expect that this trend will continue or level off. Our internal analysis demonstrated little effect on the rankings of OPOs from including or excluding pancreata for research when calculating performance on both the donor and transplant measures of this final rule. A particular OPO may move up or down 1-3 ranking spots based on the inclusion of this data, but no OPOs moved in such a way that it impacted whether they would be eligible for automatic recertification or would be automatically decertified. We will continue to monitor the trends of pancreata procured for research and will use the survey process to conduct further investigation into any anomalies that such monitoring reveal.
Comment:
We solicited comments as to whether we should consider assessing OPO performance based on organ-specific transplant rates and received a comment that broadly supported this approach, but we did not receive details about how we would measure success for the organ-specific transplant rates or how it could be implemented.
Response:
We are not including organ-specific transplantation rates in our outcome measures because we do not believe that organ-specific transplantation rates would provide an additive assessment of OPO performance and achieve our goals of increasing organs available for transplantation.
2. Donation Rate §§ 486.302 and 486.318(d)(1)
In the December 2019 OPO proposed rule, we proposed to include at § 486.302 the definition the “donation rate” as the number of donors as a percentage of the donor potential. In current regulations at § 486.318(d)(1), we define the donation rate as being the eligible donors as a percentage of the eligible deaths.
In addition, we proposed that § 486.318(d)(1) specifies that the donation rate is calculated as the number of donors in the DSA as a percentage of the donor potential.
Comment:
The majority of the comments received supported the use of the donation rate to measure OPO performance.
Response:
We appreciate the comments received. We consider the donation rate to be an important outcome measure because it assesses the ability of the OPO to obtain consent from the donor family, successfully manage the donor, procure and place at least one organ for transplantation (or pancreas for research), and ensure the safe transport of that organ for transplantation. However, despite all these aspects of the OPO's role that the donation rate measures, for patients waiting for a life-saving organ
transplant, the primary measure of interest is the organ transplantation rate. Therefore, the donation rate can be seen as augmenting the organ transplantation rate. By including the donation rate, we incentivize OPOs to pursue all donors, especially the single organ donors. An OPO is more likely to meet the donation rate measure if they also procure organs from donors after cardiac death (DCD) or marginal donors where relatively fewer organs may be transplantable.
Comment:
One commenter suggested CMS measure the OPOs ability to obtain consent in calculating the donation rate. The commenter did not suggest how the consent would be used as an outcome measure.
Response:
Although obtaining consent is one part of the donation process, we are not adopting the commenter's suggestion to use obtaining consent as an outcome measure. We recognize the critical role of obtaining consent as the first part of donation, and without it, the rest of the donation process cannot occur. Our regulation at § 486.342 requires OPOs to have a written protocol to ensure that, in the absence of a donor document, the individual(s) responsible for making the donation decision are informed of their options to donate organs or tissues (when the OPO is making a request for tissues) or to decline to donate. As with our other CfCs, we survey to ensure compliance with this requirement, and those surveys typically occur every 4 years. However, we cannot verify success in obtaining consent relative to the donor referrals through independent, objective data on an annual basis and instead, rely on surveys. It would be unduly burdensome to OPOs to be routinely surveyed every year for us to identify and verify this information for purposes of the frequent assessment periods in which these outcome measures are calculated to trigger revisions to the Quality Assessment and Performance Improvement (QAPI) plan under the requirements at § 486.348 (also discussed in section II.E of this final rule).
Final Rule Action:
We are finalizing as proposed the definition of donation rate at § 486.302, which is defined as the number of donors as a percentage of the donor potential. Furthermore, we are finalizing at § 486.318(d) that an OPO is evaluated by measuring the donation rate in their DSA and at § 486.318(d)(1) the donation rate is calculated as the number of donors in the DSA as a percentage of the donor potential.
3. Donor Definition (§ 486.302) and the “Zero Organ Donors”
In the December 2019 OPO Proposed rule, we proposed to revise the definition of “donor” at § 486.302 to mean a deceased individual from whom at least one vascularized organ (heart, liver, lung, kidney, pancreas, or intestine) is transplanted. An individual also would be considered a donor if only the pancreas is procured for research or islet cell transplantation. In general, the proposed definition of donor would change the current regulatory definition, requiring that at least one organ be transplanted, rather than being recovered for the purpose of transplantation, in order for the donor to be included in the donation rate. We also included donors who had pancreata procured for islet cell transplantation and research in the definition of donor to respond to the requirements of section 371(c) of the PHS Act.
Comment:
Several commenters stated that our proposed new definition of “donor” excluded “zero organ donors.” Some commenters had different definitions of “zero organ donors” including: (1) Donors who are taken to the operating room but cannot be a donor for one or more reasons; and (2) are donors in which the transplantable organs are turned down by transplant programs. These commenters claim that excluding “zero organ donors” in the donation rate would discourage OPOs from pursuing extended criteria or marginal and complex donors, which would be inconsistent with our goal of trying to increase donations, particularly of single organ donors.
Response:
We acknowledge that the general effect of our proposed definition of donor at § 486.302 would be that, a patient must donate at least one organ that is actually transplanted to qualify as a “donor.” We note that the definition also includes a patient who donates a pancreas for research. Although “zero organ donors” would not fall under this definition, we are not persuaded by comments that OPOs will not pursue the extended criteria or marginal, complex donors if we do not include “zero organ donors” in the outcome measures. Not only did we receive comments from some OPOs stating that they are committed to “pursuing every organ every time even if no organs are transplanted,” but an OPO that does not pursue these donors will be at risk of being identified as a poorer performer compared to other OPOs and could possibly face the prospect of being de-certified.
Evidence demonstrates that the top performing OPOs are pursuing extended criteria and single-organ donors, and those OPOs are also successfully placing the organs at programs that transplant them. Some OPOs are relatively successful in recovering organs from more marginal candidates, saving those donors from being “zero organ donors.” We accessed the OPTN database on August 12, 2020 and found that from 2018 and 2019, the OPO in Nevada had procured 80 kidneys that were categorized as having the highest Kidney Donor Profile Index (KDPI) scores of 86 through 100. These types of kidneys are primarily from extended criteria or marginal donors that are more likely to end up as “zero organ donors.” Meanwhile, the local kidney transplant programs in their DSA transplanted zero kidneys with the highest KDPI scores, meaning that these organs were transplanted by programs outside of their DSA; this example suggests that the local demand was not driving the Nevada OPO's performance. In order for other OPOs to follow this example, they must also pursue the extended criteria and marginal donors, even if the local transplant program does not accept them. Using the comparative performance methodology and holding all OPOs to the performance of these top performing OPOs, we intend to incentivize all OPOs to pursue extended criteria and marginal donors, even if they may become zero organ donors.
Comment:
Some commenters recognized that “zero organ donor” counts are self-reported data and are difficult to verify, but suggested that CMS review the charts or use triggers to lead to a chart audit as a means of verifying these donors.
Response:
In changing the definition of donor, we are adhering to the principles described in the December 2019 OPO proposed rule that the outcome measures be more transparent, reliable, and objective measures of OPO performance. Part of ensuring reliability is moving away from self-reported data as much as is feasible and using data that can be easily verified. It would require an extraordinary effort for CMS to verify the zero organ donors as frequently as needed to calculate the annual assessments of OPO outcome measures that will be used to trigger revisions of the QAPI program that can spur OPOs to improve their performance, and to rank OPOs for certification purposes.
Comment:
We received several comments stating that because it is ultimately the transplant programs that decide whether an organ is transplanted (not the OPO) that redefining “donor” to require that the organ be transplanted would not be appropriate.
Response:
Transplant programs decide whether an organ will be transplanted; however, it is the OPO's responsibility to ensure that placement and transport of organs happen in a fast and effective manner. If the OPO engages in best practices for placement, packaging, and transportation of organs, such as using RFID tags to track organs in transit and assure that they are not forgotten or diverted, there should not be significant differences in the frequency of zero organ donors among OPOs because the occurrence of unexpected anatomical issues which contraindicate donation that arise after consent is secured are random and not statistically significant in one DSA compared to another.
Comment:
Several commenters stated that OPOs are obligated to the allocation system and that sometimes they run out of time trying to place certain organs. Therefore, the commenters stated that the OPOs should not be punished if they cannot place a transplantable organ.
Response:
We respectfully disagree with the commenters' assertion. The OPTN final rule has a section on wastage that explicitly allows transplant programs to transplant an organ into any medically suitable candidate to do otherwise would result in the organ not being used for transplantation (42 CFR 121.7(f)). Therefore, we do not believe the constraints of the allocation system justify not successfully placing a transplantable organ. We believe that this final rule will allow OPOs the opportunity to improve the placement of organs, and drive the transplant community to adopt the technologies necessary to optimize placement.
Comment:
We received a comment that there are some events, such as loss of an organ during transport, which OPOs cannot control.
Response:
By requiring that the organ be actually transplanted (with the exception of pancreata procured for research) in order to be counted in the donation rate, we are supporting those OPOs that work to successfully complete the final step of the donation process. Unfortunately, we are aware of incidences where organs are lost or damaged during transport.
8
It is the responsibility of the OPO to arrange the appropriate transport to the transplant program. (See § 486.346 of this final rule.) Therefore, it is important that any measure of OPO performance not stop at the procurement of a transplantable organ, but also holds OPOs accountable for the subsequent steps for successful placement and transport of organs.
8
Kaiser Health News, “How Lifesaving Organs for Transplant Go Missing in Transit,” February 10, 2020
https://khn.org/news/how-lifesaving-organs-for-transplant-go-missing-in-transit/.
Comment:
One commenter suggested that donor families would be disappointed if they consented to the donation, but we did not allow the zero organ donor to be called a donor.
Response:
We appreciate every potential donor and are not discouraging OPOs from referring to “zero organ donors” as “organ donors” even if no organs are transplanted when speaking with families. The use of the term “organ donors” has different meanings in different settings. Many families consider their loved ones to be organ donors if they are eye and tissue donors or if the organs are donated for research. Therefore, we do not think our definition, used solely for assessing OPO performance for regulatory purposes, should affect the donor families' perception of the value of their loved one's donation or the terms used by OPOs or other organizations when liaising with families of potential donors.
Comment:
One commenter stated that it was important to encourage OPOs to pursue all donors and suggested that we include these “zero organ donors” in the performance measures even if they are not included in the outcome measures.
Response:
We will continue to work with OPOs on a more refined reporting process to capture information about zero organ donors and the reason for the organs not being retrieved or transplanted. We intend to continue the dialog with OPOs about the necessary data to collect and how to do so in a verifiable, burden neutral manner, and our CfCs may be revised in the future based on such modifications and further analysis of the data.
Comment:
One commenter supported our proposed definition of donor because they agreed that OPOs could “game the system” if we included “zero organ donors.”
Response:
We appreciate the comment and, as explained above, we are not including “zero organ donors” in the definition of “donor.” As we stated in the December 2019 OPO proposed rule, we have concerns with self-reported data. Our internal analysis of the OPTN data found a variation in the frequency of “zero organ donors” as defined as a donor in which an organ was procured for transplantation, but no organ was transplanted. We did not see an association between the OPO's performance and the percentage of these donors, however, we retain the concerns expressed in the December 2019 OPO proposed rule. The OPTN data show that in 2018, there were 1,255 organs procured from these zero organ donors, but never transplanted. In that same year 31,203 organs were transplanted. Among the top performing OPOs, zero organ donors represented 2.73 percent to 11.86 percent of donors (the range among all OPOs was 0 percent to 17.02 percent) with counts ranging from 3 to 59 zero organ donors.
We do not understand the significance of this variation, but will continue to examine the data about “zero organ donors” and assess whether we can appropriately capture and verify the data for future inclusion in our outcome measures.
Comment:
We received a comment raising concerns that the change in the definition of donor could affect reimbursement from Medicare since the previous definition allowed OPOs to be reimbursed for the efforts to procure transplantable organs.
Response:
Our revised definition of donor does not impact Medicare reimbursement for organ procurement costs. We did not propose to change our rules for reimbursing OPOs for organ procurement costs. Our payment policies for organ procurement costs do not rely on our definition of donor under § 486.302.
Final Rule Action:
We are finalizing as proposed in the December 2019 OPO proposed rule, the definition for donor at § 486.302 to mean a deceased individual from whom at least one vascularized organ (heart, liver, lung, kidney, pancreas, or intestine) is transplanted. An individual also would be considered a donor if only the pancreas is procured and is used for research or islet cell transplantation.
4. Organ Transplantation Rate (§ 486.302 and § 486.318(d))
For our second measure, we proposed to assess the OPO's organ transplantation rate, which is defined as the number of organs transplanted from donors in the DSA as a percentage of the donor potential.
Comment:
We received a number of comments supporting our organ transplantation measure because it was a volumetric measure (that is, reflects the volume of organs transplanted). We had one OPO commenter provide an example of how they increased the procurement of lungs for transplantation, but the SRTR method for measuring observed to expected performance in organ transplantation did not capture their improved performance.
Response:
We appreciate the comments in support of the volumetric organ transplantation measure. As
stated earlier, the organ transplantation rate is an important measure as it directly measures the benefit for patients from OPO performance.
Comment:
Several commenters opposed the organ transplantation rate because it was too similar to, and not independent of, the donation rate since it shared the same denominator as the donation rate.
Response:
In both of our outcome measures, the denominator represents a reasonable effort to estimate of the donor potential and other related factors for the DSA, as required by the OPO Certification Act of 2000.
9
The numerators measure OPO performance (through the number of donors and organs transplanted) and are somewhat correlated because if there are more donors, there are likely to be more organs transplanted. It is CMS' expectation that high-performing OPOs will likely perform well on both measures and low-performing OPOs will perform poorly on both measures. However, these numerators are not the same and each donor has a range of potential organs that could be transplanted. For example, OPOs that focus primarily on DCD and marginal donors may need to seek more donations in order to have sufficient organs transplanted to mathematically meet the organ transplantation rates. On the other hand, OPOs that are very effective at placing all possible organs from younger, healthier donors may achieve the targeted organ transplantation rate, but not the donation rate, if they choose not to pursue the marginal, complex and DCD donors with only one or two transplantable organs.
9
42 U.S.C. 273(b)(1)(D)(ii)(II).
Comment:
We received a number of comments from the OPO community recommending that we use the SRTR's Donor Yield model, which calculates an observed:expected (O:E) ratio for placing organs for transplantation (also called the SRTR O:E model). These commenters preferred the O:E measures because it includes 34 to 50 risk-adjustments, changes over time, and measures a different part of an OPO's performance from the donation rate (part that involves placement and transport an organ).
Response:
While the SRTR's O:E ratio may have value for understanding potential areas for improvement and may be used by the OPTN and OPOs for internal performance assessment, it is derived from the donation rate and is not capable of assessing the number of organ transplants. The SRTR O:E model “uses a series of complex statistical models” and relies on coefficients from a multinominal regression model.
10
The validity of the model is dependent upon frequent updates to the regression analyses to determine which predictors are in the model (hence range of 34 to 50 risk-adjustments). Because of the complexity of the model and the need for frequent updating, it is not feasible for us to continually update the methodology through notice and comment rulemaking, which is necessary in order to enforce the current version of the model. Use of the SRTR O:E model in regulation has not been practicable. The mathematical complexity of the risk-adjustments creates an opacity that is inconsistent with our goal of increasing transparency in our outcome measures.
10
https://www.srtr.org/about-the-data/guide-to-key-opo-metrics/opoguidearticles/donor-yield/.
Comment:
We received comments suggesting we add the SRTR's donor yield model, which measures observed to expected performance in O:E measure with the organ transplantation rate or increase the level of performance on the O:E measure.
Response:
We appreciate the commenters recognizing that the O:E measure is based on the average performance of an OPO and suggesting that we retain the measure but increase the level of performance above what was expected so that OPOs would be held to the O:E ratio of the top performing OPOs. As previously discussed, we are not using the O:E measure because it is not capable of measuring volume, is directly correlated to the donation rate, and does not directly capture increases in organs being transplanted. Finally, adding this measure to the organ transplantation rate would introduce additional regulatory complexity and reduce the transparency of these outcome measures. Therefore, we are finalizing the organ transplantation rate as the second measure.
Final Rule Action:
We are finalizing the definition of organ transplantation rate with revisions at § 486.302. The revised definition of “Organ transplantation rate” is the number of organs transplanted from donors in the DSA as a percentage of the donor potential. Organs transplanted into patients on the OPTN waiting list as part of research are included in the organ transplantation rate. This modification is a clarification that if the organ is transplanted, regardless whether it is part of normal clinical practice or research, it will be counted in the organ transplantation rate. We are also making further modifications to the definition of “organ transplantation rate,” which are discussed in more detail in section II.B.7 of this final rule. We are also finalizing as proposed at § 486.318(d) that the OPO is evaluated by measuring the organ transplantation rate in their DSA.
5. Organ Definition (§ 486.302)
In the December 2019 OPO proposed rule, we specified how we would count the organs that would constitute the numerator for the organ transplantation rate. We proposed to include pancreata procured for islet cell transplantation and research in the definition of “organ” to meet the requirements of the Pancreatic Islet Cell Transplantation Act of 2004, which amended the PHS Act to include section 371(c). (84 FR 70631) Section 371(c) of the PHS Act states that “[p]ancreata procured by an organ procurement organized and used for islet cell transplantation or research shall be counted for purposes of certification or recertification under subsection (b).”
Comment:
Several commenters opposed our inclusion of pancreata for research in our outcome measures since procuring pancreata for research is not a normal function of OPOs and is highly dependent upon the demands of the local researchers. Some commenters supported the inclusion of pancreata procured and placed for research in the organ count. We received comments that including the pancreata for research would lead to artificial inflation of the organ transplantation rate; that we should use a third performance metric to assess performance for pancreata procured for research; and that we did not properly define the scope of “pancreata procured for research.”
Response:
We agree with the commenters that pancreata for research are specific to the local research demands and may not reflect universal OPO practice. Nonetheless, their inclusion in the outcome measures is consistent with the requirements of the statute, and we are finalizing them as such. We intend to verify the accuracy of the data reported related to pancreata procured for research during the survey process, and believe that this is a sufficient disincentive for inflating the reported data. We considered creating a third outcome measure specifically for pancreata procured for research. However, there is no data source currently available to enable us to analyze performance and establish a meaningful measure. We will continue exploring ways to develop a data source and meaningful measure for consideration in future rulemaking.
Comment:
Commenters recommended CMS to include vascular composite allografts in the organ count for the organ transplantation rate.
Response:
We appreciate this suggestion but are not including vascular composite allografts (VCA) in our definition of organ. VCA transplantation is very localized and rarely performed. In 2019, approximately 15 such transplants occurred, the vast majority being the transplantation of a uterus (12 transplants)
11
. Inclusion of VCAs as organs would require a separate assessment throughout all CMS policies and regulations that is beyond the scope of this rule.
11
OPTN database accessed on July 11, 2020 and number of transplants for abdominal wall, head & neck (cranial facial), head & neck (scalp), GU: Penile, GU: Uterus, upper limb: Bilateral, upper limb: Unilateral, and VCA were counted for 2018 and 2019. In 2018, there were 11 transplants.
Final Rule Action:
After consideration of public comments, we are finalizing our proposed definition of organ at § 486.302 to mean a human kidney, liver, heart, lung, pancreas, or intestine (or multivisceral organs when transplanted at the same time as an intestine). The pancreas counts as an organ even if it is used for research or islet cell transplantation.
6. Donor Potential (§§ 486.302 and 486.318(d)(4))
Under § 486.318(d)(4), the donation rate, organ transplantation rate, and kidney transplantation rate use the “Donor potential” as defined under § 486.302 as the denominator. In our December 2019 OPO proposed rule, we proposed to define the donor potential (denominator) as total inpatient deaths in the DSA among patients 75 years of age or younger with any cause of death that is not an absolute contraindication to organ donation. We proposed to use death certificate information that can currently be obtained from the Center for Disease Control and Preventions' (CDC), National Center for Health Statistics' (NCHS's) Detailed Multiple Cause of Death (MCOD) file as described in our December 2019 OPO proposed rule. The MCOD is published annually and is publicly available upon request. The MCOD comprises county-level national mortality data that include a record for every death of a U.S. resident recorded in the U.S. The MCOD files contain an extensive set of variables derived from the death certificates which are standardized across the 57 jurisdictions that provide CDC with the data (50 states, New York City, the District of Columbia and the five territories). The jurisdictions use the U.S. Standard Certificate of Death as a template for their forms.
In order to provide a most robust and detailed discussion, we have divided the comments and responses to our definition of “Donor potential” into separate sections: The use of state death certificates for estimating the donor potential; the specific ICD-10-CM codes used to identify the donor potential; the age threshold for the deaths; the inpatient aspect of the deaths; and the effect of waiver hospitals.
a. Death That Is Consistent With Organ Donation (§ 486.302)
Under § 486.302, within our proposed definition of “Donor potential,” we use a separately defined term, “Death that is not an absolute contraindication to organ donation.” This term is characterized by two major parts: (1) The data source for calculating these deaths (state death certificate data) and (2) the ICD-10-CM codes for identifying these deaths.
We proposed to use state death certificate information that can currently be obtained from the CDC, NCHS's MCOD file as described in our December 2019 OPO proposed rule to determine the donor potential. The MCOD is published annually and is publicly available upon request. The second part of the definition of “Death that is not an absolute contraindication to organ donation” describes all deaths except those identified by the specific ICD-10-CM codes listed in our definition that would preclude donation under any circumstance. As part of our proposed rule, we also considered the alternative of using the ICD-10-CM codes that are consistent with organ donation in the methodology developed by Goldberg, et al.
12
(84 FR 70662), also knowns as the “CALC” methodology.
12
Goldberg D, Karp S, et al., “Importance of incorporating standardized, verifiable, objective metrics of organ procurement organization performance into discussions about organ allocation,”
AmJTransplant.
2019;00:1-6.
We received numerous comments on both of these components and discuss responses to these comments separately.
i. Death Certificate Data
Comment:
We received numerous and varied comments regarding our use of the death certificate information reported to the CDC and currently found in the MCOD files. Many commenters supported the use of data derived from death certificates because it represents the best available option for obtaining objective data at this time to estimate the donor potential. However, several commenters referenced literature that found error rates of the death certificates ranging from 30 to 60 percent. In addition, numerous commenters from the medical examiner/coroner community questioned the accuracy of the death certificates.
Response:
We appreciate the commenters supporting our use of the MCOD file. As discussed in the December 2019 OPO proposed rule, we are aware of the error rates in the death certificate data reported in the literature, but continue to believe this data is the most complete information that is readily and publicly available, that can be used for estimating the donor potential at this time. Every state submits death certificate data to the CDC and the elements collected in the death certificates are standardized to the greatest degree possible. Errors in reporting on the death certificates are primarily from user error, where the individual completing the form makes a mistake. The same user errors likely plague other potential data sources, such as hospital records, and those data sources would come with significant added reporting burdens with limited to no additional benefit. We are not aware of differences in the error rates that would disadvantage one DSA over another DSA (84 FR 70632). In addition, we are not aware of any research that describes such differences. Based on our understanding of which professionals are responsible for completing the death certificates and comments from the public, we do not see a compelling reason why there would be a consistent disparity in the error rates from one DSA to another. Furthermore, no commenters have suggested a source of empirical evidence that could be obtained by reasonable effort of organ donor potential in each designated service area sufficient to meet our needs and expectations. The peer-reviewed research developed by Goldberg, et al. discussed throughout our December 2019 OPO proposed rule and this final rule supports the use of the death certificate data as the best and most comprehensive source for estimating the donor potential at this time.
We appreciate the comments and knowledge from the coroner/medical examiner community about the error rates in the death certificates.
Comment:
We received one comment from the OPO in Alabama about errors in the electronic reporting of death certificate data resulting in misreporting inpatient deaths.
Response:
We thank the commenter for pointing out the error in reporting inpatient deaths in Alabama. We understand that the reporting error has been resolved for 2019 and was unique to Alabama. We do not have any reason
to believe that other states made this error. For purposes of the regulatory impact analysis in this final rule, which uses data from 2018, we have made adjustments to the inpatient deaths in Alabama to be more consistent with historical rates of inpatient deaths prior to the error occurring. If there are future occurrences in which there are similar such errors, we have added an extraordinary circumstances exception (ECE) under § 486.316(f) to allow OPOs to request a 1-year extension to their agreement cycle if there are extraordinary circumstances beyond the control of the OPO that would affect the data being used. This ECE request is discussed in greater detail in section II.C.5 of this final rule, which discusses the data length used for calculating the outcome measures.
Comment:
We received a comment describing in detail the process by which the death certificate is completed in their particular state.
Response:
We thank the commenter for providing us with this detailed information. States have slightly different processes for completing the death certificates even though all states use the standardized death certificates. The MCOD comprises county-level national mortality data that include a record for every death of a U.S. resident recorded in the U.S. The MCOD files contain an extensive set of variables derived from the death certificates which are standardized across the 57 jurisdictions that provide CDC with the data (50 states, New York City, the District of Columbia and the five territories). The jurisdictions use the U.S. Standard Certificate of Death as a template for their forms. Although commenters expressed concerns with our use of the MCOD, they did not suggest a different source of empirical evidence that could be obtained without undue reporting burden and was of greater accuracy. Furthermore, this commenter did not provide any information to suggest that this different process for their state would result in less accurate information for that jurisdiction and we do not have reason to believe that a different process would disadvantage one OPO compared to another.
Comment:
We received a comment questioning whether our donor potential reflected the DSA because the publicly available CDC data on death certificates has the location of the death based on the individual's home rather than the location of the hospital.
Response:
Based upon this comment, we believe that the commenter is referring to the CDC Wide-ranging Online Data for Epidemiologic Research (WONDER) database, which is available on the CDC public website. This database has the person's residence at the time of death instead of the location of the death. The MCOD file, which we are using to calculate our outcome measures, has information on the location (county) of the inpatient death. The location of the patient's death is more relevant to attributing donor potential for each DSA. A CMS contractor will use information from the MCOD file to attribute deaths to each DSA.
Comment:
We received a comment that “death certificate source is limited solely to statistical uses and cannot be used for regulatory purposes” because section 308(d) of the PHS Act (42 U.S.C. 242m) provides that data collected by NCHS “may be used only for the purpose of health statistical reporting and analysis.”
Response:
We have consulted with the CDC and concluded that our use of the MCOD represents a statistical analysis to characterize OPO performance and is consistent with the PHS Act. We are not using the data from the MCOD file to directly take legal, administrative or other actions against the hospitals and states that submit the data, nor are we taking regulatory action on the inpatient deaths in the DSA. Rather, we are using the data to “normalize” our outcomes of interest: The number of donors and the number of transplants in the DSA. The section of the PHS Act cited by the commenter refers to the confidentiality of the NCHS data and the limitations of the use of the data if “an establishment or person supplying the information or described in it is identifiable.” Our calculations use county level data that does not identify the specific hospitals submitting the death certificate data.
Comment:
We received comments that the death certificate data does not include information about co-morbidities or other chronic conditions that may make it unlikely for the person to be an organ donor.
Response:
Our goal in using the death certificate data was to use the best information available to calculate organ donor potential in each DSA. We are using the death certificate data to adjust the denominator to better reflect the population in the DSA that will more closely resemble individuals likely to become a deceased organ donor (individuals who are 75 and younger and died in the hospital with a cause of death consistent with organ donation). No risk-adjustment method is precise and we do not have evidence that the rate of co-morbidities associated with these causes of deaths is significantly different from one DSA to another and would be the reason for the differences in performance.
Comment:
Several commenters suggested alternative sources for estimating the donor potential: Data from electronic health records, data from hospital chart reviews, insurance billing codes, and hospital reported data of ventilated neurological deaths. Commenters also raised concerns about the burdens of asking donor hospitals to report potential donors and ventilated deaths, a concern that applies to all of the suggested alternatives.
Response:
We thank the commenters for the suggestions of alternative data sources. All of the suggested data sources are subject to the same user error inaccuracies as the MCOD files. Furthermore, we note that none of these suggested alternative sources for estimating organ donor potential could be obtained by reasonable efforts and would not be feasible or practical for calculating the outcome measures. Many of the suggested data sources are not feasible to use or sufficiently comprehensive to estimate the donor potential for various reasons. First, not all hospitals have electronic health records that can transmit data or be shared; not all OPOs have the ability to receive electronic health record transmissions. Second, collecting data via hospital chart reviews would likely be burdensome. Third, there is no national or comprehensive database of all insurance claims, and collecting data from insurance claims would inappropriately not count those decedents who did not have insurance. We agree with those comments that raised concerns about the burden on donor hospitals if we asked them to report data on ventilated deaths, and agree that requiring those additional reporting requirements or combining all these disparate data sources to estimate the donor potential could not be obtained by reasonable efforts. CMS will continue to evaluate the benefit and applicability of future data sources as they become available.
Final Rule Action:
We are finalizing the death certificate information reported to the CDC and currently found in the MCOD files as the data source for calculating the donor potential of each DSA.
ii. International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM)
Comment:
The vast majority of comments supporting the use of the state death certificate data from the CDC files also preferred using ICD-10-CM codes that represented the causes of
death that is consistent with organ donation rather than our proposed approach based on defining “death that is not an absolute contraindication to organ donation.” Some commenters suggested adding other ICD-10-CM codes to the list of ICD-10-CM codes we would exclude. One commenter stated that the ICD-10-CM codes consistent with organ donation provided a donor potential that was consistent with their own internal calculations. Another commenter provided an estimate of 187,500 donor referrals in the U.S. based on extrapolation of their own data.
Response:
Given the overwhelming comments supporting the use of the ICD-10-CM codes from the methodology which is based on the cause, age, and location consistent with organ donation (CALC), we are finalizing the use of ICD-10-CM codes from the CALC methodology, which are inclusion codes, in estimating the donor potential. Our proposed methodology used the ICD-10-CM codes that are exclusion criteria and included significantly more codes. The ICD-10-CM codes that are considered as causes consistent with organ donation were identified in section V.G “Alternatives Considered” of the December 2019 OPO proposed rule, and were confirmed by the developers of the CALC methodology, although they were not specified in the published literature. As discussed in the December 2019 OPO proposed rule, the ICD-10-CM codes in the CALC methodology captures 98-99 percent of all donors. (84 FR 70666). The advantage of this inclusion method over the one we proposed is that given the description of the ICD-10-CM codes chosen, there are unlikely to be new causes of death that would lead to organ donation. However, as we have discovered during the COVID-19 public health emergency (PHE), there is a likelihood of new, unanticipated contraindications for organ donation. If we used exclusion criteria in estimating the donor potential, we could have to update and change our rules much more frequently to adjust for these new contraindications to organ donation. We believe that these unplanned changes could be disruptive to OPO operations and efficiency. Therefore, we agree with commenters that the CALC methodology which identifies ICD-10-CM codes consistent with organ donation is preferable to the methodology we proposed. We discuss this methodology and the calculations that result from using this methodology in greater detail in the discussions of our regulatory impact analysis under section V of this final rule.
In § 486.302, we have added the definition “Death that is consistent with organ donation” as all deaths from the state death certificates with the primary cause of death listed as the ICD-10-CM codes I20-I25 (ischemic heart disease); I60-I69 (cerebrovascular disease); V-1-Y89 (external causes of death): Blunt trauma, gunshot wounds, drug overdose, suicide, drowning, and asphyxiation. From our calculations using 2017 data, the CALC methodology resulted in a donor potential of 101,479, which would be a reasonable estimate of the total U.S. donor potential if the donor referral is approximately 187,500 as suggested by a commenter who estimated this donor referral population based on their own referral data.
We also conducted preliminary analyses examining whether there was additive value to excluding the ICD-10-CM codes that were contraindications to organ donation to the causes of death that is consistent with organ donation. We found little difference in the ranking and identification of OPOs at the different thresholds of interest. Therefore, we are not using any ICD-10-CM codes to exclude additional inpatient deaths from the ICD-10-CM codes that are consistent with organ donation.
Comment:
We received several comments questioning the donor potential and providing references that cited donor potential varying as low as 10,500 and as high as 272,000 (our estimate).
Response:
We know the donor potential cannot be as low as 10,500 because there were 11,870 deceased donors in 2019, according to the OPTN (
https://optn.transplant.hrsa.gov/news/organ-donation-again-sets-record-in-2019/
). Our “donor potential” was not designed to identify an actual donor potential size as we have discovered that the true donor potential changes constantly as technology and demand for organ transplantation changes. Instead, our proposed methodology was designed to estimate the likely donor referral population to normalize the inpatient deaths across the different DSAs. Since the donor potential is part of a rate calculation, identifying the exact, true donor potential is less relevant than providing standardized, reasonable, and objective criteria to estimate it. We know that as public health crises occur, such as the opioid crisis or COVID-19, the donor potential may change. Also, as technology and practices change, the donor potential may change. When the 2006 OPO final rule was published, DCD was so infrequent that those potential donors were not included in the definition of an eligible death; yet in 2019, almost 23 percent of all deceased donors were DCD donors. Based on public comments, we believe the CALC methodology produces a very close estimate to the current donor potential for each DSA and it also has the flexibility to adjust to changes in the number of these causes of death in the DSA.
Final Rule Action:
We are finalizing under § 486.302 that “Death that is consistent with organ donation” means all deaths from state death certificates with the primary cause of death listed as the ICD-10-CM codes I20-I25 (ischemic heart disease); I60-I69 (cerebrovascular disease); V-1-Y89 (external causes of death): Blunt trauma, gunshot wounds, drug overdose, suicide, drowning, and asphyxiation. We will not include in the final rule a definition of “death that is not an absolute contraindication to organ donation.”
b. Age 75 and Younger
Comment:
We received comments that the proposed age cut-off of age 75 in our definition of “donor potential” was too high and suggested that we should use age 65 instead. On the other hand, we also received a comment that the proposed age cut-off of age 75 was too low because OPOs have procured livers from donors older than 75.
Response:
We proposed that the denominator for calculating the donation and organ transplant rates will be based on the number of inpatient deaths of someone 75 years old or younger because our previous definition of eligible death uses the age of 75 years old or younger. We do not concur with commenters' suggestion to lower the age threshold for the donor potential for our new outcome measures. Although we are aware that it is possible to for liver donors to be older than 75 years of age, we also recognize that the practice of using organs from older donors is still relatively new. Data from the OPTN lists the maximum age of liver donors as 65+. The number of living donations from this group between 2014 and 2019 ranged from 571 to 732 with gradual increase over time.
13
It is, however, a practice we want to encourage in order to increase the number of successful transplants; therefore, we are keeping our age limit at 75 years in order to reward OPOs who are successful with the donation and transplantation of organs from deceased individuals
greater than 75. OPOs who are successful in procuring these organs, particularly livers, from older donors may be able to count the donors and organs transplanted in the numerator of our outcome measures without having the death counted in the denominator.
13
https://optn.transplant.hrsa.gov/data/view-data-reports/national-data/
.
Final Rule Action:
We are finalizing that the age cutoff of age 75 for the donor potential definition in § 486.302, as proposed without modification.
c. Inpatient Deaths
We did not receive any comments as to whether the deaths should be limited to inpatient deaths. We are aware of preliminary studies suggesting that potential donors are identified in the emergency department as well as the inpatient setting. However, we believe those individuals are likely to become inpatients and thus, the location where they are identified, may not always correlate with where they die. Our data source is based on the location of death.
Final Rule Action:
We are finalizing that the definition of donor potential under § 486.302 be limited to inpatient deaths.
d. Waiver Hospitals
Comment:
We received a number of comments inquiring how CMS is addressing the donor potential estimates in DSAs where some donor hospitals sought waivers to work with a different OPO. One commenter raised concerns that we made an error in calculating the donor potential because we assigned the donor potential to the wrong OPO.
Response:
Historically, DSAs have been divided based on counties and metropolitan statistical areas. However, donor hospitals can request the ability to work with an OPO outside their DSA through a waiver request (we refer to these donor hospitals as “waiver hospitals” under § 486.308(e)). In estimating the donor potential for each DSA, we relied on the listing of counties found in the SRTR's OPO-specific reports, which listed both counties to an OPO when more than one OPO was servicing the county because of the waiver hospital. As a result, we erroneously double-counted the donor potential in several DSAs in the December 2019 OPO proposed rule. This inaccurate ranking would not have significantly altered our projections of the number of OPOs that would be automatically certified or decertified in accordance with the measure parameters set forth in the proposed rule.
While there are no data sources which we can use to precisely attribute non-Medicare inpatient deaths to these waiver hospitals, we can apportion the donor potential to each OPO by calculating the percentage of Medicare inpatient deaths at each acute care and critical care hospital in the county as a proxy, and use that percentage to divide the donor potential and assign the percentage of the donor potential based on the Medicare percentage of inpatient deaths. At this time, the apportionment method we have described is the best solution to addressing donor potential for OPOs that work with waiver hospitals. We intend to explore other options that could improve the data about deaths that should be counted for waiver hospitals.
Final Rule Action:
In response to public comments, we are amending the definition of the donor potential at § 486.302 to apportion the donor potential in a county where there is a donor hospital that has received a waiver to work with an OPO out of their DSA. For OPOs servicing a hospital with a waiver under § 486.308(e), the donor potential of the county for that hospital will be adjusted using the proportion of Medicare beneficiary inpatient deaths in the hospital compared with the total Medicare beneficiary inpatient deaths in the county.
7. Risk-Adjustments §§ 486.302 and 486.318(d)(2)
In the December 2019 OPO proposed rule (84 FR 70628), we did not propose other risk-adjustments to the proposed outcome measures, but sought comments as to the accuracy of our assessment and whether additional risk adjustments were necessary. We sought comments on whether risk-adjustments are necessary and which ones, such as donor demographic characteristics (race, gender, age, disease condition) or DSA characteristics (number of ICU beds or level I and II trauma centers) would be significant and clinically appropriate in the context of our proposed approach to identifying OPOs in need of improved performance. Specifically, we requested public comments that provide evidence-based support, such as peer-reviewed literature, that would support those suggestions, as well as data sources that would be necessary to calculate the risk-adjustments recommended.
a. Chronic Diseases
Comment:
We received comments from some OPOs about the incidence of certain diseases in their DSA that would make their general population less likely to be organ donors or have more organs available for transplantation. We received comments describing the different incidences of diseases in the different parts of the country.
Response:
We appreciate the comments about the different incidences of disease in the different geographic areas and recognize that different DSAs may have different population characteristics. However, these differences are population-based differences, and we did not receive any data that these differences were reflected in the donor potential, resulting in a disadvantage to one OPO compared with other OPOs. As part of our proposed rule, we analyzed whether there was a correlation between the performance of the OPO and the number of patients on the waiting list in the DSA (84 FR 70633). We conducted the analysis to determine whether “demand” in the form of the number of patients on the waiting list for the transplant centers within the OPO's DSA, is correlated with performance. We did not find any correlation. We reviewed the original analysis to determine whether there was a negative correlation between the waiting list and OPO performance (that is, OPOs flagged were more likely to have a sicker population in its DSA). Here, we treated the waiting list as a surrogate for the magnitude of end-stage organ failure in the DSA. Again, there was no correlation between OPO performance and end-stage organ failure in the DSA. As discussed earlier, we had compared using just the CALC versus the CALC plus our exclusionary criteria. There was no additive value to removing these contraindications to organ donation.
Comment:
One commenter suggested that we use data from the U.S. Renal Data System (USRDS) to risk-adjust for chronic kidney disease because people with chronic kidney disease are less likely to be organ donors.
Response:
Although we examined the USRDS data, we did not consider using it to risk-adjust for chronic kidney disease because it is population data and does not necessarily reflect the donor potential. Furthermore, the USRDS data does not delineate the different levels of chronic kidney disease. People with early stage chronic kidney disease can donate extra-renal organs for transplantation as well as the kidneys.
b. Race
Comment:
We received several comments from OPOs describing the racial/ethnic characteristics of people in their DSA and claiming that if we risk-adjusted for race, their performance would be improved because they serve a smaller percentage of white people than the national average. We also received comments opposing risk-adjustments based on race because of
concern that these risk-adjustments would mask past poor performance in adopting practices that are responsive to the racial/ethnic composition of the DSA served.
Response:
As we stated in our December 2019 OPO proposed rule, we decided not to risk-adjust for race because of concerns that it reflects historically poorer performance with certain racial/ethnic populations, and that studies suggest that OPOs can adopt policies and practices responsive to the community they serve and have better results. When we assess OPO performance, as seen in Table 3 in our regulatory impact analysis of this final rule, we see a diversity in the population served by the highest performing OPOs. We also see poor performance among OPOs servicing predominantly white populations. We agree with commenters who raise concerns that risk-adjusting for race could mask poorer performance, and we have concerns that racial risk-adjustments could perpetuate the stereotypes of different racial/ethnic groups and their willingness and ability to be organ donors.
We have reviewed the analysis conducted by the SRTR implying that racial risk-adjustment would ensure that a “correct” decision is made when comparing OPO measures.
14
We do not find these analyses compelling since the risk-adjustments reflect the biases and shortcomings of current OPO organ procurement practices, and we are not aware of a biological reason why race, as an independent factor, would affect the decision to be an organ donor or the number of organs transplanted. We agree with public comments and other literature opposing risk-adjustments for race.
15
We believe OPOs should be adjusting their practices to meet the characteristics of the DSA. Based on the diverse populations serviced by the top performing OPOs, we believe that racial characteristics of the DSA should not be a reason for risk-adjusting OPO performance.
14
Snyder JS, Wey A, et al., “The Centers for Medicare and Medicaid Services' proposed metrics for recertification of organ procurement organizations: Evaluation by the Scientific Registry of Transplant Recipients,” Am J Transplant, 11 Mar 2020.
15
Goldberg D, Doby B, Siminoff L, et al., “Rejecting bias: The case against race adjustment for OPO performance in communities of color,” Am J Transplant, 2020;00:1-6.
Although one of our previous outcome measure (the O:E measure) includes multiple risk-adjustments, such as for race, we are not including racial risk-adjustments in our final rule. The literature since 2005 (described in the December 2019 OPO rule), the public comments we received, and our examination of the demographics of the top performing OPOs, suggest that these factors, while they potentially pose hurdles for each OPO in their DSA, they are insufficient justification for additional risk adjustment. Therefore, we expect all OPOs to adjust their practices to overcome these hurdles and best service the populations within their respective DSAs.
c. Gender and Age
Comment:
We received comments that we should risk-adjust for identifiable variables in the donor potential data such as gender and age.
Response:
We do not know of a biological basis for why gender would be an independent factor in predicting the likelihood for being an organ donor or the number of organs transplanted.
We do, however, agree that there is biological basis for age to predict the likelihood of being a donor and the number of organs transplanted from the donor potential. Our internal analysis found statistically significant differences in the average age of the donor potential when we ranked OPOs based on their outcome measures, suggesting that age has an effect the number of donors and organ that are transplanted.
Since we are already including the age cut-off of 75 years and younger in our donor potential, we do not intend to further risk-adjust the donation rate for age. It is possible that the differences we see in performance based on the average age of the donor potential reflects OPO biases against older potential donors. Further risk-adjustments could mask these biases. Based on our methodology, in the DSAs where the population is older, OPOs have the opportunity to perform better because they have more opportunities for a donor who is older than 75 years of age—and these donors count in the numerator, but not in the denominator.
For the organ transplantation rate, there is no current risk adjustments for the average age of the donor potential. Our own internal analysis found that the average age of the donor potential correlated with performance on the organ transplantation rate, we will risk-adjust the organ transplantation rate based on the average age of the donor potential using the following method, provided here for full transparency and to allow others to replicate our methodology and calculations:
1. The age groups used for the adjusted transplantation rates are: <1, 1-5, 6-11, 12-17, 18-24, 25-29, 30-34, 35-39, 40-44, 45-49, 50-54, 55-59, 60-64, 65-69, 70-75.
2. Calculate a national age-specific transplantation rate for each age group. An expected transplantation rate for each OPO is calculated as ∑(g=1) Gdg*Rg/∑gdg,
16
where dg is the number of potential donors in the OPO in age group g, Rg is the age-specific national transplantation rate in age group g, and ∑gdg is the OPO's total number individuals in the donor potential. This can be interpreted as the overall expected transplantation rate for an OPO if each of its age-specific transplantation rates were equal to the national age-specific.
16
∑ is a mathematical symbol indicating summation.
3. Calculate the age-adjusted organ transplantation rate as (O/E)*P, where O is the OPO's observed unadjusted transplantation rate, E is the expected transplantation rate calculated in Step 2, and P is the unadjusted national transplantation rate.
d. Ventilator Status
Comment:
We received comments stating that there were differences in ventilators in ICUs based on geography, and that including ventilator status would be important in deriving the donor potential.
Response:
While there are differences in ventilators in ICUs based on geography, we do not have evidence that additional information about ventilator use would improve the CALC methodology. Since publication of our proposed measures, there has been a published study confirming our analysis that additional adjustments on cancers, sepsis and ventilator status to the CALC measure does not alter the ranking of OPO performance.
17
17
Goldberg DS, Doby B, Lynch R, “Addressing Critiques of the Proposed CMS Metric of Organ Procurement Organ Performance: More Data Isn't Better,” Transplantation; 2019 Nov 26.
Final Rule Action:
We are finalizing with modification the definition of “organ transplantation rate” at § 486.302 to be risk-adjusted for the average age of the donor potential using the following methodology:
(1) The age groups used for the adjusted transplantation rates are: <1, 1-5, 6-11, 12-17, 18-24, 25-29, 30-34, 35-39, 40-44, 45-49, 50-54, 55-59, 60-64, 65-69, 70-75.
(2) Calculate a national age-specific transplantation rate for each age group. An expected transplantation rate for each OPO is calculated as ∑(g=1)Gdg*Rg/∑gdg, where dg is the number of potential donors in the OPO in age group g, Rg is the age-specific national transplantation rate in age
group g, and ∑gdg is the OPO's total number of individuals in the donor potential. This can be interpreted as the overall expected transplantation rate for an OPO if each of its age-specific transplantation rates were equal to the national age-specific.
(3) Calculate the age-adjusted organ transplantation rate as (O/E)*P, where O is the OPO's observed unadjusted transplantation rate, E is the expected transplantation rate calculated in Step 2, and P is the unadjusted national transplantation rate.
We are also finalizing at § 486.318(d)(2) that the organ transplantation rate is calculated as the number of organs transplanted from donors in the DSA as a percentage of the donor potential. The organ transplantation rate is adjusted for the average age of the donor potential.
8. OPO Performance on Outcome Measures § 486.318(e) and § 486.302
In our December 2019 OPO proposed rule, we proposed to use our outcome measures in the context of a comparative donation rate and organ transplantation rate relative to the highest-performing OPOs. Our proposed definition of success would have been based on how OPOs perform on the outcome measures of donation rate and organ transplantation rate compared with the top 25 percent of donation and transplantation rates in DSAs with the goal of driving all OPO performances to cluster with the top performing OPOs. We proposed that OPOs would be assessed annually on these outcome measures and those whose outcome measures were below the top 25 percent would need to revise their QAPI to improve their performance. In the final year of the re-certification cycle, we proposed that OPOs whose outcome measures were below the top 25 percent will have failed their outcome measures for purposes of re-certification. We proposed to generate a one-tailed confidence interval for rates in each DSA to determine whether the OPO's outcome measures were statistically significantly the same or above the threshold rate of the top 25 percent. The top 25 percent rate would be generated using the rates established in the prior assessment period.
It is important to note that the outcome measures requirement does not require an OPO's performance be at or above the lowest rate for the top 25 percent of all of the OPOs. By determining confidence intervals, there is a range of values and the OPOs must not be statistically significantly difference from that range of values. For example, there are currently 58 OPOs. For the 58 current OPOs, twenty-five percent would be 15 OPOs (58 × .25 = 14.5). However, as discussed below, based on 2018 data, we estimate that 24 OPOs would meeting the criteria in § 486.318 to be designated as a Tier 1 top performing OPO.
We solicited public comments on whether or not comparing OPO performance should be based solely on the performance of the top 25 percent of OPOs within these two outcome measures, whether a different percentile or calculation of OPO performance should be used, or whether additional outcome, structure, or process criteria could be used to inform stakeholders of OPO performance over time (84 FR 70634).
The comments and responses are below.
Comment:
We received a diversity of comments in response to our proposed approach of establishing a threshold rate at the top 25 percent performance for OPOs to achieve. Some commenters supported our aggressive threshold rate of performance to drive an increase and improvement in OPO performance.
Response:
We thank the commenters who support our aggressive threshold rates. We intend to finalize, as proposed at § 486.318(e)(4) that OPOs whose donation rate and organ transplantation rate in the DSA is statistically significantly at or above the top 25 percent threshold rate will be considered have met the outcome measures for re-certification and their top performance will be recognized with a Tier 1 assignment. As a Tier 1 OPO, they will be rewarded by not being required to revise their QAPI to improve their performance in the outcome measures and their DSAs will not be opened for competition at the end of the re-certification cycle as long as they meet the other Conditions for Coverage during the re-certification survey.
Comment:
We received comments suggesting alternative threshold rates such as 50 percent or a tiered approach to ranking OPOs with different changes that must occur based on where the OPO falls in the tier system.
Response:
We thank the commenters for these suggestions. As we stated in the discussions of our alternatives in the December 2019 OPO proposed rule, we considered using a threshold rate based on the median or the geometric mean, but were concerned that this lower threshold rate would not incentivize OPOs to be higher performing. Furthermore, we ran the risk of top performing OPOs not being sufficiently incentivized to maintain their current performance level if we did not use an aggressive rate.
However, we also recognized that our aggressive threshold rate could result in too many OPOs being de-certified, particularly in the first re-certification cycle, without enough OPOs with organizational capacity and interest to assume responsibility for those open DSAs. We also recognize that if we set a threshold rate too difficult to attain, we risked incentivizing poorer performing OPOs to not strive to improve while remaining certified for a full 4-year cycle. Therefore, we are modifying our proposal and finalizing a 3-tier system based on public comments whereby OPOs are stratified into different tiers based on their outcome measures. The consequences of being in each tier differ based on whether the performance occurs as part of the annual assessment or if it occurs during the final assessment period. OPOs in Tier 1 are the OPOs that would have reached the goal performance of the top 25 percent threshold rates. We consider these OPOs to be top performing Tier 1. Based on data from 2018, we estimate that 24 OPOs would be in Tier 1.
The next tier will be identified as Tier 2 and will include OPOs in which both measures, donation rate and organ transplantation rate, have reached the median threshold rate or above (but are not in Tier 1). We estimate that there are 12 OPOs that would fall into Tier 2 based on 2018 data. Tier 2 OPOs will be considered to have met the outcome measures under § 486.318, and would not be decertified, but these OPOs will not be automatically re-certified. Since they have not reached the outcome measure requirements for Tier 1 status, their DSAs will be opened for competition and they will have to compete to retain their DSAs. Greater details about the competition process are discussed in section II.C of this final rule.
And finally, the lowest tier will be identified as Tier 3 and will include OPOs who have one or both outcome measures that are statistically significantly below the median threshold rates. We estimate that there are 22 OPOs who fall into Tier 3 based on 2018 data. Tier 3 OPOs will be considered as failing the outcome measures and will be de-certified. Greater details about the competition process are discussed in section II.C of this final rule.
This 3-tier system was designed based on a combination of comments that we use the 50 percent threshold rate instead of the top 25 percent threshold rate and the comments to use a tier system with varying consequences to OPOs based on the tier they were in. Instead of using a 50 percent rate or a mean rate, we chose
the median rate because both the top 25 percent threshold rate and the median rate represent the actual rates performed by one or two OPOs (when there is an even number, the median is calculated by averaging the two rates in the median). The mean rate, on the other hand, is a mathematical rate that may not reflect the performance of an actual OPO. A median, however, is not affected by extremes in performance. By identifying a specific rate of an OPO, other OPOs can directly compare their performance with another OPO.
Our goal in creating these tiers is to reward the top performing OPOs (Tier 1), while giving OPOs in Tiers 2 and 3 sufficient incentives to improve their performance and achieve ranking in the next level up and give Tier 2 OPOs the opportunity to demonstrate that they deserve to retain their DSA. These rewards and incentives are described in greater detail in this section and in our discussion about competition in section II.C of the final rule and our regulatory impact analysis (RIA).
Comments:
We also received comments that the 25 percent threshold rate was too aggressive and too many OPOs would be de-certified (discussed in detail at section II.C of this final rule), resulting in chaos in the system. Some commenters suggested that if we were to use such an aggressive threshold rate, we should not automatically de-certify OPOs who did not meet the threshold rates. Instead, we should consider a systems improvement agreement (SIA) similar to the ones for transplant program or the substantial changes they have made as part of their QAPI to avoid the disruption from de-certifying the OPO. In contrast, we received a comment that despite our aggressive threshold rate for performance, we should implement outcome measures that continually drive all OPOs to improve their performance.
Response:
We agree with some of the comments relating to a tiered approach. OPOs are not automatically decertified the first time that they do not meet the threshold rates. The performance of each OPOs will be assessed annually, this information will be provided to each OPO, and each OPO will then have an opportunity to improve and receive information about its performance following those improvements. Our annual outcome assessment is designed to inform the OPO regularly about their performance. Therefore, OPOs identified as being lower performing at the final assessment period of the agreement cycle would have a history of working with CMS to improve performance, as they would have been provided with their own performance information and making adjustments to their QAPI to improve their performance in the previous assessment periods. We expect to provide notice to OPOs of their performance and make the results public within 15 months of the end of each assessment period. For instance, performance on data from 2020 will be provided to OPOs and made public by the end of the first quarter in 2022. This period is necessary to accommodate the timeframe for the CDC to process the data and make the MCOD available for public use as well as for CMS calculate the performance measures. Additionally, during this timeframe, CMS will share preliminary results with each OPO to provide the opportunity to review the information and raise any concerns prior to the results being made publicly available and taking any enforcement action. This preliminary review is consistent with past performance updates and while this is an informal process, it does afford each OPO the opportunity to address concerns regarding its results. We acknowledge the time lag in making this information available, however, all data sources have inherent delays in making their information available to the public. Additionally, OPOs should not be relying on any single source of information to conduct self-assessments of their performance and should be employing a variety of information as part of a comprehensive QAPI program for this purpose.
We are not adopting public comments suggesting that poorly performing OPOs should be permitted to continue under an SIA. Allowing poorer performing OPOs the opportunity to continue servicing the DSA through a SIA would not benefit patients if there is a better performing OPO willing and able to service the DSA and provide patients with a higher standard of service.
However, we recognize that there are some OPOs that fall below Tier 1 but have made substantial changes designed to improve performance and have started to improve their performance. It may not be in our or patients' best interest to de-certify those OPOs, unless there is a better OPO prepared and capable of taking over the DSA. Thus, we created Tier 2 in response to the comments that we should lower our threshold rate for performance, and should not automatically de-certify OPOs who cannot reach Tier 1.
The commenter who suggested the tier system proposed that we undertake certain administration actions (like require change in leadership) based on the OPO's tier. While we appreciate the suggestions, we do not believe that there is a one-size fits all approach for all the OPOs in Tier 2, or that the federal government should dictate the specific steps needed to increase the rates in a particular DSA. Based on our assessment of outcome measures for these OPOs in Table 3, the range of performance is quite varied, with some OPOs very close to Tier 1 and others at the bottom of Tier 2. We are reluctant to follow the comments suggesting that OPOs be given an opportunity to continue as the designated OPO for another cycle subject to an SIA. That suggestion assumes that all OPOs in Tier 2 are capable of improving their performance and that they just need more time to implement best practices and improvements. However, because all OPOs receive interim reports on performance levels, we do not agree that this is always the case. Moreover, we recognize that patients in the DSA well-served by a marginal OPO that is allowed to continue without facing competition from a high performing OPO. Requiring that OPOs in Tier 2 to engage in a competitive process with other interested OPOs, on the other hand, would incentivize continual improvement to the benefit of patients.
Section 1138(b) of the Act and section 371 of the PHS Act required that the Secretary establish performance and outcome measures to be able to evaluate an OPO's performance prior to recertification. Requiring that Tier 2 OPOs compete for their DSA incentivizes best practices and optimizes organ donation and transplant rates. As already discussed and proposed, OPOs whose rates in the DSA fall under Tier 1 are considered to have met the outcome measure requirements and their DSA is protected from competition. OPOs identified as being in Tier 3 are considered to have failed the outcome measures under § 486.318 and will be de-certified, and following any administrative appeals, their DSAs will be open to competition.
Instead of automatically de-certifying OPOs in Tier 2 (those who have a statistically significant donation and organ transplantation rate at or better than the median rate) or implementing an SIA, we will allow these OPOs to compete to retain their DSAs by opening their DSA for competition to all OPOs that have been identified as being in Tier 1 and 2. In summary, all the DSAs for OPOs identified in Tiers 2 and 3 will be open for competition as proposed in our December 2019 OPO proposed rule and all the OPOs who are identified in Tier 1 and 2 will be able to compete for an open DSA. Broadening the number of DSAs open to competition and the number of OPOs eligible to compete will result in greater improvements
among all OPOs. OPOs in Tier 1 will need to maintain or improve their performance if they want to successfully compete to take over a new DSA, and OPOs in Tier 2 will need to improve their performance to retain their DSA or takeover another open DSA. Since OPOs identified under Tier 2 would have been de-certified under our original proposed methodology, this new approach will give mid-performing OPOs, who otherwise would have been de-certified, the opportunity to demonstrate, through the competition process, that they have implemented the requisite changes to progress to becoming a Tier 1 OPO.
Because OPOs have a 4-year exclusive agreement for each DSA with each re-certification cycle (see § 486.308(a)), it is critical that we select the most capable OPO that we can find to service the area, rather than automatically re-certify the incumbent OPO in Tier 2 or trying to “fix” an OPO that has not been able to reach the same performance as the top performing OPO through an SIA. We believe a competition process whereby all OPOs have sufficient incentives to continue to improve will drive all OPOs to cluster near the top.
Comments:
We also received many comments suggesting we use a standard deviation from the mean because it was statistically superior.
Response:
We disagree with the comments that the standard deviation from the mean methodology is statistically superior for our purposes of calculating OPO performance measures.
Under our methodology, all OPOs have the opportunity to cluster at the top because we generate confidence intervals for their donation and organ transplantation rates. The threshold rate is based on the previous year's rate and represents a specific rate to achieve or exceed. If all the remaining OPOs (below the top 25 percent threshold rate) had rates close to the threshold rate, their confidence interval could have all of them equal or exceed the threshold rate, resulting in clustering near the top. In Table 3, we show that 24 of 58 OPOs meet the top 25 percent threshold rate and this is 41 percent of all OPOs.
The standard deviation from the mean method, on the other hand, generates a list of OPOs that are a certain distance from the mean. As we discussed earlier, the mean is problematic because several lower performing OPOs could skew the calculated mean. The mean and the standard deviations are also generated contemporaneously with the ranking of the OPOs, giving OPOs no notice of their targeted performance. And, by nature of the statistical method of standard deviation, there will always be an OPO below the targeted standard deviation from the mean, meaning that not all OPOs would have the opportunity to be a top performing OPO unless they all had identical rates.
Comments:
We received comments implying that our goal was to reduce the number of OPOs and our methodology would result in an ever increasing threshold rate and ever-shrinking number of OPOs after each cycle.
Response:
We understand the concerns expressed by these comments, and want to reassure the public that our goal is to improve oversight of OPOs by reducing the variability in performance among OPOs and in the DSAs, not necessarily reducing the number of OPOs or forcing consolidation. Our methodology allows all OPOs the opportunity to perform as well as the top OPOs. We have proposed to change the outcomes measures because we agree with the public comments that the current OPO outcome measures are not sufficiently objective and transparent to ensure public trust in assessing OPO performance, nor do they properly incentivize the adoption of best practices and optimization of donation and organ transplantation rates.
Our methodology may result in increasing the threshold rate without shrinking the number of OPOs or DSAs significantly. Our internal analysis reveals demonstrated improvements in OPO performance from 2017 to 2018 and we anticipate OPO performance will continue to improve when incentivized by more transparent and accountable measures provided under this final rule. But, we recognize that there may be a rate at which OPOs cannot improve anymore and rates may cluster at the top. However, we intend to incentivize increases in the threshold rates for the top 25 percent and median as it would indicate that OPOs are procuring more organs for transplantation. Our methodology does not presume or require an increase in the threshold rates, and accounts for the performance of OPOs under similar circumstances or extraordinary circumstances.
In order for there to be an “ever-increasing threshold rate and ever-shrinking number of OPOs,” the commenter assumed that we would require that DSAs merge when an OPO takes it over. Our methodology for assessing OPO performance is based on the outcome measures for the OPO in each DSA. In our December 2019 OPO proposed rule (84 FR 70636), we stated that our regulations do not require that DSAs merge when a new OPO takes over. It would be our preference to not merge DSAs so that we can properly assess whether the new OPO is improving performance in each DSA since merging DSAs would result in merging the data on performance. Since DSAs are not required to merge, one OPO could run several DSAs. If an OPO with multiple DSAs cannot reach the outcome measures to be re-certified for one DSA, they will be de-certified for that DSA, but could be re-certified for other DSAs (assuming their performance supports it). Using our estimates from 2018 data, the result after conclusion of the first certification cycle that implements the new measures (2022-2026) could be approximately 36 OPOs servicing 58 DSAs with reductions in OPOs but not in DSAs. With 58 DSAs being served by top performing OPOs each cycle, we would expect the threshold rate to increase until all DSAs have donation and organ transplantation rates that cluster near the top. Even if consolidation were to occur in the industry, we believe that the certification process would retain a sufficiently large number of OPOs s to maintain an adequately diversified market in U.S.
Comment:
We received some comments that our threshold rate of 25 percent was arbitrary. We also received comments pointing out parts of the country where no OPO was top tier such as the New England area or the Gulf Coast.
Response:
We respectfully disagree with the commenter that our proposed threshold rate was arbitrary. It was chosen to mathematically achieve the Secretary's goal of doubling kidney transplants by 2030. It was also chosen because, when we assessed which OPOs were top performing, we found that that threshold rate of 25 percent provided us a diversity of OPOs serving a range of geographic areas and different donor potentials. The 25 percent threshold rate and our inclusion of a confidence interval was chosen to accommodate any uncertainty about what constitutes a top performing OPO.
In our December 2019 OPO proposed rule, we presented maps stratifying OPO performance in quartiles. The purpose of these maps was to show that even though many OPOs did not meet the threshold rate, they were quite close. Our current data analyses in Tables 1 through 3 also show that it is likely achievable for many more OPOs to reach the Tier 1 threshold rates. Additionally, our internal analysis indicates that the number of OPOs historically achieving Tier 1 status increased from 16 in 2017 to 24 in 2018, without any regulatory incentives, demonstrating that OPOs have the ability to improve their performance.
Comment:
We received a comment that the 95 percent confidence intervals (CI) were biased against large OPOs because they would likely have a narrow interval.
Response:
The purpose of the confidence interval was to ensure that the use of the threshold rate does not bias against small OPOs who may be prone to greater variability of rates due to smaller volumes. We do not concur with the commenters' assertion that our methodology is biased against large OPOs; they have a CI generated, but because they have more data, their CIs are proportionally smaller.
We did not receive any comments on the proposed mathematical methodology which we use to calculate the “Lowest rate among the top 25 percent” or the time period in which the rate will be calculated. Thus, we will be finalizing as proposed that the threshold rates for the donation and organ transplantation rates would be based on the 12-month period immediately prior to the period being evaluated and finalizing the definition of the Lowest rate among the top 25 percent with technical edits to clarify that the rate is based on the donation and organ transplantation rates in the DSAs.
Final Rule Action:
Under § 486.302, we are finalizing as proposed the definition that the “Lowest rate among the top 25 percent
”
will be calculated by taking the number of total DSAs in the time period identified for establishing the threshold rate. The total number of DSAs will be multiplied by 0.25 and rounded to the closest integer (0.5 will round to the higher integer). The donation rates and organ transplantation rates in each DSA will be separately ranked and the threshold rate will be the rate that corresponds to that integer when counting down the ranking.
We are finalizing § 486.318(e) with revisions, that (1) For each assessment period, threshold rates will be established based on donation rates during the 12-month period immediately prior to the period being evaluated: The lowest rate among the top 25 percent in the DSAs (paragraph (e)(1)(i)), and the median rate among the DSAs (paragraph (e)(1)(ii)) and, (2) For each assessment period, threshold rates will be established based on the organ transplantation or kidney transplantation rates (as applicable) during the 12-month period prior to the period being evaluated: The lowest rate among the top 25 percent in DSAs (paragraph (e)(2)(i)), and the median rate among the DSAs (paragraph(e)(2)(ii)).
We are finalizing as proposed at § 486.318(e)(3) that the 95 percent confidence interval for each DSA's donation and organ transplantation rates will be calculated using a one-sided test.
In response to public comments, we are finalizing § 486.318(e)(4) through (6), the creation of three tiers to identify OPO performance.
Tier 1—OPOs that have an upper limit of the one-sided 95 percent confidence interval for their donation and organ transplantation rates that are at or above the top 25 percent threshold rate established for their DSA will be identified at each assessment period.
Tier 2—OPOs that have an upper limit of the one-sided 95 percent confidence interval for their donation and organ transplantation rates that are at or above the median threshold rate established for their DSA but is not in Tier 1 as described in paragraph (e)(4) will be identified at each assessment period.
Tier 3—OPOs that have an upper limit of the one-sided 95 percent confidence interval for their donation or organ transplantation rates that are below the median threshold rate established for their DSA will be identified at each assessment period. OPOs that have an upper limit of the one-sided 95 percent confidence interval for their donation and organ transplantation rates that are below the median threshold rate for their DSA are also included in Tier 3.
9. Non-Contiguous States, Commonwealths, Territories, or Possessions § 486.318(e)(7)
In the December 2019 OPO proposed rule (84 FR 70628), we did not propose different outcome measures for OPOs exclusively serving non-contiguous states, commonwealths, territories, or possessions because we believe that OPOs servicing those areas should perform at the same level as the top 25 percent of OPOs. That being said, we sought comments on the burden and unique challenges that may face OPOs in the noncontiguous states, commonwealths, territories, or possessions, and whether using just the kidney transplantation rate for the Hawaii OPO would be an appropriate measure of performance as discussed in section V.G “Alternatives Considered” of the December 2019 OPO proposed rule.
Comment:
We received numerous comments in support of using a different standard for OPOs exclusively serving non-contiguous states, commonwealths, territories, or possessions. Both the Hawaii and Puerto Rico OPOs submitted comments describing the difficulty in placing extra-renal organs because of the geographic hurdles.
Response:
Based on information from the commenters regarding the unique geographical challenges of the OPO servicing the Hawaii DSA, we are persuaded to use one different outcome measure to evaluate the OPO's performance in the Hawaii DSA. Instead of using the organ transplantation rate as one measure, we will use the kidney transplantation rate for only the OPO serving the Hawaii DSA. We agree with the commenters that the OPO for this DSA has a clear geographic hurdle to placing extra-renal organs. We will use the same general methodology as the organ transplantation rate for calculating the kidney transplantation rate. We will not age-adjust the kidney transplantation rates for the same reason that we do not age-adjust the donation rates. The age of 75 cutoff provides sufficient age-adjustments for kidney transplantations.
Although we are not using the organ transplant rate for the Hawaii DSA, we will continue to monitor the development and FDA clearance of organ transport devices and expect the OPO serving the Hawaii DSA to adopt these new technologies when they are available. Moreover, we will also use the same donation rate measure for the Hawaii DSA in assessing the OPO's performance since almost all donors of other organs are also kidney donors. Like all of the other OPOs, the Hawaii DSA will be evaluated based on two outcome measures.
We do not intend to give the OPO servicing Puerto Rico any special consideration for their organ transplantation rates. We made the initial decision to not provide special consideration to the Puerto Rico OPO because of its geographic proximity to parts of the continental U.S. that have significant need for organ transplants. Our analysis of 2018 data confirmed our assessment that the OPO based in Puerto Rico does not need special consideration because that OPO would be assigned as a Tier 1 OPO if the metrics where in effect at that time. We suspected that their performance in 2017 had been significantly hampered by the multiple, strong hurricanes, rather than by sustained geographic disparities that do not change from year to year. This suspicion was confirmed by the significantly higher level of performance that the Puerto Rico OPO attained in 2018 when the island was not as impacted by hurricane activity.
Final Rule Action:
We are finalizing in response to comments at § 486.318(e)(7) that for the OPO exclusively serving the Hawaii DSA, the
kidney transplantation rate will be used instead of the organ transplantation rate. The comparative performance and designation to a tier will be the same as in paragraphs (e)(4), (5), and (6) except kidney transplantation rates will be used.
10. Assessment and Data for the Outcome Measures §§ 486.302 and 486.318(f)
In the December 2019 OPO proposed rule, we proposed to assess OPO performance every year, using the most recent 12 months of data from the CDC's MCOD files. Based on the typical timing of the release of the MCOD files, we expect to calculate the outcome measures near the beginning of each calendar year, and the assessment period data will have a 1-year lag. We explained that the reason we were using only 1 year of data is that we did not want to penalize OPOs who have made the effort to improve performance by using their older data in the outcome measure calculations.
Comment:
Some commenters stated that 1 year of data was appropriate for the assessment period for purposes of QAPI remediation, but felt that 3 years of data should be used for re-certification. Other commenters supported our use of 1 year of data for re-certification stating that 36 months of data was too long.
Response:
In the December 2019 OPO proposed rule, we stated that the reason we are using just 1 year of data is that we want to encourage and reward OPOs who make substantial efforts to improve their performance. If we use all the data from the agreement cycle in our QAPI and re-certification, the older data could mask the current performance of the OPO. It is CMS' belief that using the older data from the agreement cycle to assess OPO performance for re-certification may not accurately reflect the practices of the new OPO.
Comment:
We received a comment that OPOs who takeover a DSA should not be held accountable for the performance of the former poorer performing OPO.
Response:
Our assessment periods are normally from January 1 to December 31 based on the state death certificate data files that we receive. In our December 2019 OPO proposed rule at § 486.318(f)(3), we proposed that if an OPO takes over another OPO's DSA on a date later than January 1st, we will hold the OPO accountable for its performance on the outcome measures in the new area once 12 months of data are available. This paragraph recognizes that we need 12 months of data to conduct our analysis and that the new OPO needs the opportunity to be serving the area before they can make changes in response to the outcome measures. Based on the timing of the state death certificate data, it is very likely that most, if not all, of the data at the beginning of a new agreement cycle for a new OPO, will reflect the practices of the prior OPO. However, since we believe it is important that the OPO be aware of the past performance in the DSA and can use that performance as a benchmark for improvement, we will continue to do the evaluation of the assessment period for purposes of ranking and assessing the new OPO and other OPOs. The new OPO would not be required to take actions in its QAPI program in response to the outcome measure, as required at § 486.348(d), until 12 months of data are available. Since we are only using 1 year of data and outcome measures for the final assessment will include data from the middle of the re-certification cycle, the new OPO will not be judged on the performance of the prior OPO and will have had 1-2 years to improve performance in the DSA.
Comment:
We received comments that use of only 1 year of data would be problematic for some OPOs servicing smaller DSAs that happened to have a “bad year” during the final assessment period of their agreement cycle. Because these OPOs are smaller, they have less data for analysis and their DSA could have greater variability in the number of deaths.
Response:
We recognize that OPOs serving smaller DSAs are mathematically subject to greater variability in their inpatient deaths and number of donors and organ transplants. For this reason, the one-tailed confidence interval that we generate in calculating the donation and organ transplantation rates will account for the potential variability when we are using less data in the smaller OPOs.
As also discussed in section II.C of this final rule, for OPOs receiving an ECE extension, their data will continue to be part of the annual calculations of the outcome measures, and the OPOs' performance will be ranked with the other OPOs; the difference is that they will not be up for re-certification in that particular year. All requests for an ECE extension must occur within 90 days after the end of the extraordinary circumstance but no later than the last day of the final assessment period. To seek an ECE exception, the OPO needs to describe the extraordinary circumstance, the time period in which it occurred, why it was beyond the control of the OPO, and why it affected the OPO's performance in such a way that the data does not accurately capture the OPO's performance.
The intention of the ECE extension is to allow for those rare exceptions in which a natural disaster (such as a hurricane), a public health emergency or other similar catastrophe would disproportionately affect an OPO. We could also allow situations in which there are errors in the transmission of data to the CDC.
We believe that OPOs will use the option of seeking the extension judiciously because the request to extend their agreement by 1 year is not without risk. Once an OPO is up for recertification off-cycle from the other OPOs, their DSA could potentially be opened for competition at a time when other OPOs are not up for re-certification. While this would not matter for an OPO in Tier 1, a Tier 2 OPO may be more vulnerable to losing its DSA in competition with other OPOs who have more capacity and interest in competing in an off-cycle year.
Comment:
We received comments that something could happen with staffing during that final year, such as a loss of a high-performing transplant coordinator, which could adversely affect outcomes during that final assessment period.
Response:
Loss of key staff would not be considered an event outside of the OPOs' control and are inevitable in all organizations. Staffing, contingency planning, and other such activities are within the control of an OPO. As such, staffing changes would not constitute an extraordinary event.
Comment:
We also received comments raising concerns about the data lag from CDC, with some commenters assuming that we are calculating rates using numerators and denominators from different time periods. We also received comments that the data lag would result in OPOs being re-certified based on data that is more than 2 years old.
Response:
While there is a lag in the data from CDC, the numerator and denominator will be based on data from the same time period. We are adding clarifying language in our regulatory text at § 486.318(d)(3) to recognize that “for calculating each measure, the data used is from the same time period as the data for the donor potential.” Based on availability of the data from the CDC, the threshold rate determination and the final assessment period will use data from the middle of the agreement cycle. Therefore, OPOs would be notified of their performance on outcome measures for recertification approximately 15 months after the final assessment period just prior to the end of the
recertification cycle. Despite the lags in reporting death certificate data to the CDC, and even the lag in reporting donor and transplant information to the OPTN, the data is the best information available to empirically and transparently evaluate the OPOs' performance.
Final Rule Action:
We are finalizing as proposed at § 486.318(f)(1) that an OPO's performance on the outcome measures is based on an evaluation at least every 12 months, with the most recent 12 months of data available from the OPTN and state death certificates, beginning January 1 of the first year of the agreement cycle and ending December 31, prior to the end of the agreement cycle.
We are finalizing as proposed at § 486.318(f)(3) that if an OPO takes over another OPO's DSA on a date later than January 1 of the first year of the agreement cycle so that 12 months of data are not available to evaluate the OPO's performance in its new DSA, we will evaluate the OPO's performance on the outcome measures in the new area once 12 months of data are available.
In response to the comments and to provide additional clarity, we are also adding a new definition, “Assessment period” at § 486.302 to be a 12-month period in which an OPO's outcome measures will be evaluated for performance. The final assessment period is the 12-month assessment period used to calculate outcome measures for re-certification. We are finalizing at § 486.318(f)(2) that the assessment period is the most recent 12 months prior to the evaluation of the outcome measures in which data is available.
We are also finalizing under § 486.318(d) a clarification for calculating each measure. All OPOs will be evaluated based on two measures. For all OPOs, the numerator for the donation rate is the number of donors in the DSA. For most OPOs, the numerator for the organ transplantation rate is the number of organs transplanted from donors in the DSA. For the OPO servicing the Hawaii DSA only, the donation rate will be the same as for all other OPOs but the kidney transplantation rate will be utilized in lieu of the organ transplantation rate. The numerator for the kidney transplantation rate is the number of kidneys transplanted from kidney donors in the DSA. The numbers of donors, organs transplanted, and kidneys transplanted are based on the data submitted to the OPTN as required in § 486.328 and § 121.11. For calculating each measure, the data used is from the same time period as the data for the donor potential.
11. Implementation Timeline
We requested comments on this proposed change in the December 2019 OPO proposed rule to the applicability of the outcome measure requirements for the cycle beginning in 2022 and ending in 2026. The current OPO certification cycle is due to end on July 31, 2022 however, the OPO agreements for the certification period extend until January 31, 2023. This extra timeframe in the agreement affords the opportunity for any appeals or competition that may occur from any potential enforcement action for non-compliance with the CfCs, including the outcome measures. Normally, absent enforcement action on the part of CMS, the OPO agreements are renewed on August 1 or shortly thereafter to coincide with the start of the next certification period.
Comment:
We received a number of comments from the general public and others that encouraged us to implement these new measures as soon as possible and to hold OPOs accountable now. We also received numerous comments from OPOs, supporting a delay of implementation of the new outcome measures to begin in 2022 and end in 2026.
Response:
We appreciate the robust comments related to this topic including the desire to drive performance improvements sooner while also being responsive to providing OPOs time to adapt to the new measures and improve performance where needed. We considered the option of extending the current agreements by 2 years and assessing OPOs based on data from 2023 holding OPOs accountable to the new performance measures in 2024. However, the effects of the current COVID-19 PHE are still uncertain in regards to the impact to the organ donation and transplantation system. We note that current data from the OPTN indicate that as of November 7, 2020, there were 28,506 deceased organ transplants conducted compared to 27,658 at this same time the year prior suggesting the impacts may not be as severe as originally anticipated.
18
Therefore, we intend to implement the new measures as proposed, beginning in the 2022 recertification cycle. We believe extending implementation beyond this timeframe will negatively impact our efforts drive improvements to make these critically important life-saving organs sooner.
18
https://unos.org/covid/
.
OPOs will continue to receive performance measures under the current metrics until the end of the current certification cycle in 2022. However, we intend to also begin providing OPOs an assessment of their performance under the new metrics in each DSA immediately using data from 2019. OPOs will receive results of their performance on the outcome measures from 2019 in the first quarter of 2021 with additional assessments being provided annually. We will rank OPO performance to provide information that may be utilized for purposes of QAPI programs interventions leading up to implementation of the new measures. OPOs will receive performance assessments in the first quarter of the year for their performance 2 years prior. As previously stated, this time lag is inherent the use of objective, reliable, and transparent publically available data sources. It affords the CDC time to collect all information and develop the report for public posting. Additionally, it provides time for CMS to receive and process information, conduct analysis, share preliminary results with OPO, and make the files public. Therefore, for the 2022-2026 certification period, the threshold rate will be established based on data from 2023 and the final assessment period will utilize data from 2024. CMS will conduct activities for recertification in early 2026, including publication of tier ranking in performance measures and conducting onsite surveys of OPO operations. While we acknowledge that OPOs will not know the actual threshold rate that will be utilized for the final assessment period until after it is complete, they will have the results of prior years from which to trend and incorporate into their QAPI program to assist in improving performance. Additionally, we expect that OPOs implement a comprehensive data-driven QAPI program to monitor and evaluate their performance. Therefore, OPOs should already be including a range of data and activities for this purpose that will inform and drive performance toward success in achieving Tier 1 status on the outcome measures and the new QAPI requirement at § 486.348(d) will be one component of that comprehensive plan.
Final Rule Action:
This final rule will be effective 60 days after the publication date and the new outcome measures will be implemented on August 1, 2022 to coincide with the start of the next certification period.
12. Definitions § 486.302
In the December 2019 OPO rule, we proposed to remove several definitions from § 486.302, since these terms would no longer apply. Specifically, we proposed to remove the definitions of
“eligible death,” “eligible donor,” “expected donation rate,” “observed donation rate”, and “Standard criteria donor (SCD)”. We proposed to revise the definition of “Donor” described in section II.B.3 of this final rule and we will add the terms “Assessment period” described in section II.B.10 of this final rule, “Death that is consistent with organ donation” described in section II.B.6 of this final rule, “Donation rate” described in section II.B.2 of this final rule, “Donor potential” described in section II.B.6 of this final rule, “Kidney transplantation rate” described in section II.B.9 of this final rule, “Lowest rate among the top 25 percent” described in section II.B.8 of this final rule, and “Organ transplantation rate” described in section II.B.4 of this final rule. Public comments related to these definitions and our responses are addressed in sections II.B of this final rule as described above. The addition of “assessment period” and “kidney transplantation rate” were not proposed, and are being added in response to public comments and to provide convenience in understanding the other definitions being defined in the regulation text. The term “Lowest rate among the top 25 percent” was proposed, and we did not receive any comments regarding our methodology for calculating this rate. Therefore, we are finalizing with technical edits. We will define these terms as follows:
• “
Assessment period”
is a 12-month period in which an OPO's outcome measures will be evaluated for performance. The final assessment period is the 12-month assessment period used to calculate outcome measures for re-certification.
• “
Death that is consistent with organ donation”
means all deaths from the state death certificates with the primary cause of death listed as the ICD-10-CM codes I20-I25 (ischemic heart disease); I60-I69 (cerebrovascular disease); V-1-Y89 (external causes of death): Blunt trauma, gunshot wounds, drug overdose, suicide, drowning, and asphyxiation.
• “
Donor potential”
is the number of inpatient deaths with in the DSA among patients 75 and younger with a primary cause of death that is consistent with organ donation. For OPOs servicing a hospital with a waiver under § 486.308(e), the donor potential of the county for that hospital will be adjusted using the proportion of Medicare beneficiary inpatient deaths in the hospital compared with the total Medicare beneficiary inpatient deaths in the county.
• “
Donation rate”
is the number of donors as a percentage of the donor potential.
• “
Kidney transplantation rate”
is the number of kidneys transplanted from donors in the DSA as a percentage of the donor potential.
• “
Lowest rate among the top 25 percent”
will be calculated by taking the number of DSAs in the time period identified for establishing the threshold rate. The total number of DSAs will be multiplied by 0.25 and rounded to the closest integer (0.5 will round to the higher integer). The donation rates and organ transplantation rates in each DSA will be separately ranked and the threshold rate will be the rate that corresponds to that integer when counting down the ranking.
•
Organ
means a human kidney, liver, heart, lung, pancreas, or intestine (or multivisceral organs when transplanted at the same time as an intestine). The pancreas counts as an organ if it is used for research or islet cell transplantation.
Organ type
Number of
organs transplanted
Right or Left Kidney
1
Right and Left Kidney
2
Double/En-Bloc Kidney
2
Heart
1
Intestine
1
Intestine Segment 1 or Segment 2
1
Intestine Segment 1 and Segment 2
2
Liver
1
Liver Segment 1 or Segment 2
1
Liver Segments 1 and Segment 2
2
Right or Left Lung
1
Right and Left Lung
2
Double/En-bloc Lung
2
Pancreas (transplanted whole, research, islet transplant)
1
Pancreas Segment 1 or Segment 2
1
Pancreas Segment 1 and Segment 2
2
•
Organ transplantation rate
is the number of organs transplanted from donors in the DSA as a percentage of the donor potential. Organs transplanted into patients on the OPTN waiting list as part of research are included in the organ transplantation rate. The organ transplantation rate will be risk-adjusted for the average age of the donor potential using the following methodology:
(1) The age groups used for the adjusted transplantation rates are: <1, 1-5, 6-11, 12-17, 18-24, 25-29, 30-34, 35-39, 40-44, 45-49, 50-54, 55-59, 60-64, 65-69, 70-75.
(2) Calculate a national age-specific transplantation rate for each age group. An expected transplantation rate for each OPO is calculated as ∑(g=1)Gdg*Rg/∑gdg, where dg is the number of potential donors in the OPO in age group g, Rg is the age-specific national transplantation rate in age group g, and ∑gdg is the OPO's total number of individuals in the donor potential. This can be interpreted as the overall expected transplantation rate for an OPO if each of its age-specific transplantation rates were equal to the national age-specific.
(3) Calculate the age-adjusted organ transplantation rate as (O/E)*P, where O is the OPO's observed unadjusted transplantation rate, E is the expected transplantation rate calculated in Step 2, and P is the unadjusted national transplantation rate.
Comment:
We received several public comments related to the deletion of definitions.
Response:
We have addressed all comments related to the deletion of definitions in our discussion about the outcome measures throughout section II.B of this final rule. Comments and responses were addressed in the manner to how they applied to the related new or revised definitions. Eligible death was described in the context of the donor potential in section II.B.6; eligible donor and standard donor criteria in the context of donor definition at section II.B.3; and expected donation rate in the context of risk adjustments at section II.B.7 of this final rule.
Final Rule Action:
Under § 486.302, we are finalizing as proposed, the removal of the following definitions: “
Eligible death,”
“
Eligible donor,”
“
Expected donation rate,”
“
Observed donation rate,”
and “
Standard criteria donor
(SCD).” We are also finalizing as proposed, by adding the definition of “
Donation rate.”
We are finalizing as proposed with modifications, the definitions of “
Donor potential”
and “
Organ transplantation rate.”
And we are finalizing the new definitions: “
Assessment period,”
“
Death that is consistent with organ donation,”
and “
Kidney transplantation rate.”
C. Re-Certification and Competition Processes (§ 486.316)
1. Re-Certification and Competition Processes § 486.316(a)
In the December 2019 OPO proposed rule, we proposed to revise § 486.316(a) to provide that the OPO must meet the performance requirements of the outcome measures at § 486.318 at the end of the re-certification cycle; and has been shown by survey to be in compliance with the requirements for certification at § 486.303, including the conditions for coverage at §§ 486.320 through 486.360.
We proposed revisions at § 486.316(a)(1) to correspond to our proposed outcome measures that were
set forth at § 486.318 in the December 2019 OPO proposed rule. To be consistent with the tier system finalized in this rule, we also need revised § 486.316(a)(1), (a)(2) and (a)(3) to reflect that the OPO has been shown by a survey to be in compliance with the conditions for coverage from “§§ 486.320 through 486.360,” so that it is included § 486.360 Conditions for Coverage: Emergency Preparedness, which was effective on November 15, 2016 (81 FR 63859). We are finalizing the inclusion of § 486.360 in § 486.316(a)(1)(i), (a)(2)(i) and (a)(3)(i).
In addition, we proposed to remove § 486.316(a)(3), which provided that for the 2022 recertification cycle only, an OPO is recertified for an additional 4 years and its service area is not opened for competition when the OPO meets one out of the two outcome measure requirements described in § 486.318(a)(1) and (3) for OPOs not operating exclusively in the noncontiguous States, Commonwealths, Territories, or possessions; or § 486.318(b)(1) and (3) for OPOs operating exclusively in noncontiguous States, Commonwealths, Territories, and possessions. An OPO is not required to meet the second outcome measure described in § 486.318(a)(2) or (b)(2) for the 2022 recertification cycle. We intend to relocate paragraphs (a)(3) and (b)(2) to paragraph (g) due to the proposed new outcome measures set forth at § 486.318 becoming effect at the start of the next recertification cycle in 2022.
As described in sections II.B “Proposed Changes to Definitions (§ 486.302)” and “Proposed Changes to Outcome Requirements (§ 486.318)” of this final rule, we are not only finalizing new outcome measures, but we are also finalizing a tier system. The tier system will determine whether the OPO is immediately re-certified, must compete to retain its DSA, or will receive an initial de-certification determination. Thus, we are amending our proposal and finalizing § 486.316(a) to incorporate the tier system.
Final Rule Action:
We are finalizing § 486.316(a) by incorporating the language for the tier system to indicate the requirements for each tier. We are also finalizing the inclusion of § 486.360 in the CfCs that all OPOs must meet for re-certification. We are also revising § 486.316(a)(3) as discussed above.
2. De-Certification and Competition Processes § 486.316(b)
In the December 2019 OPO proposed rule, we proposed that if an OPO does not meet the performance requirements or the outcome measures as described in paragraph (b) of this section at the final assessment prior to the end of the re-certification cycle or the requirements described in paragraph (b)(2) of this section, the OPO would be de-certified. If the OPO does not appeal, or the OPO appeals and the reconsideration official and CMS hearing officer uphold the decertification, the OPO's service area is opened for competition from other OPOs. The de-certified OPO is not permitted to compete for its open area or any other open area. An OPO competing for an open service area must submit information and data that describe the barriers in its service area, how they affected organ donation, what steps the OPO took to overcome them, and the results.
As discussed in section II.B of this final rule and based on the comments we received, we are finalizing new outcome measures, for all OPOs, and except for the Hawaii DSA, those measures are the donation and the organ transplantation rates. Based on public comments, we are also establishing a tier system that will be used to classify OPOs for purposes of re-certification, decertification, appeals, and competition. The outcome measures and tier system are discussed in detail in sections II.B.2, II.B.4, and II.B. 6 through 10 of this final rule.
We requested comments on competition, including whether all DSAs should be opened at the end of the re-certification cycle for competition under § 486.316. Only one of the commenters wanted all of the DSAs open for competition each re-certification cycle regardless of the OPO's performance. Most of the commenters, however, wanted more competition than existed under our prior rules and contended that more competition would improve OPO performance. Some commenters suggested that OPOs that were doing well should not have to compete to retain their DSAs because it would divert resources from their primary mission of procuring organs. This finalized rule does provide for more competition to drive improvements in performance. Prior to this finalized rule, OPOs were either re-certified or de-certified based on their outcome measures. In this final rule, OPOs will be assigned to a tier based on their outcome measures. Only those OPOs that are designated as Tier 1 OPOs will not have their DSAs opened for competition (§ 486.316(a)). Tier 3 OPOs will be decertified and, following any appeals, their DSAs will be opened for competition, unless the de-certification is reversed as a result of the appeals process. With respect to Tier 2 OPOs, those DSAs will also be opened for competition. The incumbent OPO will have to compete if the OPO wants to retain its DSA and the DSA will also be open for competition from any other OPO that is qualified to compete for open DSAs. If a Tier 2 OPO does not win the competition for its DSA and does not win the competition for any other open DSA it competes for, CMS will not renew its agreement with the OPO. The OPO will not be able to appeal this non-renewal, which is not a de-certification. The change to a tiered approach increases the number of DSAs open for competition and the number of OPOs eligible to compete for open DSAs, which is consistent with the recommendation of most public commenters. Although we proposed to change the criteria for competition at § 486.316(c) to correspond to the new outcome measures in § 486.318, we did not propose any changes to the selection criteria for competition at § 486.316(d). We appreciate all of the commenters that submitted comments on the competition process. Those comments have been reviewed and will be considered in any future rulemaking
Comment:
Some commenters contended that de-certification was too severe of a consequence for OPOs below the lowest rate among the top 25 percent. Those commenters do not believe that this would provide incentive for OPOs to improve their performance.
Response:
The establishment of the tier system should provide OPOs with the incentive to improve their performance. We believe that it is realistic that all OPOs, even those that we have estimated would be de-certified based on their past performance, can avoid de-certification by improving their performance. After considering public comments, we have lowered the level of performance that would lead to an OPO being decertified. We do not agree with the commenter that de-certification is too severe of a consequence for Tier 3 OPOs. If an OPO cannot achieve the outcome measures we are finalizing in this rule, or cannot demonstrate compliance with the OPO CfCs through its re-certification survey, we believe that de-certification is the appropriate consequence.
In reviewing our proposal in light of this comment, however, we believe that the language in this section should be clarified. In the December 2019 OPO proposed rule, we said, “the OPO is de-certified.” We believe that statement could be misleading. As set forth in § 486.314(a), when CMS determines that an OPO will be de-certified because of involuntary termination or non-renewal of its agreement with CMS, CMS will
mail the OPO an initial de-certification determination. The OPO then has the appeal rights set forth in § 486.314. Thus, we are revising the language from what we proposed at § 486.316(b) by removing, “the OPO is de-certified” and inserting “CMS will send the OPO a notice of its initial de-certification determination and the OPO has the right to appeal as established in § 486.314”. We have also separated the three requirements after the stem statement to improve clarity and readability.
Comment:
Some commenters contended that the OPO CfCs did not need the drastic changes we proposed. Some commenters contended that many OPOs were performing well and the system was not underperforming to the extent that the proposed rule contended.
Response:
We agree that some OPOs, as demonstrated by their performance on our assessment of their performances based the new outcome measures, are doing a great job in procuring transplantable organs and working with donor families. This is why we are finalizing the tier system that recognizes those OPO's superior performance. In addition, the estimated number of OPOs that would be de-certified under the proposed rule (refer to Table 3 in the 2019 December OPO proposed rule) was based on the past performance of the OPOs. We believe that OPOs will be incentivized to improve their performance because of the outcome measures and tier system finalized in this rule. At the end of the first re-certification that uses these outcome measures and tier system, we believe that fewer OPOs will be de-certified.
Comment:
One commenter was concerned about unintended consequences of the requirements that may come to light if the proposals were finalized.
Response:
In any change of regulation, there is always a possibility of unintended consequences. We have taken all of the appropriate steps necessary to consider and develop outcome measures that we believe will improve OPO performance and increase the number of transplantable organs for those individuals on the waiting lists. In addition, OPO performance and patient access impacts will be monitored closely. If any unintended consequences come to our attention, we will appropriately evaluate and address them at that time.
Comment:
Some commenters expressed concern that increased pressure from the new outcome measures and the threat of de-certification would damage the relationships between the OPOs so that they will no longer cooperate or share best practices with each other. The commenter noted that this was especially concerning since the OPTN is moving towards a geographical allocation system, which makes cooperation between OPOs even more important. One commenter contended that the proposal had already damaged some collaboration between OPOs.
Response:
While collaboration between OPOs is a worthy goal, such collaboration has not resolved the significant, ongoing disparities that exist in OPO outcomes. Thus, it is CMS' belief that it is necessary to revise the current policies. We believe that the need for additional organs presents such a great need as to outweigh any impacts to OPO collaboration. Thus, in order to achieve such a benefit, it is necessary for incentives for OPOs to improve performance or face competition and decertification.
By finalizing a tiered system, only OPOs that are not in compliance with the outcome measures, or found to be not in compliance with the conditions for coverage at the re-certification survey, will be designated as Tier 3 and receive a notice of de-certification. Many OPOs that would have been de-certified under the proposed outcome measures will be designated in Tier 2 and have the opportunity to compete to retain their DSAs. While this approach may change the nature of recertification, we do not believe it should change the nature of OPO relationships with each other. Cooperation among other OPOs in procuring and placing organs could not only improve an OPO's performance on the outcome measures, but also increase the number of transplantable organs.
Based upon this tiered system, OPOs that fail to meet the outcome measures as specified in § 486.318(e)(6), that is an OPO that fails to meet the median threshold for the donation or transplantation measures, fails to meet the median threshold for the donation and transplantation measures or fails to demonstrate compliance with the OPO CfCs via the re-certification survey, will be the only OPOs that are designated into Tier 3. An OPO that qualifies for Tier 3 designation will receive an initial notice of de-certification determination, has the appeal rights set forth at § 486.314, and, if decertified, cannot compete for either its own or any other open DSA.
Final Rule Action
This text is long and has been trimmed here. Open the source document for the complete record.
This is a copy of a public record, reproduced as it was published. It is not legal advice, and it may not be the version a court would rely on. Check the official source before you cite it.