Centers for Medicare and Medicaid Services
7500 Security Boulevard
Baltimore, MD 21244
RE: Draft Guidance for the Public, Industry, and CMS Staff; Coverage with Evidence Development in the Context of Coverage Decisions
Dear Administrator Tavenner:
The Medical Imaging & Technology Alliance (MITA) is pleased to comment on the draft guidance by the Centers for Medicare and Medicaid Services (CMS) issued November 29, 2012, on the use of Coverage with Evidence Development (CED) in the context of coverage decisions. As the leading trade association representing medical imaging, radiotherapy technology, and radiopharmaceutical manufacturers, we have an in-depth understanding of the significant benefits to the health of Medicare beneficiaries that medical imaging, radiotherapy, and proton therapy provide. MITA is pleased to continue working with CMS to ensure access to and appropriate use of these life-saving and life-changing technologies.
We agree with the use of CED in certain circumstances, particularly, as CMS states, when “available evidence may not support confident conclusions of clinical benefit for all uses of potentially beneficial items and services.”1 By providing coverage via CED to emerging and existing technologies, CMS gives Medicare beneficiaries access to these innovations they otherwise would not have had. We emphasize that the clinical benefit CMS seeks may take a different shape depending on the objective of the use of the technology in question. Specifically, and in adhering to the theme that emerged from MITA’s PET Endpoints Workshop in November 2012, diagnostic tools should be treated differently than therapeutic modalities when CMS considers the evidence to support their (respective) coverage policies. This comment letter will focus on the theme that diagnostics should receive differential treatment as we respond to various sections of the draft guidance.
We applaud CMS for enhancing the President’s National Bioeconomy Blueprint by identifying CED as a way to promote innovation among emerging technologies for Medicare patients. As the Blueprint stated, the CED authority has been applied sparingly over the decade of its existence.2 We encourage CMS to continue to use this authority in rare circumstances because technologies that have achieved a high level of evidence, demonstrating safety and efficacy for the Food and Drug Administration (FDA), should in most cases be considered reasonable and necessary by CMS for Medicare beneficiaries without additional evidentiary burdens. In addition, invoking CED means that costs not reimbursed by Medicare are added to the system; such costs include, but are not limited to designing a study, getting the study approved, recruiting sites and patients, collecting and analyzing data, and publishing the results.
MITA is not opposed to CED, but we believe that the use of CED should continue sparingly, when coverage is suggestive of improved health outcomes for Medicare beneficiaries, but is not conclusive. We oppose the use of CED as a first-line coverage mechanism for emerging or existing technologies; we urge CMS to continue its long-standing practice of allowing coverage at the local contractor level, and then considering alternative policies (including CED) when there is a reason to question the reasonableness and necessity of coverage under Social Security Act § 1862(a)(1)(A). We agree that the use of CED should occur to build evidence that an item or service is generalizable to the clinical situation in which Medicare beneficiaries regularly present. Using CED at the national level is generally presumed, but there are cases where local contractors have established policies that require patients to enroll in a registry to receive coverage for an item or service. We encourage CED at the local level to be used sparingly, in a transparent manner, and for the same purpose of obtaining evidence for an item or service to expand coverage.
We understand CMS’ desire to define more narrowly how CED will be applied, but we are unsure why CMS believes that “coverage with appropriateness determination” (CAD) does not “encompass the concept of research” that is central to CED. We request that CMS more clearly explain the rationale for using only “coverage with study participation” (CSP) as a more effective means of building an evidence base. With the pathways of CED more clearly defined, manufacturers will be better able to work with CMS to answer the questions about evidence that are posed.
In using either form of CED, we assume that CMS will continue to issue tracking sheets on National Coverage Analyses and will continue to use the other notifications and tools of transparency that are part of the national coverage determination (NCD) process. In this spirit, we encourage CMS to continue to solicit discussions with stakeholders who make coverage requests, as well as the providers, patient groups, and manufacturers who would implement and participate in CED studies, so that all stakeholders can provide advice on gaps in the evidence base and the resources and timelines required for meaningful results.
MITA generally supports the processes of applying CED as proposed in the draft guidance. CED must generate evidence that is relevant to the health outcomes in the Medicare population. We encourage CMS to examine the intermediate benefits of technologies on health outcomes in this population, not just final endpoints. In particular, diagnostic modalities generate important evidence that is likely to affect the way patients are managed, which will lead to changes in the ultimate health outcomes of Medicare patients. However, the ultimate outcome may not be apparent in a patient during the timeframe for a CED study, while a change in management that leads to that outcome would be immediately evident, thus providing an incremental endpoint or piece of evidence that CMS should consider. We also encourage CMS to limit the use of CED to technologies for which no evidence exists that is translatable or generalizable to the Medicare population; if such evidence exists, then we would not be supportive of the resources and time invested into CED.
We understand CMS’ concern about covering technology that could harm beneficiaries or lack significant benefit to the Medicare population. We encourage CMS to examine the other side of the evolution and reevaluation of the evidence base by seeking to confirm an item’s or service’s meaningful benefit to this population, rather than starting from an assumption against coverage. We ask that CMS use evidence to “rule-in” the use of a product, rather than build evidence to rule out its use.
We agree that the setting of care is important in determining the coverage of a product. If the evidence for a technology has been reproduced or recreated only in a research or laboratory setting, then we understand CMS’ need for evidence that is applicable to the clinical setting at large. We are confident that our membership will have that type of evidence available for their products upon FDA approval, obviating the need for CSP.
While starting CED is a huge undertaking for both manufacturers (study sponsors) and for CMS, ending CED must be a highly deliberate and methodical process as well. With all new imaging and most new diagnostic technology fundamentally based on computer science innovations following Moore’s Law doubling of computer power every 18 months, we encourage CMS to use timelines of this 18 month magnitude to gate CED study designs. This way, an 18 month CED timeline might only impact one cycle of technical innovation. Thus, if a study (using the CSP methodology) closes, there needs to be a specific timeline when CED ends and coverage begins or is not granted. We encourage CMS to close the potential non-coverage gaps by working with requesters to ensure a timely decision, within the confines of the administrative regulatory process.
MITA is adamant that the parallel review process, while potentially successful, must be approached with an eye toward FDA approval and CMS coverage simultaneously, not consecutively. We understand that the Memorandum of Understanding between CMS and FDA is in force, and there is a degree of cooperation between the agencies.3 The October 2011 Federal Register notice proposing parallel review, though, projected a more comprehensive document. Assuming this document is still under development, MITA suggests that potential CED activities be discussed with a product sponsor during the FDA evaluation so that there will not be further delay in the collection of evidence once a product receives marketing approval. Therefore, for products for which CED may be necessary, such coverage would be immediate upon approval so that the needed evidence can be developed in a timely manner, and in a fashion that provides access to the Medicare population. As we stated at the beginning of this letter, we expect that most products will receive coverage without CED because the evidence they bring through the FDA approval process demonstrates that the products are reasonable and necessary under SSA § 1862(a)(1)(A).
MITA also strongly believes that CED and the parallel review process must focus on answering the question of whether an item or service is reasonable and necessary for the Medicare population. Even though we appreciate scientific rigor (for our members would not be innovators without it), and we demand its use in determining market approval and Medicare coverage, we think more general health services research questions on the broader practice of clinical care should be left to research institutions that can evaluate these technologies as their use increases. Non-core research into certain characteristics of a product that do not show significant benefit to Medicare patients should not be part of either CED or the parallel review process.
Formal Evidentiary Criteria for CED
CMS has stated in this draft guidance and previously that CED is a way to understand whether or not promising technologies yield positive health outcomes. The evidentiary threshold for applying CED to diagnostics, though, should not be the same as for therapeutics. For diagnostics, substantive changes in patient management should be the appropriate endpoint rather than the overall health outcomes of patients. We request that CMS provide more explicit guidance on developing evidence based on diagnostic endpoints. CMS has demonstrated through the utilization of the oncologic PET tracers CED NCD (administered by the National Oncologic PET Registry, NOPR) that it understands that the the agency must consider the evidence from diagnostic tools differently than therapeutic items or services. As stated in the draft guidance, the MEDCAC meeting in May 2012 elicited several perspectives of the application of evidentiary criteria for CED. The panel conclusions deserve some comment:
An evidentiary threshold can be defined to invoke CED.4 We agree with this assertion, and add that the threshold may differ depending on the type of technology and indications under review. CMS should work with stakeholders to define the threshold for each potential application of CED so that there is a complete understanding of the question at hand. CMS also should continue to work with stakeholders to develop clear guidance that will explain the general criteria for determining whether there is enough evidence for CED, but not enough for a conclusive coverage determination. This guidance also should differentiate diagnostics and therapeutics.
An evidentiary threshold can be defined to trigger an evidentiary review to determine if CED should cease, continue or be modified.5 This threshold should be defined at the time the decision to engage in CED is announced, thus making clear the intent of the evidence development process. We assume the threshold will vary from technology to technology, and that it should be established through agreement among the technology’s sponsor, CMS, and any relevant professional societies based on the need for CED and the research protocols to be applied under CED.
An evidentiary threshold would be influenced by general and particular characteristics of the item or service, the disease, and the availability of acceptable alternatives. There may be meaningful interaction of these alternatives.6 We agree that the threshold would depend on the individual characteristics of the item or service, as well as the disease. We are unsure how the availability of acceptable alternatives would influence the determination of whether an item or service stands on its own merit (body of evidence) as part of the establishment of CED, or gaining coverage in general.
Generalizability (to additional settings, practitioners or other clinical indications) may comprise the primary evidence gap for some bodies of evidence.7 As we discussed earlier, we believe that CED should be implemented practically, and if there is a reason to close an evidence gap, including that the available evidence cannot be applied to Medicare patients, then a study may be necessary. The study, of course, should meet the requirements that we stated earlier, including close collaboration between investigators and CMS with an eye toward coverage.
Transparency of and Proprietary Information Confidentiality in CED
We appreciate that CMS intends to use the results of published CED studies to inform coverage decisions, which in our opinion is the sole reason for the undertaking. In terms of transparency, we are disappointed that CMS has not delineated the steps it will use to determine when and how to use CED, broad instructions for interacting with CMS when there is a question that CED may answer, and more specificity on timing and timelines in the use of CED. MITA also is very concerned that this draft guidance failed to mention that CMS would take proactive steps in maintaining the integrity of confidential trade and proprietary information through the CED and coverage process. We request that CMS provide such information in a future guidance document, and we hope that there will be further opportunity to provide our input.
We are encouraged that CMS is taking a deliberative approach to implementing CED for the long term. We again encourage CMS to consider each technology and each type of technology individually, particularly drawing the distinction between diagnostics and therapeutics. MITA would like to express our sincere thanks to CMS for posting this draft guidance. For further information or questions, please contact Brian Abraham, Senior Policy Director, at firstname.lastname@example.org or (703) 841-3258.
Gail M. Rodriguez, PhD
Executive Director, MITA
Vice President, NEMA
cc: Louis Jacques (CMS/CAG)
Rosemarie Hakim (CMS/CAG)
1 Draft guidance document, November 29, 2012.
2 National Bioeconomy Blueprint. The White House. April 26, 2012. Also cited in draft guidance document, Nov. 29, 2012.
3 MOU No. 225-10-0010 at 75 FedReg 48699, June 25, 2010.
4 CED Draft Guidance, Section VII. Nov. 29, 2012. At www.cms.gov/medicare-coverage-database/