Before a medical product can be widely used in the United States, it generally must first be approved or cleared for marketing by the US Food and Drug Administration (FDA). Then, payers such as the Centers for Medicare & Medicaid Services (CMS) must decide whether the product merits coverage and payment. Because the statutes governing these agencies evolved to meet the exigencies of particular moments in the history of medical product development, the degree of convergence in standards and in the underlying evidence needed to support regulatory and payment decisions is not always immediately obvious. The resulting fragmentation—perceived or real—has led to questions about whether FDA approval or clearance for marketing will necessarily result in approval for coverage and payment.
Despite these challenges, changes in the organization of health care and in the larger information ecosystem should allow the FDA and CMS to move increasingly toward use of shared sources of evidence while still applying the most appropriate criteria to their decision making. Such a move should help reduce current gaps in evidence that create uncertainty surrounding the approval or clearance of new therapies and their subsequent use in practice. It should also enable greater efficiency in medical product development and provide the higher-quality evidence needed in the emerging era of precision medicine.
Although different standards are applied for product approval or clearance for marketing (in the case of the FDA) and coverage and payment (in the case of CMS), both agencies base their determinations on scientific evidence. The FDA bases drug approval for marketing on “substantial evidence” derived from at least 1 well-conducted clinical trial,1 whereas CMS uses “reasonable and necessary” as the standard for coverage.2 In the increasing number of situations in which “coverage with evidence development” is used by CMS, coverage decisions are based on “… formal review of the medical literature … only in the context of an approved clinical study or when additional clinical data are collected to assess the appropriateness of an item or service for use with a particular beneficiary.”3
For much of the modern era of medicine, research intended to support marketing approval or clearance by the FDA has usually examined the effects of therapeutics in narrow, strictly delineated populations enrolled in trials conducted under conditions that may not reflect the clinical practice settings in which the product will actually be used. There are excellent reasons that this system evolved, including the need to control confounding variables and ensure the internal validity of study findings. Given that approximately 90% of candidate drugs that reach the point of human trials fail before attaining marketing approval4 (as do many devices designed to treat serious medical illnesses), demonstrating that a therapy can work even when administered under a strict protocol in a carefully defined population represents a major challenge.
However, when a product demonstrates promise in the rarefied setting of earlier-phase clinical research, developers should rapidly pivot toward evaluating that product in the clinical contexts for which it is intended. In other words, the United States needs to develop high-quality evidence about the risks and benefits of tangible health outcomes in clinical settings and among patients representative of those who will actually use these products. This expansion of scope earlier in the product life cycle will help establish how a given product is likely to perform once it is marketed and how best to administer that product. Early involvement of health systems and payers will also help to understand and address the kinds of evidence needed to incorporate the new product into practice, place it on formulary or device inventory, and decide whether and how much to pay for its use, thereby facilitating implementation and adoption. Notably, the European Medicines Agency has recently come to the same conclusion in concert with its health technology assessment organizations.5
Before the Kefauver-Harris Amendments were enacted by Congress in 1962, drugs were required to undergo evaluation only for safety, not efficacy, and many were sold on the basis of anecdote. However, following passage of the amendments, the FDA sought to ascertain whether drugs met new standards for effectiveness.6 This program was a tremendous success in terms of requiring that drugs demonstrate efficacy in specific, well-defined populations, and similar systems have developed for high-risk devices. But the unintended results of limitations arising from the separate evolution of clinical and research environments and the absence of a means for efficient data exchange between these 2 spheres were trials that often do not adequately inform practice and product labeling that sometimes lacks optimal empirical information to guide use of the product in clinical practice.
It is imperative that the US regulatory groups bridge this gap between evidence that is generated through trials performed primarily to obtain marketing approval for drugs and high-risk devices and the optimal kinds of evidence needed to support the intended uses of these products in practice. Therefore, to ensure that patients, clinicians, and payers have adequate evidence to guide their choices, the FDA and CMS are focusing on the following approaches:
First, these agencies are clarifying the need for including diverse populations and measuring relevant clinical outcomes within the sphere of trials conducted for regulatory approval and to inform labeling. This task is particularly important in the case of accelerated approvals based on biomarkers: because the current regulatory framework requires postmarket trials to evaluate the effect of the therapy on the clinical outcome of interest, such approvals are expected to increase in tandem with rapid progress in molecular medicine.
Second, the FDA and CMS are collaborating with other federal agencies to build functional links across a range of systems developed to capitalize on existing digital information collected in the course of health care delivery, including electronic health records, insurance claims, and data housed in clinical registries. These approaches should significantly improve the quality and reduce the cost of evidence generation and enable the inclusion of much more representative populations into clinical trials and other studies. This in turn will provide the evidence needed to better characterize the balance of risk and benefit and establish the appropriate use and value of therapies in clinical practice.
Third, this effort will require broad collaboration across public and private sectors. The Patient-Centered Outcomes Research Institute, which is overseen by a board representing key stakeholders (including the medical products industry), is one example of an entity designed to develop this knowledge. It will be particularly important for clinicians, professional societies, and health systems to engage in developing effective approaches to integrating broader evidence generation with practice.
The public expects and deserves that regulatory and coverage decisions made by the FDA and CMS are based on the most robust evidence possible. At the same time, the strong evidence base needed by both agencies can be built largely by drawing on shared data sources. As electronic health records, claims databases, and clinical data quality registries continue to advance and become accessible by multiple stakeholders, a “tipping point” will occur and will enable more inclusive trials to be done on a wider scale at much lower cost. This will allow the FDA and CMS to measure health outcomes that matter to patients and health care consumers and define the appropriate use of medical products in a more straightforward, evidence-based manner.
The phrase “personalized medicine” is often used to encapsulate the emerging opportunity to tailor treatment according to the biology and preferences of the individual patient. Scientific and technological capabilities are rapidly advancing toward a point where improved understanding of the underlying biology of disease will in many cases enable making biologically based estimates about the performance of a drug or an engineering-based estimate about the performance of a high-risk device. However, the FDA and CMS are concerned that this term could be misconstrued as implying that science will enable individual practitioners to accurately predict what will happen to patients without reference to a larger base of empirical evidence gathered from broader groups. For this reason, the term precision medicine is preferred by the National Academies of Science, in recognition that it is this aggregate knowledge about the breadth of variation in biological and clinical characteristics across populations that will enable precise therapies to be tailored to individuals.7
All decisions about the use of medical products made jointly by clinicians and patients can be considered personalized decisions that should optimize the balance of benefits and risks for patients, taking into account their needs and preferences and the totality of scientific findings about the products. The best and most cost-effective therapy is the one that is used correctly—in the right patient, at the right time, in the right manner, and at the right dose. As the United States enters the era of technology-enabled precision medicine, regulators and payers have a clear interest in and responsibility for ensuring that we have the evidence needed to support such approaches.
Despite having different missions, both the FDA and CMS require adequate information to make sound, evidence-based decisions. Virtually everyone will at some point face difficult medical decisions, either for themselves or for a family member, about the best course of action in a unique situation. For too long, physicians, patients, regulators, policy makers, and payers have depended on suboptimal data to guide decision making because of constraints, perceived or real, about what was affordable, practicable, or desirable. But now, thanks to rapidly accelerating scientific knowledge and new perspectives, the FDA and CMS are poised to change this approach for the better and improve outcomes for patients.
Corresponding Author: Robert M. Califf, MD, US Food and Drug Administration, 10903 New Hampshire Ave, Silver Spring, MD 20993 (email@example.com).
Published Online: November 7, 2016. doi:10.1001/jama.2016.16734
Conflict of Interest Disclosures: All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Dr Califf is the Commissioner of Food and Drugs of the FDA. Prior to his appointment to the FDA, Dr Califf received research grant funding from the Patient-Centered Outcomes Research Institute, the National Institutes of Health, the FDA, Amylin, and Lilly; research grants and consulting payments from Bristol-Myers Squibb, Janssen, Merck, and Novartis; consulting payments from Amgen, Bayer Healthcare, BMEB Services, Genentech, GlaxoSmithKline, Heart.org–Daiichi Sankyo, Kowa, Servier, Medscape/Heart.org, Regado, and Roche; he also held equity in N30 Pharma and Portola. No other disclosures were reported.
Additional Contributions: We thank Patrick Conway, MD, MSc, CMS, for providing valuable commentary on an earlier draft of the manuscript. We also thank Jonathan McCall, MS, Duke Clinical Research Institute, Durham, North Carolina, for editorial assistance. Neither contributor received any compensation for their contribution other than usual salary.
Califf RM, Sherman RE, Slavitt A. Knowing When and How to Use Medical ProductsA Shared Responsibility for the FDA and CMS. JAMA. Published online November 07, 2016. doi:10.1001/jama.2016.16734